The Effects of Household Electricity Access on Educational Attainment In Rural Ghana

The Effects of Household Electricity Access on Educational Attainment In Rural Ghana | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article The Effects of Household Electricity Access on Educational Attainment In Rural Ghana Edward Frimpong Baah, Dadzie Takyi Samuel This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9393811/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 18 You are reading this latest preprint version Abstract This study examines the causal relationship between houshold electricity access and educational achievement in rural Ghana. Utilising data from the seventh iteration of the Ghana Living Standards Survey (GLSS7), we assess the impact of home electricity access on the years of completed education for individuals aged 6 to 30 in rural regions. Ordinary Least Squares (OLS) regressions provide a baseline, demonstrating a positive and statistically significant effect of an additional 0.24 years of education. To mitigate endogeneity caused by omitted variable bias and reverse causality, we utilise a Two-Stage Least Squares (2SLS) instrumental variable approach. The IV estimates indicate that access to electricity enhances educational attainment by 0.52 years. Heterogeneity analysis employing interacted instrumental variable specifications indicates that the poverty gradient is the most reliable source of differential effects: the overall impact of electricity access on poor households is 1.17 additional years of education statistically significant at the 1 percent level in contrast to a minimal and insignificant effect of 0.13 years for non-poor households, highlighting electricity’s function as an equalising force for the most resource-deprived families. The findings highlight the significance of rural electrification as a policy tool for bridging poverty-related educational attainment disparities in Sub-Saharan Africa, particularly benefiting the most impoverished households. Figures Figure 1 Figure 2 Introduction The United Nations Sustainable Development Goal 4 strives to ensure equitable and inclusive education that promotes lifelong learning opportunities for everyone (UN). It can be argued that the premise of this goal does not only rest on ensuring that people attend school, but also on their progress through the various stages of education. Globally, education, especially higher education, is recognised as a hub of human capital development (Mamuli, 2020 ). This implies that educational attainment, measured by years of schooling, is crucial for human development and lifelong learning. However, educational attainment in many countries across the world has witnessed many gaps as many people struggle to attain tertiary education (Barro & Lee, 2013 ). Despite the crucial importance of educational attainment on human capital development and national development, educational attainment in many developing countries has been hindered due to factors such as financial constraint, household chores, parents’ education, land sufficiency, and student/teacher ratio (Neupane, 2017 ). While a school attainment gap exists among all genders, the problem is more severe among women and young girls due to socio-cultural barriers such as early marriages, patriarchal norms, and insensitivity of educational systems to religious misinterpretation (Kumar Sarkar et al., 2014 ). The theoretical basis for a positive relationship between household electricity access and educational attainment rests on three interrelated mechanisms. Drawing on Becker’s ( 1964 ) human capital framework and Gronau’s ( 1977 ) household production model, electricity is expected to extend productive study hours by replacing inadequate and costly kerosene lighting, improve access to educational media and digital learning resources, and reduce the domestic labour burden on school-age children, particularly girls, thereby freeing time for schooling (Greenstone and Jack, 2015 ; Kanagawa and Nakata, 2008 ). A further indirect pathway operates through household income: rural electrification raises agricultural productivity and enables non-farm enterprises, alleviating the financial constraints that prevent poor families from investing in education (Khandker et al., 2014). Despite a growing body of quasi-experimental evidence documenting these effects across Latin America, South Asia, and East Africa (Dinkelman, 2011 ; Lipscomb et al., 2013 ; Rud, 2012 ), credible causal evidence for rural Ghana remains scarce. Crucially, no prior study has systematically examined whether the educational returns to electrification differ by poverty status, gender, and age cohort within the Ghanaian context, nor benchmarked IV estimates against propensity score matching to assess the robustness of identification strategies in this setting. Despite the centrality of electricity to human capital development, access to electricity in rural communities remains critically low. Over 666 million people worldwide lack access to electricity. About 85% of these people live in Sub-Saharan Africa (World Bank, 2025). In Sub-Saharan Africa, the rates of access to electricity exhibit significant variation across countries except for South Africa, which boasts an electrification rate of about 85%. West African countries such as Côte d’Ivoire, Ghana, Cameroon, Gabon, Nigeria, and Senegal have relatively high electrification rates ranging from 45% to 75%. Conversely, Kenya, Tanzania, Uganda, and Zambia exhibit lower rates, approximately 25 percent or less (Blimpo et al., 2020 ; Gamette et al., 2024 ). In Ghana, whilst ninety percent of urban households have access to electricity, only fifty-five percent of rural households have access to electricity (GLSS 7). Data from the Ghana Living Standards Survey 7 indicate that access to electricity remains uneven both across and within poverty groups. Among households in absolute poverty, approximately 35 percent lack access to electricity, while the incidence rises to about 49 percent among moderately poor households. Even among non-poor households, a non-negligible 29 percent remain without electricity. Overall, about 37 percent of poor households (combining absolute and moderate poverty) are deprived of electricity, highlighting persistent disparities in access that are not fully explained by income differences alone. Against this backdrop, this study addresses two research questions: (1) Does household access to the national electricity grid causally improve educational attainment among individuals aged 6 to 30 in rural Ghana? (2) Are the educational returns to electrification heterogeneous across poverty status, gender, and age cohort, and if so, are benefits concentrated among the most resource-deprived households? To answer these questions, we exploit data from the seventh round of the Ghana Living Standards Survey (GLSS7) and employ a Two-Stage Least Squares (2SLS) strategy, instrumenting household grid connection with the leave-one-out community-level electrification rate within each enumeration area. We complement this with interacted IV specifications and subgroup IV regressions to characterise the distribution of treatment effects, and benchmark these estimates against five propensity score matching algorithms to assess the robustness of identification. This paper makes three contributions to the literature. First, we provide credible causal evidence on electricity’s educational returns in rural Ghana a context underrepresented in the quasi-experimental energy-education literature by exploiting community-level grid rollout as an instrument. The instrument satisfies the relevance condition (Kleibergen-Paap F-statistic of 1,651) and the exclusion restriction, since Ghana’s ECG and NEDCo expand grids based on engineering and load-forecasting criteria rather than the educational characteristics of resident households. The IV estimate of 0.52 additional years of schooling is approximately twice the OLS estimate of 0.24 years, confirming that conventional regressions substantially understate electrification’s true educational returns. Second, we demonstrate that poverty status is the dominant source of treatment effect heterogeneity: poor households gain over one additional year of schooling (1.17 years, p < 0.01), while the effect for non-poor households is negligible and statistically insignificant (0.13 years), positioning electricity as an equalising force in education rather than a benefit captured by the already-privileged. Gender disaggregation reveals broadly comparable effects for boys and girls, indicating that deep-rooted structural barriers, early marriage, patriarchal norms, and distance to school persist independently of electricity access and require complementary demand-side interventions. Third, the divergence between our IV estimates and propensity score matching results highlights the constraints of matching estimators when selection into electricity access is driven by unobservable household traits, strengthening the case for instrument-based identification in energy access research. These findings carry direct implications for Ghana’s human capital development agenda and for Sub-Saharan Africa’s pursuit of SDG 4. Literature Review Access to electricity has emerged as a significant determinant of educational outcomes in developing countries. The rapid expansion of household electricity connections in Ghana over the past two decades, driven by demographic growth, urbanisation, and the National Electrification Scheme, has created an important opportunity to examine how grid access shapes educational attainment(Ahali & Yaw, 2016 ), creating an important opportunity to examine how household electricity usage shapes children’s educational attainment. Empirical evidence demonstrates that household access to electricity exerts significant positive effects on educational outcomes. León Esteban et al. (2018) established that household electricity access produces significant and positive effects on educational attainment, with electrified households consistently recording higher school enrollment, longer study hours, and improved academic performance compared to their unelectrified counterparts. Moreover, Ye ( 2017 ) found that in Kenya, household electricity usage improves educational attainment not solely through extended study hours and better lighting, but through additional mechanisms including access to digital learning tools, reduced domestic labour burden, and improved household economic productivity. Further, households with electricity access have been shown to record higher adoption of educational digital tools, which correlates with improved academic performance (Boampong et al., 2026 ). Gender and socioeconomic inequalities are well-documented moderators of educational attainment in developing countries. Kumar Sarkar et al. ( 2014 ) document that girls in rural South Asia face compounding barriers including early marriage, patriarchal norms, and gender-insensitive school systems, which reduce their educational progression relative to boys. In Sub-Saharan Africa specifically, Brew-Hammond ( 2010 ) argues that women and girls bear a disproportionate share of household energy burdens, and that electrification programmes targeting these burdens can generate significant female human capital gains. Vecchione et al. ( 2014 ) further show that motivational orientations differ by gender, with intrinsic motivation being a stronger predictor of educational outcomes for females, implying that interventions which improve the home study environment, such as reliable evening lighting, may yield larger returns for girls than for boys. An expanding corpus of quasi-experimental research substantiates these mechanisms. Dinkelman ( 2011 ) employs land gradient as a device for electrification expenses in South Africa and reports substantial positive effects on female labour force participation, aligning with the domestic labour reallocation mechanism that has downstream implications for children's education. The most frequently referenced study in this field, conducted by Lipscomb, Mobarak, and Barham ( 2013 ), utilises a least-cost grid rollout model as an instrument for municipal electrification in Brazil. They discover that a ten percentage point increase in electrification rates correlates with an approximate increase of 0.35 years in average years of schooling, thereby establishing a direct benchmark for the estimates derived in the current study. Rud ( 2012 ) employs programme eligibility standards to ascertain the beneficial impacts of electrification on school enrolment in rural India, demonstrating more pronounced effects for girls than for boys. Bayer, Dolan, and Urpelainen ( 2020 ) synthesise research from many contexts and affirm that the beneficial educational outcomes of electrification are consistent, though contextually dependent. Data from Sub-Saharan Africa, characterised by some of the lowest electricity rates worldwide, is pertinent to the current study. Bensch, Kluve, and Peters ( 2011 ) present experimental results from Rwanda demonstrating that enhanced lighting substantially extends homework duration for primary school students. Van de Walle, Dust, Minh, and Sharma ( 2017 ) utilise panel data from five African nations, including Ghana, to illustrate the beneficial impacts of electrification on educational spending and attendance, particularly among economically disadvantaged households. Blimpo, Postepska, and Xu ( 2020 ) demonstrate that electrification rates for impoverished households in West Africa consistently trail those of affluent households, resulting in enduring energy-related educational disparities that are not entirely attributable to income, a trend clearly evident in the GLSS7 data utilised in this study. The variability of electricity's educational impacts across gender, socioeconomic status, and age has garnered heightened scrutiny. Brew-Hammond ( 2010 ) indicates that women and girls in Sub-Saharan Africa disproportionately endure home energy burdens, and that electrification programs aimed at alleviating these burdens offer significant advantages for the building of female human capital. Njiru and Letema ( 2018 ) present empirical evidence from East Africa indicating that households with electricity access exhibit elevated female school completion rates, which they attribute to the liberation of home time and enhanced study environments. Puzzolo et al. ( 2016 ) affirm via systematic study that access to sustainable energy yields substantial gender co-benefits for education in low-income contexts. Khandker et al. (2014) and Lipscomb et al. ( 2013 ) both conclude that the educational benefits of electrification are most significant for households in the lower income bracket, aligning with the expectation that the marginal utility of electricity is greatest for resource-limited families unable to afford alternative lighting or educational resources. Taken together, the literature establishes a robust theoretical and empirical case for electricity’s positive educational effects, yet evidence specific to rural Ghana remains limited, heterogeneous effects by poverty status are rarely the primary focus of analysis, and no study has directly compared IV and propensity score matching estimates within a single Ghanaian dataset. This study addresses all three gaps. Methodology 3.1. Data and variable definitions This study draws on data from the seventh round of the Ghana Living Standards Survey (GLSS7), the most comprehensive nationally representative household survey on living conditions in Ghana, covering education, health, employment, household expenditure, and housing. The GLSS7 comprises 1,000 primary sampling units, with 43.8% in rural and 56.2% in urban areas, spanning 15,000 households (8,430 rural and 6,570 urban). The survey achieved a response rate of 93.4%, yielding usable data on 14,009 households and 59,864 individuals. The analytical sample is restricted to individuals aged 6 to 30. Six years is the statutory age at which primary school education begins in Ghana, and 30 years is chosen as the upper bound to focus on the cohort most likely to be in or recently completing formal education, while excluding older individuals for whom electricity access during childhood cannot be meaningfully recovered from cross-sectional data. In this study, educational attainment was measured by the number of years of completed schooling, in accordance with the classification used in GLSS7. In accordance with established conventions in the literature, only the highest attained level of education was recorded for each respondent; incomplete levels were not acknowledged. Years of education were designated as follows: no formal education (0 years), Primary education (6 years), Junior High School/Junior Secondary School (9 years), Polytechnic, Teacher and Nursing Training colleges, or equivalent tertiary institution (15 years), Bachelor’s degree (16 years), professional qualification (16 years), and postgraduate degree (17 years). This coding system transforms an