Taking advantage of R and ggplot2 to visualize panel data.
Read more →
Junior scholars in particular need to develop the skills for publishing in academic journals AND for engaging with the community of scholars online
Ignoring average inter-item correlation is a really bad thing!
Given the credibility gap in social science research, we had all better get into the data science canoe, or we will ride our statistical row boats right over the waterfall.
The use and misuse of the margins command in Stata to visualize interaction effects.
We need to move past the idea that we can identify a ‘correct’ model, embrace uncertainty, and be humble with our measures.
I recently had a paper rejected. That’s a little irritating, but what it really irritating is that the reviewers were right.
Building a site with blogdown, RStudio, and Updog has made my online existence SO much more pleasant!
We can think of measurement error as an endogeneity problem—that means we must address it in latent variable models to yield consistent effect size estimates.
Headline grabbing effect sizes with sexy variables in complicated models are, over the long run, far more likely to be found wanting than vindicated as possibly right. Science progresses with small, incremental contributions to our knowledge base.