Reframing Short-Term Economic Research: A Mixed Methods Case Study on Inflation and Income Inequality in the U.S.

Short-term research matters today because: 1) pre-pandemic data may obscure new paradigms; 2) new studies should build on rather than duplicate long-term work; and 3) timely policy is crucial. However, short-term studies involve limited data. We examine inflation’s impact on income groups to explore methodological combinations and sequencing for short-term research. First, we assess the effectiveness of quantitative (multiple and quantile regression) and qualitative (thematic analysis of reports) approaches. We evaluate how sequencing methods influences insights, avoiding researcher bias towards preferred sequences by prompting a large language model (LLM) multiple times to interpret the combined results in each sequence: quant→qual and qual→quant. We then compare outputs. Next, we conduct small-scale, open- and closed-ended surveys and repeat the LLM sequencing experiment, this time with closed→open-ended and open→closed-ended instructions. Quantitative models link income to debt and unemployment; qualitative findings show varied views on inflation control and consumer behavior across groups. Surveys show inflation shapes coping and policy trust—especially for middle-income earners, whose unique struggles are lost when grouped with low-income peers by analyses. The quant→qual and closed→open-ended sequences provided more insightful outputs with causal connections and contextualization. Theoretically, this paper provides methodological guidelines; practically, it provides policy insights.