ordinal categorical variable into a continuous numeric scale that quantifies the extent of educational disparities among respondents. The independent variable for the study was household access to electricity. In this study, we measured household access to electricity as a binary variable with one (1) indicating that the household has access to the national grid and zero (0) indicating that the household does not use the national grid. The remaining control variables were chosen based on their established predictive power for educational attainment in the Sub-Saharan African literature (Akabayashi and Psacharopoulos, 1999 ; Khandker et al., 2014; Neupane, 2017 ). Table 1 delineates the dependent and independent variables along with their respective measurements. Table 1 Measurement of Variables Variable Measurement Dependent Variables Educational attainment Highest level of education completed measured in years of schooling. Household control Variables Access to Electricity Binary variable equals 1 if household use national grid Log Household per capita income Continuous Household size continuous Educational level of mother Continuous Educational level of father Continuous Log of household education expenditure Continuous Individual Characteristics Gender Binary variable equals 1 if male Age Continuous Authors construct Empirical model We begin with an Ordinary Least Squares (OLS) regression as a baseline specification $$\:edu(Household\:electricity,\:individual\:characteristics,\:Income,\:parental\:education,\:$$ Edu i = α + βElectricity i +X i ′​γ + HH i ′​δ + µ​+ϵ i​ (1) Where; Edu i is the highest level of education an individual has completed, measured in years. X captures individual characteristics such as age, sex, relationship to the household head The vector, HH, contains household-level control variables: age and sex of the household head, household size, household income, education status of the household head, and household expenditures on education and non-education items Identification The OLS estimate of electricity access on educational attainment is subject to endogeneity bias from two primary causes. Initially, unobserved heterogeneity: cognitively proficient parents are concurrently more capable of obtaining electrical connections and imparting academic competence to their offspring, resulting in omitted variable bias. Secondly, reverse causality: households possessing greater human capital may be more adept at advocating for grid expansion or securing private connections, so creating a feedback loop from attainment to access. The cumulative violations of the OLS orthogonality constraint result in a biased and inconsistent β^​. We tackle this using Two-Stage Least Squares (2SLS) estimation, employing the community-level electrification rate as an instrument for household electricity access, the leave-one-out mean proportion of other households within the same enumeration area (EA) that possess electricity access, excluding the household under consideration. Electricity i = α + ΩCluster i +X i ′​γ + HH i ′​δ + µ​+ϵ i​ (2) We estimate Eq. (2) with the ordinary least square model and predict Electricity ̂. In the second stage, we replace Electricity in Eq. (1) with its prediction from Eq. (2) and estimate of educational attainment in Eq. (3) Edu i = α + θElectricity i ̂ +X i ′​γ + HH i ′​δ + µ​+ϵ i​​ (3) A causal interpretation of θ requires the instrument to satisfy both the relevance condition and the exclusion restriction simultaneously. The instrument meets the relevance criterion as grid infrastructure in Ghana is implemented at the community level by ECG and NEDCo, resulting in a robust within-EA correlation in electricity access, which we confirm using the Kleibergen-Paap F-statistic. The exclusion restriction is maintained as the decisions about ECG and NEDCo grid extensions are influenced by engineering and load-forecasting factors specifically, proximity to transmission lines and community size, rather than the educational attributes of the resident families. Consequently, there exists no reliable independent correlation between neighbors’ power availability and a child’s completed years of education. Secondly, we do a falsification test utilising the age of the household head as a placebo outcome, which should remain unaffected by an individuals access to electricity. The IV second stage yields a negligible coefficient on this placebo result (coefficient: -0.783, standard error 1.064, p = 0.462). This null result corroborates the exogeneity of the instrument, indicating that the variance in electricity availability is unlikely to be influenced by unobserved household factors associated with age. The instrument therefore does not predict pre-determined demographic outcomes, lending further credibility to the identifying assumptions. The study used propensity score matching (PSM) alongside instrumental variable estimations to mitigate potential endogeneity, as noted in previous research (Kofinti et al, 2022 ). The treatment variable in this study is access to electricity, utilised to evaluate the average treatment effect on educational attainment. The technique generates an estimate to assess the counterfactual effect of access to electricity on educational attainment. We employ five matching strategies to conduct sensitivity tests on our findings: Kernel Matching, Logit + Kernel Matching, Nearest Neighbour (one-to-one)and Radius matching methods. Descriptive Statistics Table 2 presents the descriptive statistics of the primary variables of the study and examines the disparities between homes with access to the national grid and those without. The estimation sample consists of 3,219 individuals, including 1,157 (36%) from houses without access to electricity and 2,062 (64%) from households having electricity. Households with electricity have an average educational attainment of 8 years, but those without electricity average 7 years, and this disparity is statistically significant. Concerning gender, there is essentially no distinction between houses with electricity and those without. Individuals in houses with electricity tend to be marginally older than those in households without power. Households with electricity exhibit higher per capita income and allocate more funds to education compared to those lacking power access. Households lacking electricity exhibit bigger sizes than those with electricity. We additionally observe that both fathers and mothers of individuals from households with electricity possess greater levels of education than those from households without electricity. Ultimately, heads of families lacking electricity are, on average, older than those with electricity access, and all these disparities are statistically significant at the 1% level, with the exception of gender. Table 2 Descriptive statistics Educational attainment No Electricity (N = 1,157) Electricity (N = 2,062) Difference p-value 6.93 8.03 -1.10 0.000 Access to electricity 0.00 1.00 -1.00 . Gender 0.42 0.43 -0.01 0.634 Age in years 16.99 18.17 -1.18 0.000 Log of household per capita income 6.44 7.19 -0.75 0.000 Log of educational expenditure 5.60 6.54 -0.94 0.000 Household size 6.54 6.09 0.44 0.000 Years of Father’s education 2.37 4.68 -2.31 0.000 Years of mothers education 1.57 3.26 -1.69 0.000 Age of household head 50.12 47.31 2.81 0.000 Source: Computed from GLSS 7 Results Prior to the regression analysis, we conducted a brief descriptive analysis of the distribution of electricity across Ghana. In 2017, electricity remained unequally distributed across Ghana, with only 32 per cent of rural households connected to the national grid, while 68 per cent of urban households were connected to the national grid. Model Results Table 3 presents the OLS results of the effects of electricity on educational attainment. The results reveal that access to electricity (national grid) in rural Ghana has a statistically significant positive effect on educational attainment (0.239, p < 0.05). This implies that, holding all other factors constant, individuals aged 6 to 30 in rural households with electricity attain 0.24 years of education more than individuals of the same age group in rural households without access to electricity. This confirms our a priori hypothesis that electricity access enhances educational attainment, possibly through extended hours of study, access to information through television, radio, and digital learning platforms. The results further revealed that being male is associated with 0.63 additional years of schooling compared to females (p < 0.01). While statistically significant, this modest gap suggests that gender barriers to educational progression in rural Ghana are driven by structural factors that household-level controls only partially capture. Each additional year of age is associated with 0.2 more years of completed schooling, consistent with the cumulative nature of educational attainment. Moreover, the results revealed that a one percent increase in household per capita income is associated with 0.14 additional years of schooling. Wealthier households can afford learning materials such as textbooks, computers, furniture, school uniforms, school fees, and home tuition, which translates into higher educational attainment. We found that educational expenditure was the biggest predictor of educational attainment. A one percent increase in household education expenditure is associated with 0.42 additional years of schooling, making it the strongest predictor among the controls. Household size reduces educational attainment by 0.11 years. This affirms that larger households have lower per capita resources needed to spend on education and electricity. Effects of access to electricity on educational attainment in Ghana (Baseline-OLS) Table 3 Effect of electricity on educational attainment (1) OLS National electricity 0.239 ** (0.110) Male 0.626 *** (0.110) Age in years 0.200 *** (0.008) Log Household Income Per Capita 0.136 *** (0.033) Log of educational expenditure 0.423 *** (0.041) Household size -0.110 *** (0.018) Educational level of Father 0.016 (0.013) Educational level of Mother) 0.029 * (0.015) Age of household head 0.012 *** (0.003) Constant -0.199 (0.422) Observation 3219.000 R squared 0.340 Standard errors in parentheses * p < .1, ** p < 0.05, *** p < .01 Source: Computed from GLSS 7 Effects of access to electricity on educational attainment in Ghana (IV) As discussed in Section 3, the OLS estimate of electricity’s effect on educational attainment is subject to endogeneity bias from omitted variable bias and reverse causality. To address this, we instrument household electricity access with the leave-one-out community-level electrification rate within each enumeration area. The 2SLS results indicate that electricity access increases completed years of schooling by 0.52 years (p < 0.01), approximately double the OLS estimate of 0.24 years. This upward revision relative to OLS is consistent with two mechanisms discussed in Section 3: classical attenuation bias from measurement error in the binary electricity indicator, and a Local Average Treatment Effect (LATE) that captures households with electricity at the margin of grid access, who may derive particularly large educational returns from connection. The OLS estimate of electricity access on educational attainment is influenced by endogeneity bias from two main sources: omitted variable bias due to the positive correlation between household socioeconomic status and both electricity access and educational investment, and reverse causality, where households with greater human capital may be more capable of advocating for grid connection. Both sources generally induce an upward bias in Ordinary Least Squares (OLS). Nonetheless, our IV estimates surpass the OLS estimates, a trend we view as aligned with two supplementary causes. The measurement error in the binary electrical access variable, which just reflects grid connection status instead of duration or dependability of service, may diminish OLS coefficients towards zero. Secondly, the instrument determines a Local Average Treatment Effect (LATE) for compliers, specifically homes whose connection status is influenced by community-level electrification. These homes are on the periphery of grid access and may experience significant educational advantages from connection, resulting in a Local Average Treatment Effect (LATE) that surpasses the overall population average treatment effect. Disentangling these two mechanisms is beyond the scope of the present study and remains a direction for future research. Table 4 Effects of access to electricity on educational attainment in Ghana (IV) Variables (2) IV National electricity 0.521 *** (0.172) Male 0.623 *** (0.110) Age in years 0.199 *** (0.008) Log of Household Income Per Capita 0.125 *** (0.034) Log of expenditure on education 0.403 *** (0.042) Household size -0.108 *** (0.018) Educational level of Father 0.014 (0.013) Educational level of Mother 0.027 * (0.015) Age of household head 0.012 *** (0.003) Constant -0.189 (0.422) Observation 3219.000 R squared 0.338 Idstat 651.259 Idp 0.000 Widstat 1651.270 Source: Computed from GLSS 7 Heterogeneity analysis We evaluate diverse effects across two dimensions for which interacting instrumental variable estimates are accessible: gender and socioeconomic status (poverty classification). The heterogeneity is analysed by interacting the instrumented electricity access variable with each moderating characteristic in the second stage of the 2SLS, with the aggregated subgroup effect calculated through a linear combination (lincom) of the primary electricity coefficient and the pertinent interaction term. Gender Heterogeneity Column (1) of Table 5 indicates that the baseline effect of electricity access on male years of schooling is 0.585 years (SE: 0.267, p < 0.05). The interaction term between electricity and female status is negative yet statistically insignificant (coefficient: -0.117, SE: 0.317, p = 0.711), suggesting that females do not experience a statistically distinct effect compared to boys. The cumulative electricity effect for females, derived from the linear combination of the primary electricity coefficient and the interaction term, is 0.468 years (SE: 0.285, p = 0.101), which is just below conventional significance levels. The data indicate that although access to electricity enhances educational attainment for both boys and girls, the extent of this effect does not significantly vary by gender in this sample. This outcome diverges from earlier anticipations that girls would gain disproportionately from electricity via increased study hours and less household labour responsibilities. An acceptable view is that the structural and social impediments limiting females’ education in rural Ghana, such as early marriage, cultural norms, and distance to educational institutions, are so deeply rooted that mere access to electricity cannot produce a significant advantage. Consequently, complementary gender-specific interventions may be necessary in conjunction with electrification to completely bridge the gender gap in educational achievement. Heterogeneity of poverty Column (2) of Table 5 delineates the poverty heterogeneity specification, categorising poor families as those situated inside the lowest two income quintiles. The baseline electricity effect for non-impoverished households is minimal and statistically negligible (coefficient: 0.134, SE: 0.231, p = 0.562). The relationship between electricity access and impoverished household status is positive and statistically significant (coefficient: 1.038, SE: 0.486, p < 0.05), suggesting that impoverished households get considerably greater educational benefits from power access compared to their non-impoverished counterparts. The cumulative impact of electricity on impoverished households, calculated as the linear amalgamation of primary electricity and interaction coefficients, amounts to 1.172 years of supplementary education (SE: 0.429, p < 0.01). This is a substantial and accurately quantified effect, approximately twice the baseline estimate, and constitutes the most significant discovery in the heterogeneity study. This outcome aligns with the view that electricity serves as a partial equaliser: households with less resources have the most significant obstacles to studying and being productive after dark, hence electrification yields the greatest marginal benefits for these households. The discovery that the non-poor baseline effect is negligible further substantiates this interpretation. Households with more resources can more effectively compensate for the lack of electricity through alternatives like candles, generators, or private tutoring, thereby mitigating the marginal effect of grid access. The substantial negative coefficient for the poor indicator (-0.688, SE: 0.334, p < 0.05) independently corroborates the current educational disadvantage experienced by impoverished households, highlighting the critical role of electrification as a focused development tool. Table 5 Heterogeneous