Data Collectionhard
0:00.0
A panel survey tracks 1000 workers for 5 years. Response rates: Year 1 = 100%, Year 2 = 95%, Year 3 = 89%, Year 4 = 80%, Year 5 = 72%. Analysis reveals: workers with lower education attrit at 20% per year; higher education workers attrit at 8% per year. Wage growth in the population is similar across education levels (mean ~3% annually), but lower-educated workers tend to have lower wages initially.
What is the most severe bias introduced by this attrition pattern?
A panel survey tracks 1000 workers for 5 years. Response rates: Year 1 = 100%, Year 2 = 95%, Year 3 = 89%, Year 4 = 80%, Year 5 = 72%. Analysis reveals: workers with lower education attrit at 20% per year; higher education workers attrit at 8% per year. Wage growth in the population is similar across education levels (mean ~3% annually), but lower-educated workers tend to have lower wages initially.
What is the most severe bias introduced by this attrition pattern?