Collecting the perfect data to answer a research question is optimal but not always feasible. Sometimes existing data can be leveraged to answer some questions, but this opportunity comes with important potential pitfalls. To test the effect of early life exposures such as being born preterm or living in a materially deprived neighborhood on early…

This case study evaluated the impact of science, technology, engineering, and mathematics initiatives in improving educational outcomes in the United Kingdom. Using secondary data in the form of official datasets approved by the national pupil database, Year 7 cohort was longitudinally followed from the beginning of Key Stage 3 till the end of Key…

This case study presents an example of using a generalized linear model with a log-linear link to calculate an adjusted risk ratio to be able to assess the association between educational attainment and obesity in a cohort study of American adults. Both risk ratios and odds ratios can be calculated based on cohort study data, and risk ratios are…

This study presents a case for applying a transformation (Box and Cox, 1964) of the criterion used in predictive validity studies. The goals of the transformation were to better meet the assumptions of the linear regression model and to reduce the residual variance of fitted (i.e., predicted) values. Using data for the 2008 cohort of first-time,…