Effects of Electricity on Years of Education: Interacted IV (1) (2) (3) Gender Poverty Age National electricity 0.585 ** 0.134 0.597 ** (0.267) (0.231) (0.300) Electr*female -0.117 (0.317) Electr*poor 1.038 ** (0.486) Electr*young -0.302 (0.309) female -0.537 ** (0.254) household in bottom 2 income quintiles -0.688 ** (0.334) aged 6–15 -1.884 *** (0.260) Age in years 0.199 *** 0.195 *** 0.075 *** (0.009) (0.009) (0.014) Log household Income Per Capita 0.126 *** 0.132 ** 0.112 *** (0.039) (0.060) (0.038) Log total Education Expenditure 0.404 *** 0.403 *** 0.388 *** (0.050) (0.049) (0.050) Household size -0.108 *** -0.108 *** -0.114 *** (0.022) (0.022) (0.022) Educational level of Father 0.014 0.016 0.021 (0.014) (0.014) (0.014) Educational level of Mother 0.027 0.026 0.034 * (0.017) (0.017) (0.017) Age of household head 0.012 *** 0.013 *** 0.013 *** (0.003) (0.003) (0.003) Constant 0.373 0.364 3.343 *** (0.489) (0.603) (0.516) Subgroup effect 0.468 1.172 0.295 (SE) 0.285 0.429 0.229 [p-value] 0.101 0.006 0.198 Observations 3219 3219 3219 KP F-stat (1st stage) 107.400 60.149 92.590 AR stat 123.217 85.186 87.346 AR p-value 0.000 0.000 0.000 Clustered standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01. Subgroup effect = lincom of nat_electr + interaction term. Subgroup Heterogeneity Analysis To examine the distributional nature of the baseline estimate, we conduct distinct subgroup IV regressions across six mutually exclusive or complimentary categories: females, males, impoverished families, non-impoverished households, children aged 6–15, and young people aged 16–30. This method enhances the interacting IV specifications by permitting all slope coefficients, not solely the electricity effect, to fluctuate between subgroups, thus circumventing the limiting homogeneity assumptions associated with a singular pooled interaction factor. The results demonstrate a markedly diverse pattern that is both economically significant and statistically rigorous. The most accurately calculated and significantly large effect is noted among impoverished households (column 3), where access to electricity increases years of schooling by 1.206 years (SE: 0.397, p < 0.01). This figure exceeds double the equivalent estimate for non-poor families (0.176, SE: 0.244, p = 0.470), which is economically insignificant and statistically indistinguishable from zero. The distinction is clear and logically sound: non-poor households can compensate for the lack of grid electricity with generators, battery lighting, and private tuition, thus diminishing the marginal educational benefits of electrification. In contrast, impoverished households encounter stringent limitations on study time during the evening and are disproportionately dependent on domestic child labour, both of which electricity alleviates directly. This discovery establishes electricity access as a true equalising factor inside the Ghanaian education system, yielding the greatest benefits precisely in areas of severe educational deprivation. The gender disaggregation is also instructive. The electrical impact for girls (0.540, SE: 0.279, p < 0.10) and males (0.583, SE: 0.272, p < 0.05) is statistically comparable, as seen by overlapping confidence intervals and a coefficient difference of less than 0.05 years. The pronounced negative baseline disparity in girls’ education identified in the interacted specification due to deeply rooted social norms, early marriage, and cultural role expectations, persist regardless of electricity access, indicating that electrification alone cannot eradicate the structural factors contributing to the gender gap. This null result is a significant finding: it challenges the optimistic assumption that electricity would disproportionately benefit girls by freeing them from nighttime domestic restrictions, and underscores the need for complementary demand-side interventions targeting the structural roots of the gender gap. The results of the age subgroup results are also instructive. Children aged 6 to 15 experience a substantial and precisely estimated effect of 0.543 years (SE: 0.184, p < 0.01); the Kleibergen-Paap Wald F-statistic of 297 for this subsample is well above standard weak-instrument thresholds. The older group (ages 16–30) produces a similar point estimate of 0.528 years (SE: 0.305, p < 0.10), albeit with significantly broader confidence intervals, indicating increased variability in educational pathways among young individuals. The sustained economic relevance of the effect for the older generation aligns with the ongoing prevalence of extended education, adult literacy initiatives, and vocational training in rural Ghana, where delayed educational advancement is prevalent. The comparable magnitudes across age groups refute the notion of a singular study-lighting channel as the exclusive mechanism and endorse a more comprehensive interpretation that includes reduced child labour, increased household productivity, and improved access to educational media. The subgroup analysis indicates that poverty is the principal factor of heterogeneity in the educational returns of electricity. The targeted electrification of impoverished rural areas in Ghana is expected to yield significant educational benefits, positively impacting various genders and age groups within such communities. We present a graph of the coefficients of heterogeneity in Fig. 2 . Table 6 Heterogeneous Effects of Electricity: Subgroup IV Regressions (1) (2) (3) (4) (5) (6) Females Males Poor Household Non-Poor household Age 6–15 Age 16–30 National electricity 0.540 * 0.583 ** 1.206 *** 0.176 0.543 *** 0.528 * (0.279) (0.272) (0.397) (0.244) (0.184) (0.305) Age in years 0.170 *** 0.242 *** 0.176 *** 0.204 *** 0.422 *** -0.011 (0.012) (0.013) (0.015) (0.011) (0.025) (0.019) Log Househol Income Per Capita 0.109 ** 0.136 *** 0.042 0.106 0.023 0.162 *** (0.050) (0.051) (0.111) (0.069) (0.046) (0.053) Log Total Expenditure on education 0.347 *** 0.471 *** 0.227 *** 0.493 *** 0.106 ** 0.522 *** (0.061) (0.066) (0.067) (0.065) (0.042) (0.070) Househould size -0.104 *** -0.106 *** -0.026 -0.148 *** -0.034 * -0.181 *** (0.029) (0.025) (0.032) (0.027) (0.019) (0.036) Educational level of Father 0.016 0.013 0.040 ** 0.008 0.022 0.029 (0.018) (0.021) (0.019) (0.018) (0.013) (0.020) Educational level of Mother 0.025 0.031 0.016 0.032 0.001 0.052 ** (0.024) (0.022) (0.025) (0.021) (0.018) (0.026) Age of household head 0.012 *** 0.012 *** 0.013 ** 0.013 *** 0.006 0.007 (0.005) (0.004) (0.005) (0.004) (0.004) (0.005) Constant 0.827 -0.891 0.985 0.018 -0.107 4.726 *** (0.632) (0.580) (0.767) (0.715) (0.556) (0.713) Observation 1849.000 1370.000 1211.000 2008.000 1463.000 1756.000 widstat 274.948 302.123 117.478 285.694 297.010 239.566 Standard errors in parentheses * p < .1, ** p < 0.05, *** p < .01 Source: Computed from GLSS 7 Propensity Score Matching Analysis Table 7 presents PSM estimates as a robustness check against the IV strategy. We employ five matching algorithms, kernel, logit-kernel, nearest neighbour, radius, and bootstrapped kernel, to assess convergence across identification approaches. Across all five specifications, the ATT estimates range from 0.038 to 0.181 years and are statistically indistinguishable from zero, falling far short of the IV estimate of 0.52 years. This divergence is theoretically expected: PSM eliminates bias only from observable confounders, whereas the IV approach additionally removes bias from unobservable household traits, such as parental motivation and social networks, that simultaneously drive electricity access and educational investment. The gap between the two estimators therefore reinforces our preferred IV approach and confirms that instrument-based identification is essential in this setting. Table 7 Propensity Score Matching using different matching algorithms Matching Algorithm ATT Observations Kernel Matching (Baseline) 0.038 (0.170) 3,219 Logit + Kernel Matching 0.046 (0.177) 3,219 Nearest Neighbour (one-to-one) 0.038 (0.170) 3,218 Radius 0.1807 (0.135) 3,218 Bootstrapped Kernel PSM (50 replications) 0.038 (0.200) 3,219 Notes: Standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.10. ATT = Average Treatment Effect on the Treated. All specifications use national electricity grid access as the treatment variable and years of education as the outcome variable. Propensity scores estimated using probit unless otherwise stated. Table 8 Placebo Test:Effect of Electricity on Household Head Age Variables (1) Outcome: Household Head Age Electricity (nat_electr) -0.783 (1.064) Male 0.407 (0.748) Age (individual) -1.163*** (0.053) Log HH Income per capita -1.139*** (0.366) Log Education Expenditure 0.592* (0.311) Household Size 0.742*** (0.146) Father’s Education 0.093 (0.088) Mother’s Education 0.136 (0.102) Constant 67.707*** (3.353) Observations: 3,930 Clusters: 545 Controls: Yes Clustered standard errors in parentheses. * p < 0.10, ** p < 0.05, *** p < 0.01 Conclusion and Policy Recommendations This study provides credible causal evidence that household access to grid electricity significantly enhances educational attainment in rural Ghana. Using a Two-Stage Least Squares strategy instrumented by the leave-one-out community-level electrification rate, we find that electricity access raises completed years of schooling by 0.52 years, approximately twice the naïve OLS estimate of 0.24 years, confirming that conventional regression approaches substantially understate the true educational returns to electrification. Heterogeneity analysis reveals that poverty status is the principal source of differential effects: poor households gain over one additional year of schooling (1.17 years, p < 0.01), while the effect for non-poor households is negligible and statistically insignificant (0.13 years). Gender disaggregation yields broadly comparable effects for boys and girls, indicating that deep-rooted structural barriers. early marriage, patriarchal norms, and distance to school, persist independently of electricity access and require complementary demand-side interventions. Effects are consistently significant for children aged 6–15 and older youth (16–30), suggesting that mechanisms beyond study-lighting, including domestic labour reallocation and access to educational media, are operative across the lifecycle. These findings carry direct policy implications. First, rural electrification should be explicitly integrated into Ghana’s human capital development agenda, with targeted grid expansion prioritising communities in the lowest income quintiles where educational returns are largest. Second, since electrification alone does not close the gender gap, complementary interventions, conditional cash transfers, school proximity programmes, and community-based gender sensitisation, must accompany electrification rollouts to translate energy access into equitable educational outcomes. These results position targeted rural electrification as a high-return, inequality-reducing policy instrument with broad implications for Sub-Saharan Africa’s pursuit of SDG 4. Several limitations of the present study merit acknowledgement. First, the GLSS7 is cross-sectional, precluding the estimation of within-individual or within-household fixed effects; the IV strategy mitigates but does not fully eliminate concerns about time-invariant unobservables. Second, the binary measure of grid connection does not capture the reliability, duration, or quality of electricity supply, which may attenuate estimated effects. Third, the PSM results are consistently small and statistically insignificant, suggesting residual confounding from unobservable household traits that matching cannot address — a finding that reinforces the superiority of the IV approach in this context. Future research employing panel data and more granular electricity quality measures would further sharpen our understanding of electrification’s educational returns across Sub-Saharan Africa. Declarations Consent to Participate: Not applicable. Consent to Publish: Not applicable. Ethics approval: This study used publicly available secondary data from the Ghana Living Standards Survey Round 7 (GLSS7). No ethics approval was required as no primary data collection involving human subjects was conducted. Competing interests: The authors declare no competing interests. Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Author Contribution BFE: Conceptualisation, methodology, formal analysis, writing of original draft. DTS: Conceptualisation, data curation, writing of review literature and editing. all aruhors reviews the manuscript Data Availability The data used in this study are from the Ghana Living Standards Survey Round 7 (GLSS7), publicly available from the Ghana Statistical Service at www.statsghana.gov.gh References Ahali AY, Yaw F. Household energy demand in Ghana: The role of household characteristics. Energy Policy. 2016;94:235–44. Akabayashi H, Psacharopoulos G. The trade-off between child labour and human capital formation: A Tanzanian case study. J Dev Stud. 1999;35(5):120–40. Barro RJ, Lee JW. A new data set of educational attainment in the world, 1950–2010. J Dev Econ. 2013;104:184–98. https://doi.org/10.1016/j.jdeveco.2012.10.001 . Bayer P, Dolan L, Urpelainen J. The effect of household electricity access on women’s empowerment: Evidence from rural India. World Dev. 2020;130:104895. https://doi.org/10.1016/j.worlddev.2020.104895 . Becker GS. Human capital: A theoretical and empirical analysis, with special reference to education. University of Chicago Press; 1964. Bensch G, Kluve J, Peters J. Impacts of rural electrification in Rwanda. J Dev Eff. 2011;3(4):567–88. https://doi.org/10.1080/19439342.2011.629521 . Blimpo MP, Postepska A, Xu Y. Why is household electricity uptake low in Sub-Saharan Africa? World Dev. 2020;133:105002. https://doi.org/10.1016/j.worlddev.2020.105002 . Boampong O, Frimpong EB, Dadzie TS. (2026). Digital learning tools, electricity access and academic performance in rural Ghana. J Afr Educ. (Forthcoming). Brew-Hammond A. Energy access in Africa: Challenges ahead. Energy Policy. 2010;38(5):2291–301. https://doi.org/10.1016/j.enpol.2009.12.016 . Dinkelman T. The effects of rural electrification on employment: New evidence from South Africa. Am Econ Rev. 2011;101(7):3078–108. https://doi.org/10.1257/aer.101.7.3078 . Fessler P, Schneebaum A. Gender and educational attainment across generations in Austria. Fem Econ. 2012;18(1):161–88. https://doi.org/10.1080/13545701.2011.637759 . Gamette P, Asante FA, Amoako-Tuffour J. Electrification and welfare outcomes in West Africa: A cross-country assessment. Energy Sustain Dev. 2024;78:101370. Ghana Statistical Service. Ghana Living Standards Survey Round 7 (GLSS 7): Main report. Ghana Statistical Service; 2019. Greenstone M, Jack BK. Envirodevonomics: A research agenda for an emerging field. J Econ Lit. 2015;53(1):5–42. https://doi.org/10.1257/jel.53.1.5 . Gronau R. Leisure, home production, and work: The theory of the allocation of time revisited. J Polit Econ. 1977;85(6):1099–123. https://doi.org/10.1086/260629 . Kanagawa M, Nakata T. Assessment of access to electricity and the socioeconomic impacts in rural areas of developing countries. Energy Policy. 2008;36(6):2016–29. https://doi.org/10.1016/j.enpol.2008.01.041 . Khandker SR, Barnes DF, Samad HA. Welfare impacts of rural electrification: A panel data analysis from Vietnam. Econ Dev Cult Change. 2013;62(4):659–92. https://doi.org/10.1086/676930 . Kofinti RE, Baako-Amponsah J, Danso P. Household National Health Insurance Subscription and Learning Outcomes of Poor Children in Ghana. Child Indic Res. 2022;1–38. https://doi.org/10.1007/s12187-022-09980-y . Kumar Sarkar S, Islam Khatun M, Ray S. Girls’ education in developing countries: Issues and challenges. Int J Educ Res. 2014;2(7):343–60. Lipscomb M, Mobarak AM, Barham T. Development effects of electrification: Evidence from the geologic placement of hydropower plants in Brazil. Am Economic Journal: Appl Econ. 2013;5(2):200–31. https://doi.org/10.1257/app.5.2.200 . Mamuli LC. The impact of higher education on human capital development in developing countries. Int J Educational Sci. 2020;28(1–3):1–8. https://doi.org/10.31901/24566322.2020/28.1-3.1175 . Ministry of Energy. National Electrification Scheme (NES)-Master Plan Review (2011–2020); Ministry of Energy. Accra,Ghana; 2010. Murillo FJ, Román M. Latin America: School bullying and academic achievement. CEPAL Rev. 2011;104:37–54. Neupane S. Factors affecting educational attainment in developing countries. Int J Social Sci Manage. 