Katkat
Back

Test regional response-rate differences rigorously

ThumbtackData Scientist · Onsite
MediumData ScientistOnsiteStatistics & Math

Last updated

This question evaluates applied statistical modeling and causal-inference competencies, including adjustment for confounding, analysis of proportion/count outcomes, variance estimation under clustering, multiple-comparison control, diagnostic detection of Simpson’s paradox, and clear effect-size reporting; it is categorized under Statistics & Math and emphasizes practical application with necessary conceptual understanding. Such problems are commonly asked to assess the ability to justify and interpret comparative analyses that account for covariate mix, temporal structure, and multiple-testing concerns, and this summary is in English.

Using the same job-level data (job_category, region, invitations_sent, provider_responses, created_at), determine whether response rates differ by region after adjusting for job_category mix and time. Specify and justify a statistical approach (e.g., logistic regression with response as successes/tries, or a mixed-effects model with region random effects). Include: (1) model formula, link function, and any fixed effects (job_category, month) and interactions; (2) how you will cluster or regularize standard errors; (3) a multiple-comparison strategy across many regions (e.g., Holm–Bonferroni or Benjamini–Hochberg on region coefficients); (4) a diagnostic to detect Simpson’s paradox driven by job_category mix; and (5) how you would report effect sizes and practical significance, not just p-values.