Executing a Fully Crossed Stratified Sampling Design Exploring Fear of Happiness, Intersectional Discrimination, and Depression

Though online sampling methods have progressed, psychology literature continues to include primarily white, educated, upper-class samples. This work will outline the methodology of a stratified community study examining demographic differences in fear of happiness and depression among U.S. adults. Drawing on prior research suggesting variation in emotional processes across social identities, we implemented a fully crossed stratified sampling design to ensure equal representation across race/ethnicity (Black, Asian, Latine, white), gender (cisgender male, cisgender female), and socioeconomic status (SES). Eligible participants were 18-65 years old and residing in the U.S. SES categories were defined as ≤ $50,000 (low), $50,000 - $100,000 (middle), and ≥ $100,000 (high). A priori power analysis (G*Power) indicated a target of 42 participants per cell (N=1,008) to detect medium effects. Following the exclusion of 75 uncategorizable cases, the per-cell target was adjusted to 38, maintaining adequate power for medium-to-large effects. Participants were recruited in SES-based blocks via Prolific. Procedures to preserve stratification, confidentiality, and analytic rigor will be discussed, highlighting methodological considerations for equity-focused psychological research. These methods can ensure adequate power and representation when examining relationships between intersectional identities and psychological phenomena in online samples.