2017;4(3):189–96. https://doi.org/10.3126/ijssm.v4i3.17423 . Njiru C, Letema SC. (2018). Energy poverty and its implication on standard of living in Kirinyaga, Kenya. Journal of Energy, 2018 , 3196567. https://doi.org/10.1155/2018/3196567 Puzzolo E, Pope D, Stanistreet D, Rehfuess EA, Bruce NG. Clean fuels for resource-poor settings: A systematic review of barriers and enablers to adoption and sustained use. Environ Res. 2016;146:218–34. https://doi.org/10.1016/j.envres.2016.01.015 . Rud JP. Electricity provision and industrial development: Evidence from India. J Dev Econ. 2012;97(2):352–67. https://doi.org/10.1016/j.jdeveco.2011.06.010 . Schultz TW. Investment in human capital. Am Econ Rev. 1961;51(1):1–17. Van de Walle D, Dust M, Minh NT, Sharma M. Poor households’ electricity connections and use in five African countries. World Bank Economic Rev. 2017;31(3):615–41. https://doi.org/10.1093/wber/lhw020 . Vecchione M, Alessandri G, Marsicano G, Caprara GV. Academic motivation predicts educational attainment: Does gender make a difference? Learn Individual Differences. 2014;32:124–31. https://doi.org/10.1016/j.lindif.2014.01.003 . Ye J. Household electricity use and children’s educational outcomes in rural Kenya. Energy Sustain Dev. 2017;37:45–54. https://doi.org/10.1016/j.esd.2017.01.003 . Additional Declarations No competing interests reported. Supplementary Files StataCodes.do Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 13 May, 2026 Reviews received at journal 10 May, 2026 Reviews received at journal 07 May, 2026 Reviews received at journal 06 May, 2026 Reviews received at journal 06 May, 2026 Reviews received at journal 05 May, 2026 Reviews received at journal 02 May, 2026 Reviewers agreed at journal 26 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers invited by journal 24 Apr, 2026 Editor invited by journal 22 Apr, 2026 Editor assigned by journal 15 Apr, 2026 Submission checks completed at journal 15 Apr, 2026 First submitted to journal 12 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9393811","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":633875669,"identity":"15260196-3584-4140-80e6-fcba05b4b6f6","order_by":0,"name":"Edward Frimpong Baah","email":"data:image/png;base64,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","orcid":"","institution":"University of Cape Coast","correspondingAuthor":true,"prefix":"","firstName":"Edward","middleName":"Frimpong","lastName":"Baah","suffix":""},{"id":633875670,"identity":"525bb14c-205e-4b82-bc24-f9bef17eb9bd","order_by":1,"name":"Dadzie Takyi Samuel","email":"","orcid":"","institution":"University of Cape Coast","correspondingAuthor":false,"prefix":"","firstName":"Dadzie","middleName":"Takyi","lastName":"Samuel","suffix":""}],"badges":[],"createdAt":"2026-04-12 11:38:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9393811/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9393811/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108804220,"identity":"4ced6f78-ffaf-45e9-9bf4-72d04c09bcca","added_by":"auto","created_at":"2026-05-08 15:18:08","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":24604,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of national electricity access across Ghana in 2017.\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9393811/v1/33edc5015fb93a2a3ef4089b.jpg"},{"id":108545843,"identity":"b40aab5b-6f43-4236-a673-ec29f506a447","added_by":"auto","created_at":"2026-05-05 20:28:32","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":464118,"visible":true,"origin":"","legend":"\u003cp\u003eCoefficient plot of heterogeneous IV estimates of electricity access on years of education, by gender, poverty status, and age cohort. Point estimates with 95% confidence intervals. Poor household and Age 6–15 subgroups show the largest and most precisely estimated effects.\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9393811/v1/69b5d25168faab75fe8391c7.jpg"},{"id":108809332,"identity":"0f6aaaae-b5ac-4210-a290-baebe164ab2e","added_by":"auto","created_at":"2026-05-08 15:52:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1115116,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9393811/v1/6bbd8f4f-9ddc-432f-80b5-745a7ddd36e8.pdf"},{"id":108545841,"identity":"56b7c25d-8ac7-4e39-ae0e-4d20566bf22b","added_by":"auto","created_at":"2026-05-05 20:28:32","extension":"do","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":27076,"visible":true,"origin":"","legend":"","description":"","filename":"StataCodes.do","url":"https://assets-eu.researchsquare.com/files/rs-9393811/v1/1257176392b203883be9e97d.do"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Effects of Household Electricity Access on Educational Attainment In Rural Ghana","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe United Nations Sustainable Development Goal 4 strives to ensure equitable and inclusive education that promotes lifelong learning opportunities for everyone (UN). It can be argued that the premise of this goal does not only rest on ensuring that people attend school, but also on their progress through the various stages of education. Globally, education, especially higher education, is recognised as a hub of human capital development (Mamuli, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This implies that educational attainment, measured by years of schooling, is crucial for human development and lifelong learning. However, educational attainment in many countries across the world has witnessed many gaps as many people struggle to attain tertiary education (Barro \u0026amp; Lee, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite the crucial importance of educational attainment on human capital development and national development, educational attainment in many developing countries has been hindered due to factors such as financial constraint, household chores, parents\u0026rsquo; education, land sufficiency, and student/teacher ratio (Neupane, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). While a school attainment gap exists among all genders, the problem is more severe among women and young girls due to socio-cultural barriers such as early marriages, patriarchal norms, and insensitivity of educational systems to religious misinterpretation (Kumar Sarkar et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe theoretical basis for a positive relationship between household electricity access and educational attainment rests on three interrelated mechanisms. Drawing on Becker\u0026rsquo;s (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1964\u003c/span\u003e) human capital framework and Gronau\u0026rsquo;s (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1977\u003c/span\u003e) household production model, electricity is expected to extend productive study hours by replacing inadequate and costly kerosene lighting, improve access to educational media and digital learning resources, and reduce the domestic labour burden on school-age children, particularly girls, thereby freeing time for schooling (Greenstone and Jack, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Kanagawa and Nakata, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). A further indirect pathway operates through household income: rural electrification raises agricultural productivity and enables non-farm enterprises, alleviating the financial constraints that prevent poor families from investing in education (Khandker et al., 2014). Despite a growing body of quasi-experimental evidence documenting these effects across Latin America, South Asia, and East Africa (Dinkelman, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Lipscomb et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Rud, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), credible causal evidence for rural Ghana remains scarce. Crucially, no prior study has systematically examined whether the educational returns to electrification differ by poverty status, gender, and age cohort within the Ghanaian context, nor benchmarked IV estimates against propensity score matching to assess the robustness of identification strategies in this setting.\u003c/p\u003e \u003cp\u003eDespite the centrality of electricity to human capital development, access to electricity in rural communities remains critically low. Over 666\u0026nbsp;million people worldwide lack access to electricity. About 85% of these people live in Sub-Saharan Africa (World Bank, 2025). In Sub-Saharan Africa, the rates of access to electricity exhibit significant variation across countries except for South Africa, which boasts an electrification rate of about 85%. West African countries such as C\u0026ocirc;te d\u0026rsquo;Ivoire, Ghana, Cameroon, Gabon, Nigeria, and Senegal have relatively high electrification rates ranging from 45% to 75%. Conversely, Kenya, Tanzania, Uganda, and Zambia exhibit lower rates, approximately 25 percent or less (Blimpo et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Gamette et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn Ghana, whilst ninety percent of urban households have access to electricity, only fifty-five percent of rural households have access to electricity (GLSS 7). Data from the Ghana Living Standards Survey 7 indicate that access to electricity remains uneven both across and within poverty groups. Among households in absolute poverty, approximately 35 percent lack access to electricity, while the incidence rises to about 49 percent among moderately poor households. Even among non-poor households, a non-negligible 29 percent remain without electricity. Overall, about 37 percent of poor households (combining absolute and moderate poverty) are deprived of electricity, highlighting persistent disparities in access that are not fully explained by income differences alone.\u003c/p\u003e \u003cp\u003eAgainst this backdrop, this study addresses two research questions: (1) Does household access to the national electricity grid causally improve educational attainment among individuals aged 6 to 30 in rural Ghana? (2) Are the educational returns to electrification heterogeneous across poverty status, gender, and age cohort, and if so, are benefits concentrated among the most resource-deprived households? To answer these questions, we exploit data from the seventh round of the Ghana Living Standards Survey (GLSS7) and employ a Two-Stage Least Squares (2SLS) strategy, instrumenting household grid connection with the leave-one-out community-level electrification rate within each enumeration area. We complement this with interacted IV specifications and subgroup IV regressions to characterise the distribution of treatment effects, and benchmark these estimates against five propensity score matching algorithms to assess the robustness of identification.\u003c/p\u003e \u003cp\u003eThis paper makes three contributions to the literature. First, we provide credible causal evidence on electricity\u0026rsquo;s educational returns in rural Ghana a context underrepresented in the quasi-experimental energy-education literature by exploiting community-level grid rollout as an instrument. The instrument satisfies the relevance condition (Kleibergen-Paap F-statistic of 1,651) and the exclusion restriction, since Ghana\u0026rsquo;s ECG and NEDCo expand grids based on engineering and load-forecasting criteria rather than the educational characteristics of resident households. The IV estimate of 0.52 additional years of schooling is approximately twice the OLS estimate of 0.24 years, confirming that conventional regressions substantially understate electrification\u0026rsquo;s true educational returns. Second, we demonstrate that poverty status is the dominant source of treatment effect heterogeneity: poor households gain over one additional year of schooling (1.17 years, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), while the effect for non-poor households is negligible and statistically insignificant (0.13 years), positioning electricity as an equalising force in education rather than a benefit captured by the already-privileged. Gender disaggregation reveals broadly comparable effects for boys and girls, indicating that deep-rooted structural barriers, early marriage, patriarchal norms, and distance to school persist independently of electricity access and require complementary demand-side interventions. Third, the divergence between our IV estimates and propensity score matching results highlights the constraints of matching estimators when selection into electricity access is driven by unobservable household traits, strengthening the case for instrument-based identification in energy access research. These findings carry direct implications for Ghana\u0026rsquo;s human capital development agenda and for Sub-Saharan Africa\u0026rsquo;s pursuit of SDG 4.\u003c/p\u003e"},{"header":"Literature Review","content":"\u003cp\u003eAccess to electricity has emerged as a significant determinant of educational outcomes in developing countries. The rapid expansion of household electricity connections in Ghana over the past two decades, driven by demographic growth, urbanisation, and the National Electrification Scheme, has created an important opportunity to examine how grid access shapes educational attainment(Ahali \u0026amp; Yaw, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), creating an important opportunity to examine how household electricity usage shapes children\u0026rsquo;s educational attainment.\u003c/p\u003e \u003cp\u003eEmpirical evidence demonstrates that household access to electricity exerts significant positive effects on educational outcomes. Le\u0026oacute;n Esteban et al. (2018) established that household electricity access produces significant and positive effects on educational attainment, with electrified households consistently recording higher school enrollment, longer study hours, and improved academic performance compared to their unelectrified counterparts. Moreover, Ye (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) found that in Kenya, household electricity usage improves educational attainment not solely through extended study hours and better lighting, but through additional mechanisms including access to digital learning tools, reduced domestic labour burden, and improved household economic productivity. Further, households with electricity access have been shown to record higher adoption of educational digital tools, which correlates with improved academic performance (Boampong et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2026\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGender and socioeconomic inequalities are well-documented moderators of educational attainment in developing countries. Kumar Sarkar et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) document that girls in rural South Asia face compounding barriers including early marriage, patriarchal norms, and gender-insensitive school systems, which reduce their educational progression relative to boys. In Sub-Saharan Africa specifically, Brew-Hammond (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) argues that women and girls bear a disproportionate share of household energy burdens, and that electrification programmes targeting these burdens can generate significant female human capital gains. Vecchione et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) further show that motivational orientations differ by gender, with intrinsic motivation being a stronger predictor of educational outcomes for females, implying that interventions which improve the home study environment, such as reliable evening lighting, may yield larger returns for girls than for boys.\u003c/p\u003e \u003cp\u003eAn expanding corpus of quasi-experimental research substantiates these mechanisms. Dinkelman (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) employs land gradient as a device for electrification expenses in South Africa and reports substantial positive effects on female labour force participation, aligning with the domestic labour reallocation mechanism that has downstream implications for children's education. The most frequently referenced study in this field, conducted by Lipscomb, Mobarak, and Barham (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), utilises a least-cost grid rollout model as an instrument for municipal electrification in Brazil. They discover that a ten percentage point increase in electrification rates correlates with an approximate increase of 0.35 years in average years of schooling, thereby establishing a direct benchmark for the estimates derived in the current study. Rud (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) employs programme eligibility standards to ascertain the beneficial impacts of electrification on school enrolment in rural India, demonstrating more pronounced effects for girls than for boys. Bayer, Dolan, and Urpelainen (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) synthesise research from many contexts and affirm that the beneficial educational outcomes of electrification are consistent, though contextually dependent.\u003c/p\u003e \u003cp\u003eData from Sub-Saharan Africa, characterised by some of the lowest electricity rates worldwide, is pertinent to the current study. Bensch, Kluve, and Peters (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) present experimental results from Rwanda demonstrating that enhanced lighting substantially extends homework duration for primary school students. Van de Walle, Dust, Minh, and Sharma (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) utilise panel data from five African nations, including Ghana, to illustrate the beneficial impacts of electrification on educational spending and attendance, particularly among economically disadvantaged households. Blimpo, Postepska, and Xu (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) demonstrate that electrification rates for impoverished households in West Africa consistently trail those of affluent households, resulting in enduring energy-related educational disparities that are not entirely attributable to income, a trend clearly evident in the GLSS7 data utilised in this study.\u003c/p\u003e \u003cp\u003eThe variability of electricity's educational impacts across gender, socioeconomic status, and age has garnered heightened scrutiny. Brew-Hammond (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) indicates that women and girls in Sub-Saharan Africa disproportionately endure home energy burdens, and that electrification programs aimed at alleviating these burdens offer significant advantages for the building of female human capital. Njiru and Letema (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) present empirical evidence from East Africa indicating that households with electricity access exhibit elevated female school completion rates, which they attribute to the liberation of home time and enhanced study environments. Puzzolo et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) affirm via systematic study that access to sustainable energy yields substantial gender co-benefits for education in low-income contexts. Khandker et al. (2014) and Lipscomb et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) both conclude that the educational benefits of electrification are most significant for households in the lower income bracket, aligning with the expectation that the marginal utility of electricity is greatest for resource-limited families unable to afford alternative lighting or educational resources. Taken together, the literature establishes a robust theoretical and empirical case for electricity\u0026rsquo;s positive educational effects, yet evidence specific to rural Ghana remains limited, heterogeneous effects by poverty status are rarely the primary focus of analysis, and no study has directly compared IV and propensity score matching estimates within a single Ghanaian dataset. This study addresses all three gaps.\u003c/p\u003e"},{"header":"Methodology","content":"\u003cp\u003e3.1. Data and variable definitions\u003c/p\u003e \u003cp\u003eThis study draws on data from the seventh round of the Ghana Living Standards Survey (GLSS7), the most comprehensive nationally representative household survey on living conditions in Ghana, covering education, health, employment, household expenditure, and housing. The GLSS7 comprises 1,000 primary sampling units, with 43.8% in rural and 56.2% in urban areas, spanning 15,000 households (8,430 rural and 6,570 urban). The survey achieved a response rate of 93.4%, yielding usable data on 14,009 households and 59,864 individuals.\u003c/p\u003e \u003cp\u003eThe analytical sample is restricted to individuals aged 6 to 30. Six years is the statutory age at which primary school education begins in Ghana, and 30 years is chosen as the upper bound to focus on the cohort most likely to be in or recently completing formal education, while excluding older individuals for whom electricity access during childhood cannot be meaningfully recovered from cross-sectional data. In this study, educational attainment was measured by the number of years of completed schooling, in accordance with the classification used in GLSS7. In accordance with established conventions in the literature, only the highest attained level of education was recorded for each respondent; incomplete levels were not acknowledged. Years of education were designated as follows: no formal education (0 years), Primary education (6 years), Junior High School/Junior Secondary School (9 years), Polytechnic, Teacher and Nursing Training colleges, or equivalent tertiary institution (15 years), Bachelor\u0026rsquo;s degree (16 years), professional qualification (16 years), and postgraduate degree (17 years). This coding system transforms an ordinal categorical variable into a continuous numeric scale that quantifies the extent of educational disparities among respondents.\u003c/p\u003e \u003cp\u003eThe independent variable for the study was household access to electricity. In this study, we measured household access to electricity as a binary variable with one (1) indicating that the household has access to the national grid and zero (0) indicating that the household does not use the national grid.\u003c/p\u003e \u003cp\u003eThe remaining control variables were chosen based on their established predictive power for educational attainment in the Sub-Saharan African literature (Akabayashi and Psacharopoulos, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Khandker et al., 2014; Neupane, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e delineates the dependent and independent variables along with their respective measurements.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMeasurement of Variables\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeasurement\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDependent Variables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational attainment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHighest level of education completed measured in years of schooling.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHousehold control Variables\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAccess to Electricity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBinary variable equals 1 if household use national grid\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog Household per capita income\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContinuous\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003econtinuous\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of mother\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContinuous\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContinuous\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog\u0026nbsp;of\u0026nbsp;household\u0026nbsp;education expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContinuous\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eIndividual Characteristics\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBinary variable equals 1 if male\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContinuous\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003eAuthors construct\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEmpirical model\u003c/h2\u003e \u003cp\u003eWe begin with an Ordinary Least Squares (OLS) regression as a baseline specification\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:edu(Household\\:electricity,\\:individual\\:characteristics,\\:Income,\\:parental\\:education,\\:$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eEdu\u003csub\u003ei\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;α\u0026thinsp;+\u0026thinsp;βElectricity\u003csub\u003ei\u003c/sub\u003e +X\u003csub\u003ei\u003c/sub\u003e\u0026prime;​γ\u0026thinsp;+\u0026thinsp;HH\u003csub\u003ei\u003c/sub\u003e\u0026prime;​δ\u0026thinsp;+\u0026thinsp;\u0026micro;​+ϵ\u003csub\u003ei​\u003c/sub\u003e (1)\u003c/p\u003e \u003cp\u003eWhere;\u003c/p\u003e \u003cp\u003eEdu\u003csub\u003ei\u003c/sub\u003e is the highest level of education an individual has completed, measured in years.\u003c/p\u003e \u003cp\u003eX captures individual characteristics such as age, sex, relationship to the household head\u003c/p\u003e \u003cp\u003eThe vector, HH, contains household-level control variables: age and sex of the household head, household size, household income, education status of the household head, and household expenditures on education and non-education items\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIdentification\u003c/h3\u003e\n\u003cp\u003eThe OLS estimate of electricity access on educational attainment is subject to endogeneity bias from two primary causes. Initially, unobserved heterogeneity: cognitively proficient parents are concurrently more capable of obtaining electrical connections and imparting academic competence to their offspring, resulting in omitted variable bias. Secondly, reverse causality: households possessing greater human capital may be more adept at advocating for grid expansion or securing private connections, so creating a feedback loop from attainment to access. The cumulative violations of the OLS orthogonality constraint result in a biased and inconsistent β^​. We tackle this using Two-Stage Least Squares (2SLS) estimation, employing the community-level electrification rate as an instrument for household electricity access, the leave-one-out mean proportion of other households within the same enumeration area (EA) that possess electricity access, excluding the household under consideration.\u003c/p\u003e \u003cp\u003eElectricity\u003csub\u003ei\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;α\u0026thinsp;+\u0026thinsp;ΩCluster\u003csub\u003ei\u003c/sub\u003e +X\u003csub\u003ei\u003c/sub\u003e\u0026prime;​γ\u0026thinsp;+\u0026thinsp;HH\u003csub\u003ei\u003c/sub\u003e\u0026prime;​δ\u0026thinsp;+\u0026thinsp;\u0026micro;​+ϵ\u003csub\u003ei​\u003c/sub\u003e (2)\u003c/p\u003e \u003cp\u003eWe estimate Eq.\u0026nbsp;(2) with the ordinary least square model and predict \u003cem\u003eElectricity\u003c/em\u003e ̂. In the second stage, we replace \u003cem\u003eElectricity\u003c/em\u003e in Eq.\u0026nbsp;(1) with its prediction from Eq.\u0026nbsp;(2) and estimate of educational attainment in Eq.\u0026nbsp;(3)\u003c/p\u003e \u003cp\u003eEdu\u003csub\u003ei\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;α\u0026thinsp;+\u0026thinsp;θElectricity\u003csub\u003ei\u003c/sub\u003e ̂ +X\u003csub\u003ei\u003c/sub\u003e\u0026prime;​γ\u0026thinsp;+\u0026thinsp;HH\u003csub\u003ei\u003c/sub\u003e\u0026prime;​δ\u0026thinsp;+\u0026thinsp;\u0026micro;​+ϵ\u003csub\u003ei​​\u003c/sub\u003e (3)\u003c/p\u003e \u003cp\u003eA causal interpretation of θ requires the instrument to satisfy both the relevance condition and the exclusion restriction simultaneously.\u003c/p\u003e \u003cp\u003eThe instrument meets the relevance criterion as grid infrastructure in Ghana is implemented at the community level by ECG and NEDCo, resulting in a robust within-EA correlation in electricity access, which we confirm using the Kleibergen-Paap F-statistic. The exclusion restriction is maintained as the decisions about ECG and NEDCo grid extensions are influenced by engineering and load-forecasting factors specifically, proximity to transmission lines and community size, rather than the educational attributes of the resident families. Consequently, there exists no reliable independent correlation between neighbors\u0026rsquo; power availability and a child\u0026rsquo;s completed years of education.\u003c/p\u003e \u003cp\u003eSecondly, we do a falsification test utilising the age of the household head as a placebo outcome, which should remain unaffected by an individuals access to electricity. The IV second stage yields a negligible coefficient on this placebo result (coefficient: -0.783, standard error 1.064, p\u0026thinsp;=\u0026thinsp;0.462). This null result corroborates the exogeneity of the instrument, indicating that the variance in electricity availability is unlikely to be influenced by unobserved household factors associated with age. The instrument therefore does not predict pre-determined demographic outcomes, lending further credibility to the identifying assumptions.\u003c/p\u003e \u003cp\u003eThe study used propensity score matching (PSM) alongside instrumental variable estimations to mitigate potential endogeneity, as noted in previous research (Kofinti et al, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The treatment variable in this study is access to electricity, utilised to evaluate the average treatment effect on educational attainment. The technique generates an estimate to assess the counterfactual effect of access to electricity on educational attainment. We employ five matching strategies to conduct sensitivity tests on our findings: Kernel Matching, Logit\u0026thinsp;+\u0026thinsp;Kernel Matching, Nearest Neighbour (one-to-one)and Radius matching methods.\u003c/p\u003e\n\u003ch3\u003eDescriptive Statistics\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the descriptive statistics of the primary variables of the study and examines the disparities between homes with access to the national grid and those without. The estimation sample consists of 3,219 individuals, including 1,157 (36%) from houses without access to electricity and 2,062 (64%) from households having electricity. Households with electricity have an average educational attainment of 8 years, but those without electricity average 7 years, and this disparity is statistically significant. Concerning gender, there is essentially no distinction between houses with electricity and those without. Individuals in houses with electricity tend to be marginally older than those in households without power. Households with electricity exhibit higher per capita income and allocate more funds to education compared to those lacking power access. Households lacking electricity exhibit bigger sizes than those with electricity. We additionally observe that both fathers and mothers of individuals from households with electricity possess greater levels of education than those from households without electricity. Ultimately, heads of families lacking electricity are, on average, older than those with electricity access, and all these disparities are statistically significant at the 1% level, with the exception of gender.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eEducational attainment\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo Electricity\u003c/p\u003e \u003cp\u003e(N\u0026thinsp;=\u0026thinsp;1,157)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eElectricity\u003c/p\u003e \u003cp\u003e(N\u0026thinsp;=\u0026thinsp;2,062)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDifference\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.93\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.03\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-1.10\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAccess to electricity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.634\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e18.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-1.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog of household per capita income\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog of educational expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYears of Father\u0026rsquo;s education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-2.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYears of mothers education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-1.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e50.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e47.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eSource: Computed from GLSS 7\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003ePrior to the regression analysis, we conducted a brief descriptive analysis of the distribution of electricity across Ghana. In 2017, electricity remained unequally distributed across Ghana, with only 32 per cent of rural households connected to the national grid, while 68 per cent of urban households were connected to the national grid.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eModel Results\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents the OLS results of the effects of electricity on educational attainment. The results reveal that access to electricity (national grid) in rural Ghana has a statistically significant positive effect on educational attainment (0.239, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). This implies that, holding all other factors constant, individuals aged 6 to 30 in rural households with electricity attain 0.24 years of education more than individuals of the same age group in rural households without access to electricity. This confirms our a priori hypothesis that electricity access enhances educational attainment, possibly through extended hours of study, access to information through television, radio, and digital learning platforms. The results further revealed that being male is associated with 0.63 additional years of schooling compared to females (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). While statistically significant, this modest gap suggests that gender barriers to educational progression in rural Ghana are driven by structural factors that household-level controls only partially capture. Each additional year of age is associated with 0.2 more years of completed schooling, consistent with the cumulative nature of educational attainment. Moreover, the results revealed that a one percent increase in household per capita income is associated with 0.14 additional years of schooling. Wealthier households can afford learning materials such as textbooks, computers, furniture, school uniforms, school fees, and home tuition, which translates into higher educational attainment. We found that educational expenditure was the biggest predictor of educational attainment. A one percent increase in household education expenditure is associated with 0.42 additional years of schooling, making it the strongest predictor among the controls. Household size reduces educational attainment by 0.11 years. This affirms that larger households have lower per capita resources needed to spend on education and electricity.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eEffects of access to electricity on educational attainment in Ghana (Baseline-OLS)\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffect of electricity on educational attainment\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOLS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNational electricity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.239\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.110)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.626\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.110)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.200\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog Household Income Per Capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.136\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.033)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog of educational expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.423\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.041)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.110\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Mother)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.029\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.015)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.012\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.199\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.422)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3219.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR squared\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.340\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eStandard errors in parentheses \u003csup\u003e*\u003c/sup\u003e \u003cem\u003ep\u003c/em\u003e \u0026lt; .1, \u003csup\u003e**\u003c/sup\u003e \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u003csup\u003e***\u003c/sup\u003e \u003cem\u003ep\u003c/em\u003e \u0026lt; .01\u003c/p\u003e \u003cp\u003eSource: Computed from GLSS 7\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEffects of access to electricity on educational attainment in Ghana (IV)\u003c/h3\u003e\n\u003cp\u003eAs discussed in Section 3, the OLS estimate of electricity\u0026rsquo;s effect on educational attainment is subject to endogeneity bias from omitted variable bias and reverse causality. To address this, we instrument household electricity access with the leave-one-out community-level electrification rate within each enumeration area.\u003c/p\u003e \u003cp\u003eThe 2SLS results indicate that electricity access increases completed years of schooling by 0.52 years (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), approximately double the OLS estimate of 0.24 years. This upward revision relative to OLS is consistent with two mechanisms discussed in Section 3: classical attenuation bias from measurement error in the binary electricity indicator, and a Local Average Treatment Effect (LATE) that captures households with electricity at the margin of grid access, who may derive particularly large educational returns from connection.\u003c/p\u003e \u003cp\u003eThe OLS estimate of electricity access on educational attainment is influenced by endogeneity bias from two main sources: omitted variable bias due to the positive correlation between household socioeconomic status and both electricity access and educational investment, and reverse causality, where households with greater human capital may be more capable of advocating for grid connection. Both sources generally induce an upward bias in Ordinary Least Squares (OLS). Nonetheless, our IV estimates surpass the OLS estimates, a trend we view as aligned with two supplementary causes. The measurement error in the binary electrical access variable, which just reflects grid connection status instead of duration or dependability of service, may diminish OLS coefficients towards zero. Secondly, the instrument determines a Local Average Treatment Effect (LATE) for compliers, specifically homes whose connection status is influenced by community-level electrification. These homes are on the periphery of grid access and may experience significant educational advantages from connection, resulting in a Local Average Treatment Effect (LATE) that surpasses the overall population average treatment effect. Disentangling these two mechanisms is beyond the scope of the present study and remains a direction for future research.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffects of access to electricity on educational attainment in Ghana (IV)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIV\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNational electricity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.521\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.172)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.623\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.110)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.199\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.008)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog of Household Income Per Capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.125\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.034)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog of expenditure on education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.403\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.042)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.108\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Mother\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.027\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.015)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.012\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.189\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.422)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3219.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR squared\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.338\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIdstat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e651.259\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIdp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWidstat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1651.270\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003eSource: Computed from GLSS 7\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003eHeterogeneity analysis\u003c/h3\u003e\n\u003cp\u003eWe evaluate diverse effects across two dimensions for which interacting instrumental variable estimates are accessible: gender and socioeconomic status (poverty classification). The heterogeneity is analysed by interacting the instrumented electricity access variable with each moderating characteristic in the second stage of the 2SLS, with the aggregated subgroup effect calculated through a linear combination (lincom) of the primary electricity coefficient and the pertinent interaction term.\u003c/p\u003e \u003cp\u003eGender Heterogeneity\u003c/p\u003e \u003cp\u003eColumn (1) of Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e indicates that the baseline effect of electricity access on male years of schooling is 0.585 years (SE: 0.267, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The interaction term between electricity and female status is negative yet statistically insignificant (coefficient: -0.117, SE: 0.317, p\u0026thinsp;=\u0026thinsp;0.711), suggesting that females do not experience a statistically distinct effect compared to boys. The cumulative electricity effect for females, derived from the linear combination of the primary electricity coefficient and the interaction term, is 0.468 years (SE: 0.285, p\u0026thinsp;=\u0026thinsp;0.101), which is just below conventional significance levels. The data indicate that although access to electricity enhances educational attainment for both boys and girls, the extent of this effect does not significantly vary by gender in this sample. This outcome diverges from earlier anticipations that girls would gain disproportionately from electricity via increased study hours and less household labour responsibilities. An acceptable view is that the structural and social impediments limiting females\u0026rsquo; education in rural Ghana, such as early marriage, cultural norms, and distance to educational institutions, are so deeply rooted that mere access to electricity cannot produce a significant advantage. Consequently, complementary gender-specific interventions may be necessary in conjunction with electrification to completely bridge the gender gap in educational achievement.\u003c/p\u003e \u003cp\u003eHeterogeneity of poverty\u003c/p\u003e \u003cp\u003eColumn (2) of Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e delineates the poverty heterogeneity specification, categorising poor families as those situated inside the lowest two income quintiles. The baseline electricity effect for non-impoverished households is minimal and statistically negligible (coefficient: 0.134, SE: 0.231, p\u0026thinsp;=\u0026thinsp;0.562). The relationship between electricity access and impoverished household status is positive and statistically significant (coefficient: 1.038, SE: 0.486, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), suggesting that impoverished households get considerably greater educational benefits from power access compared to their non-impoverished counterparts. The cumulative impact of electricity on impoverished households, calculated as the linear amalgamation of primary electricity and interaction coefficients, amounts to 1.172 years of supplementary education (SE: 0.429, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). This is a substantial and accurately quantified effect, approximately twice the baseline estimate, and constitutes the most significant discovery in the heterogeneity study. This outcome aligns with the view that electricity serves as a partial equaliser: households with less resources have the most significant obstacles to studying and being productive after dark, hence electrification yields the greatest marginal benefits for these households. The discovery that the non-poor baseline effect is negligible further substantiates this interpretation. Households with more resources can more effectively compensate for the lack of electricity through alternatives like candles, generators, or private tutoring, thereby mitigating the marginal effect of grid access. The substantial negative coefficient for the poor indicator (-0.688, SE: 0.334, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) independently corroborates the current educational disadvantage experienced by impoverished households, highlighting the critical role of electrification as a focused development tool.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eHeterogeneous Effects of Electricity on Years of Education: Interacted IV\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePoverty\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNational electricity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.585\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.134\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.597\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.267)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.231)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.300)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElectr*female\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.317)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElectr*poor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.038\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.486)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElectr*young\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.302\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.309)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003efemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.537\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.254)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehousehold in bottom 2 income quintiles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.688\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.334)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eaged 6\u0026ndash;15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-1.884\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.260)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.199\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.195\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.075\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.009)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.009)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.014)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog household Income Per Capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.126\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.132\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.112\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.039)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.060)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.038)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog total Education Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.404\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.403\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.388\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.050)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.049)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.050)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.108\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.108\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.114\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.014)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.014)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.014)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Mother\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.034\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.017)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.017)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.017)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.012\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.013\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.013\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.003)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.373\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.364\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.343\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.489)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.603)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.516)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSubgroup effect\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.468\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.172\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.295\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(SE)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.429\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.229\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[p-value]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.101\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.198\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3219\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3219\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3219\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKP F-stat (1st stage)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e107.400\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e60.149\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e92.590\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAR stat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e123.217\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85.186\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e87.346\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAR p-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eClustered standard errors in parentheses. * p\u0026thinsp;\u0026lt;\u0026thinsp;0.10, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01. Subgroup effect\u0026thinsp;=\u0026thinsp;lincom of nat_electr\u0026thinsp;+\u0026thinsp;interaction term.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSubgroup Heterogeneity Analysis\u003c/h2\u003e \u003cp\u003eTo examine the distributional nature of the baseline estimate, we conduct distinct subgroup IV regressions across six mutually exclusive or complimentary categories: females, males, impoverished families, non-impoverished households, children aged 6\u0026ndash;15, and young people aged 16\u0026ndash;30. This method enhances the interacting IV specifications by permitting all slope coefficients, not solely the electricity effect, to fluctuate between subgroups, thus circumventing the limiting homogeneity assumptions associated with a singular pooled interaction factor.\u003c/p\u003e \u003cp\u003eThe results demonstrate a markedly diverse pattern that is both economically significant and statistically rigorous. The most accurately calculated and significantly large effect is noted among impoverished households (column 3), where access to electricity increases years of schooling by 1.206 years (SE: 0.397, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). This figure exceeds double the equivalent estimate for non-poor families (0.176, SE: 0.244, p\u0026thinsp;=\u0026thinsp;0.470), which is economically insignificant and statistically indistinguishable from zero. The distinction is clear and logically sound: non-poor households can compensate for the lack of grid electricity with generators, battery lighting, and private tuition, thus diminishing the marginal educational benefits of electrification. In contrast, impoverished households encounter stringent limitations on study time during the evening and are disproportionately dependent on domestic child labour, both of which electricity alleviates directly. This discovery establishes electricity access as a true equalising factor inside the Ghanaian education system, yielding the greatest benefits precisely in areas of severe educational deprivation.\u003c/p\u003e \u003cp\u003eThe gender disaggregation is also instructive. The electrical impact for girls (0.540, SE: 0.279, p\u0026thinsp;\u0026lt;\u0026thinsp;0.10) and males (0.583, SE: 0.272, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) is statistically comparable, as seen by overlapping confidence intervals and a coefficient difference of less than 0.05 years. The pronounced negative baseline disparity in girls\u0026rsquo; education identified in the interacted specification due to deeply rooted social norms, early marriage, and cultural role expectations, persist regardless of electricity access, indicating that electrification alone cannot eradicate the structural factors contributing to the gender gap. This null result is a significant finding: it challenges the optimistic assumption that electricity would disproportionately benefit girls by freeing them from nighttime domestic restrictions, and underscores the need for complementary demand-side interventions targeting the structural roots of the gender gap. The results of the age subgroup results are also instructive. Children aged 6 to 15 experience a substantial and precisely estimated effect of 0.543 years (SE: 0.184, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01); the Kleibergen-Paap Wald F-statistic of 297 for this subsample is well above standard weak-instrument thresholds.\u003c/p\u003e \u003cp\u003eThe older group (ages 16\u0026ndash;30) produces a similar point estimate of 0.528 years (SE: 0.305, p\u0026thinsp;\u0026lt;\u0026thinsp;0.10), albeit with significantly broader confidence intervals, indicating increased variability in educational pathways among young individuals. The sustained economic relevance of the effect for the older generation aligns with the ongoing prevalence of extended education, adult literacy initiatives, and vocational training in rural Ghana, where delayed educational advancement is prevalent. The comparable magnitudes across age groups refute the notion of a singular study-lighting channel as the exclusive mechanism and endorse a more comprehensive interpretation that includes reduced child labour, increased household productivity, and improved access to educational media.\u003c/p\u003e \u003cp\u003eThe subgroup analysis indicates that poverty is the principal factor of heterogeneity in the educational returns of electricity. The targeted electrification of impoverished rural areas in Ghana is expected to yield significant educational benefits, positively impacting various genders and age groups within such communities. We present a graph of the coefficients of heterogeneity in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eHeterogeneous Effects of Electricity: Subgroup IV Regressions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(4)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(5)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e(6)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFemales\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMales\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePoor Household\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNon-Poor household\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAge 6\u0026ndash;15\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eAge 16\u0026ndash;30\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNational electricity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.540\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.583\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.206\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.176\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.543\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.528\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.279)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.272)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.397)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.244)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.184)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.305)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.170\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.242\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.176\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.204\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.422\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.011\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.015)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.011)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.025)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.019)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog Househol Income Per Capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.109\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.136\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.106\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.162\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.050)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.051)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.111)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.069)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.046)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.053)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog Total Expenditure on education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.347\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.471\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.227\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.493\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.106\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.522\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.061)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.066)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.067)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.065)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.042)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.070)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehould size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.104\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.106\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.148\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.034\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.181\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.029)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.025)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.032)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.027)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.019)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.036)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.040\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.019)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.013)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.020)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational level of Mother\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.052\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.024)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.022)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.025)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.026)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.012\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.012\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.013\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.013\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.004)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.827\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.891\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.985\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.018\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.107\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e4.726\u003csup\u003e***\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e(0.632)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e(0.580)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e(0.767)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e(0.715)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e(0.556)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e(0.713)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1849.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1370.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1211.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2008.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1463.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1756.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewidstat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e274.948\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e302.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e117.478\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e285.694\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e297.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e239.566\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eStandard errors in parentheses \u003csup\u003e*\u003c/sup\u003e \u003cem\u003ep\u003c/em\u003e \u0026lt; .1, \u003csup\u003e**\u003c/sup\u003e \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u003csup\u003e***\u003c/sup\u003e \u003cem\u003ep\u003c/em\u003e \u0026lt; .01\u003c/p\u003e \u003cp\u003eSource: Computed from GLSS 7\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003ePropensity Score Matching Analysis\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e presents PSM estimates as a robustness check against the IV strategy. We employ five matching algorithms, kernel, logit-kernel, nearest neighbour, radius, and bootstrapped kernel, to assess convergence across identification approaches. Across all five specifications, the ATT estimates range from 0.038 to 0.181 years and are statistically indistinguishable from zero, falling far short of the IV estimate of 0.52 years. This divergence is theoretically expected: PSM eliminates bias only from observable confounders, whereas the IV approach additionally removes bias from unobservable household traits, such as parental motivation and social networks, that simultaneously drive electricity access and educational investment. The gap between the two estimators therefore reinforces our preferred IV approach and confirms that instrument-based identification is essential in this setting.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePropensity Score Matching using different matching algorithms\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMatching Algorithm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eATT\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKernel Matching (Baseline)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003cp\u003e(0.170)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,219\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLogit\u0026thinsp;+\u0026thinsp;Kernel Matching\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.046\u003c/p\u003e \u003cp\u003e(0.177)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,219\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNearest Neighbour (one-to-one)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003cp\u003e(0.170)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,218\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRadius\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.1807\u003c/p\u003e \u003cp\u003e(0.135)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,218\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBootstrapped Kernel PSM (50 replications)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003cp\u003e(0.200)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3,219\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003eNotes: Standard errors in parentheses. *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.10. ATT\u0026thinsp;=\u0026thinsp;Average Treatment Effect on the Treated. All specifications use national electricity grid access as the treatment variable and years of education as the outcome variable. Propensity scores estimated using probit unless otherwise stated.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePlacebo Test:Effect of Electricity on Household Head Age\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1) Outcome: Household Head Age\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElectricity (nat_electr)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.783 (1.064)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.407 (0.748)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (individual)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.163*** (0.053)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog HH Income per capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-1.139*** (0.366)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLog Education Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.592* (0.311)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold Size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.742*** (0.146)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFather\u0026rsquo;s Education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.093 (0.088)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMother\u0026rsquo;s Education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.136 (0.102)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e67.707*** (3.353)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003eObservations: 3,930\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003eClusters: 545\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003eControls: Yes\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003eClustered standard errors in parentheses. * p\u0026thinsp;\u0026lt;\u0026thinsp;0.10, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e "},{"header":"Conclusion and Policy Recommendations","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003cp\u003eThis study provides credible causal evidence that household access to grid electricity significantly enhances educational attainment in rural Ghana. Using a Two-Stage Least Squares strategy instrumented by the leave-one-out community-level electrification rate, we find that electricity access raises completed years of schooling by 0.52 years, approximately twice the na\u0026iuml;ve OLS estimate of 0.24 years, confirming that conventional regression approaches substantially understate the true educational returns to electrification. Heterogeneity analysis reveals that poverty status is the principal source of differential effects: poor households gain over one additional year of schooling (1.17 years, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), while the effect for non-poor households is negligible and statistically insignificant (0.13 years). Gender disaggregation yields broadly comparable effects for boys and girls, indicating that deep-rooted structural barriers. early marriage, patriarchal norms, and distance to school, persist independently of electricity access and require complementary demand-side interventions. Effects are consistently significant for children aged 6\u0026ndash;15 and older youth (16\u0026ndash;30), suggesting that mechanisms beyond study-lighting, including domestic labour reallocation and access to educational media, are operative across the lifecycle.\u003c/p\u003e \u003cp\u003eThese findings carry direct policy implications. First, rural electrification should be explicitly integrated into Ghana\u0026rsquo;s human capital development agenda, with targeted grid expansion prioritising communities in the lowest income quintiles where educational returns are largest. Second, since electrification alone does not close the gender gap, complementary interventions, conditional cash transfers, school proximity programmes, and community-based gender sensitisation, must accompany electrification rollouts to translate energy access into equitable educational outcomes. These results position targeted rural electrification as a high-return, inequality-reducing policy instrument with broad implications for Sub-Saharan Africa\u0026rsquo;s pursuit of SDG 4. Several limitations of the present study merit acknowledgement. First, the GLSS7 is cross-sectional, precluding the estimation of within-individual or within-household fixed effects; the IV strategy mitigates but does not fully eliminate concerns about time-invariant unobservables. Second, the binary measure of grid connection does not capture the reliability, duration, or quality of electricity supply, which may attenuate estimated effects. Third, the PSM results are consistently small and statistically insignificant, suggesting residual confounding from unobservable household traits that matching cannot address \u0026mdash; a finding that reinforces the superiority of the IV approach in this context. Future research employing panel data and more granular electricity quality measures would further sharpen our understanding of electrification\u0026rsquo;s educational returns across Sub-Saharan Africa.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConsent to Participate:\u003c/h2\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent to Publish:\u003c/strong\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eEthics approval:\u003c/strong\u003e \u003cp\u003eThis study used publicly available secondary data from the Ghana Living Standards Survey Round 7 (GLSS7). No ethics approval was required as no primary data collection involving human subjects was conducted.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCompeting interests:\u003c/strong\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding:\u003c/h2\u003e \u003cp\u003eThis research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eBFE: Conceptualisation, methodology, formal analysis, writing of original draft. DTS: Conceptualisation, data curation, writing of review literature and editing. all aruhors reviews the manuscript\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data used in this study are from the Ghana Living Standards Survey Round 7 (GLSS7), publicly available from the Ghana Statistical Service at www.statsghana.gov.gh\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAhali AY, Yaw F. Household energy demand in Ghana: The role of household characteristics. Energy Policy. 2016;94:235\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAkabayashi H, Psacharopoulos G. The trade-off between child labour and human capital formation: A Tanzanian case study. J Dev Stud. 1999;35(5):120\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarro RJ, Lee JW. A new data set of educational attainment in the world, 1950\u0026ndash;2010. J Dev Econ. 2013;104:184\u0026ndash;98. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jdeveco.2012.10.001\u003c/span\u003e\u003cspan address=\"10.1016/j.jdeveco.2012.10.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBayer P, Dolan L, Urpelainen J. The effect of household electricity access on women\u0026rsquo;s empowerment: Evidence from rural India. World Dev. 2020;130:104895. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.worlddev.2020.104895\u003c/span\u003e\u003cspan address=\"10.1016/j.worlddev.2020.104895\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBecker GS. Human capital: A theoretical and empirical analysis, with special reference to education. University of Chicago Press; 1964.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBensch G, Kluve J, Peters J. Impacts of rural electrification in Rwanda. J Dev Eff. 2011;3(4):567\u0026ndash;88. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/19439342.2011.629521\u003c/span\u003e\u003cspan address=\"10.1080/19439342.2011.629521\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlimpo MP, Postepska A, Xu Y. Why is household electricity uptake low in Sub-Saharan Africa? World Dev. 2020;133:105002. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.worlddev.2020.105002\u003c/span\u003e\u003cspan address=\"10.1016/j.worlddev.2020.105002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoampong O, Frimpong EB, Dadzie TS. (2026). Digital learning tools, electricity access and academic performance in rural Ghana. J Afr Educ. (Forthcoming).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrew-Hammond A. Energy access in Africa: Challenges ahead. Energy Policy. 2010;38(5):2291\u0026ndash;301. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.enpol.2009.12.016\u003c/span\u003e\u003cspan address=\"10.1016/j.enpol.2009.12.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDinkelman T. The effects of rural electrification on employment: New evidence from South Africa. Am Econ Rev. 2011;101(7):3078\u0026ndash;108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/aer.101.7.3078\u003c/span\u003e\u003cspan address=\"10.1257/aer.101.7.3078\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFessler P, Schneebaum A. Gender and educational attainment across generations in Austria. Fem Econ. 2012;18(1):161\u0026ndash;88. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13545701.2011.637759\u003c/span\u003e\u003cspan address=\"10.1080/13545701.2011.637759\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGamette P, Asante FA, Amoako-Tuffour J. Electrification and welfare outcomes in West Africa: A cross-country assessment. Energy Sustain Dev. 2024;78:101370.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhana Statistical Service. Ghana Living Standards Survey Round 7 (GLSS 7): Main report. Ghana Statistical Service; 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreenstone M, Jack BK. Envirodevonomics: A research agenda for an emerging field. J Econ Lit. 2015;53(1):5\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/jel.53.1.5\u003c/span\u003e\u003cspan address=\"10.1257/jel.53.1.5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGronau R. Leisure, home production, and work: The theory of the allocation of time revisited. J Polit Econ. 1977;85(6):1099\u0026ndash;123. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1086/260629\u003c/span\u003e\u003cspan address=\"10.1086/260629\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanagawa M, Nakata T. Assessment of access to electricity and the socioeconomic impacts in rural areas of developing countries. Energy Policy. 2008;36(6):2016\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.enpol.2008.01.041\u003c/span\u003e\u003cspan address=\"10.1016/j.enpol.2008.01.041\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhandker SR, Barnes DF, Samad HA. Welfare impacts of rural electrification: A panel data analysis from Vietnam. Econ Dev Cult Change. 2013;62(4):659\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1086/676930\u003c/span\u003e\u003cspan address=\"10.1086/676930\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKofinti RE, Baako-Amponsah J, Danso P. Household National Health Insurance Subscription and Learning Outcomes of Poor Children in Ghana. Child Indic Res. 2022;1\u0026ndash;38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12187-022-09980-y\u003c/span\u003e\u003cspan address=\"10.1007/s12187-022-09980-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar Sarkar S, Islam Khatun M, Ray S. Girls\u0026rsquo; education in developing countries: Issues and challenges. Int J Educ Res. 2014;2(7):343\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLipscomb M, Mobarak AM, Barham T. Development effects of electrification: Evidence from the geologic placement of hydropower plants in Brazil. Am Economic Journal: Appl Econ. 2013;5(2):200\u0026ndash;31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1257/app.5.2.200\u003c/span\u003e\u003cspan address=\"10.1257/app.5.2.200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMamuli LC. The impact of higher education on human capital development in developing countries. Int J Educational Sci. 2020;28(1\u0026ndash;3):1\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.31901/24566322.2020/28.1-3.1175\u003c/span\u003e\u003cspan address=\"10.31901/24566322.2020/28.1-3.1175\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinistry of Energy. National Electrification Scheme (NES)-Master Plan Review (2011\u0026ndash;2020); Ministry of Energy. Accra,Ghana; 2010.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMurillo FJ, Rom\u0026aacute;n M. Latin America: School bullying and academic achievement. CEPAL Rev. 2011;104:37\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeupane S. Factors affecting educational attainment in developing countries. Int J Social Sci Manage. 2017;4(3):189\u0026ndash;96. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3126/ijssm.v4i3.17423\u003c/span\u003e\u003cspan address=\"10.3126/ijssm.v4i3.17423\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNjiru C, Letema SC. (2018). Energy poverty and its implication on standard of living in Kirinyaga, Kenya. \u003cem\u003eJournal of Energy, 2018\u003c/em\u003e, 3196567. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1155/2018/3196567\u003c/span\u003e\u003cspan address=\"10.1155/2018/3196567\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePuzzolo E, Pope D, Stanistreet D, Rehfuess EA, Bruce NG. Clean fuels for resource-poor settings: A systematic review of barriers and enablers to adoption and sustained use. Environ Res. 2016;146:218\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envres.2016.01.015\u003c/span\u003e\u003cspan address=\"10.1016/j.envres.2016.01.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRud JP. Electricity provision and industrial development: Evidence from India. J Dev Econ. 2012;97(2):352\u0026ndash;67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jdeveco.2011.06.010\u003c/span\u003e\u003cspan address=\"10.1016/j.jdeveco.2011.06.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchultz TW. Investment in human capital. Am Econ Rev. 1961;51(1):1\u0026ndash;17.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan de Walle D, Dust M, Minh NT, Sharma M. Poor households\u0026rsquo; electricity connections and use in five African countries. World Bank Economic Rev. 2017;31(3):615\u0026ndash;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/wber/lhw020\u003c/span\u003e\u003cspan address=\"10.1093/wber/lhw020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVecchione M, Alessandri G, Marsicano G, Caprara GV. Academic motivation predicts educational attainment: Does gender make a difference? Learn Individual Differences. 2014;32:124\u0026ndash;31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.lindif.2014.01.003\u003c/span\u003e\u003cspan address=\"10.1016/j.lindif.2014.01.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYe J. Household electricity use and children\u0026rsquo;s educational outcomes in rural Kenya. Energy Sustain Dev. 2017;37:45\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.esd.2017.01.003\u003c/span\u003e\u003cspan address=\"10.1016/j.esd.2017.01.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diedu","sideBox":"Learn more about [Discover Education](https://www.springer.com/journal/44217)","snPcode":"44217","submissionUrl":"https://submission.nature.com/new-submission/44217/3","title":"Discover Education","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9393811/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9393811/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study examines the causal relationship between houshold electricity access and educational achievement in rural Ghana. Utilising data from the seventh iteration of the Ghana Living Standards Survey (GLSS7), we assess the impact of home electricity access on the years of completed education for individuals aged 6 to 30 in rural regions. Ordinary Least Squares (OLS) regressions provide a baseline, demonstrating a positive and statistically significant effect of an additional 0.24 years of education. To mitigate endogeneity caused by omitted variable bias and reverse causality, we utilise a Two-Stage Least Squares (2SLS) instrumental variable approach. The IV estimates indicate that access to electricity enhances educational attainment by 0.52 years. Heterogeneity analysis employing interacted instrumental variable specifications indicates that the poverty gradient is the most reliable source of differential effects: the overall impact of electricity access on poor households is 1.17 additional years of education statistically significant at the 1 percent level in contrast to a minimal and insignificant effect of 0.13 years for non-poor households, highlighting electricity\u0026rsquo;s function as an equalising force for the most resource-deprived families. The findings highlight the significance of rural electrification as a policy tool for bridging poverty-related educational attainment disparities in Sub-Saharan Africa, particularly benefiting the most impoverished households.\u003c/p\u003e","manuscriptTitle":"The Effects of Household Electricity Access on Educational Attainment In Rural Ghana","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-05 20:28:28","doi":"10.21203/rs.3.rs-9393811/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-14T03:34:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-10T19:47:09+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T05:04:04+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T21:28:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T11:18:17+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T01:51:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-02T15:17:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"84978560162626350289992512530808068293","date":"2026-04-27T01:02:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"329740761080864089878059031628689927474","date":"2026-04-24T18:19:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"287952419437440768169351117270593838680","date":"2026-04-24T11:23:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"217343145758815968743617762661870257845","date":"2026-04-24T11:02:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"13677195049201259367821096996714145241","date":"2026-04-24T10:14:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"332071100346354455413342138140135285905","date":"2026-04-24T06:45:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-24T06:08:58+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-22T07:32:35+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-15T05:50:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-15T05:49:57+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Education","date":"2026-04-12T11:24:15+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diedu","sideBox":"Learn more about [Discover Education](https://www.springer.com/journal/44217)","snPcode":"44217","submissionUrl":"https://submission.nature.com/new-submission/44217/3","title":"Discover Education","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"826740e2-7445-41b0-b5b3-27c1c05f106a","owner":[],"postedDate":"May 5th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-14T03:34:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-10T19:47:09+00:00","index":50,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-07T05:04:04+00:00","index":49,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T21:28:39+00:00","index":48,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T11:18:17+00:00","index":47,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T01:51:36+00:00","index":46,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-02T15:17:59+00:00","index":45,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-14T03:39:22+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-05 20:28:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9393811","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9393811","identity":"rs-9393811","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}