Are there differences by race and gender? Using survival analysis, we find differences by both race and gender, but especially gender. White and BIPOC men have similar exit rates over time, but there are distinct differences b/w White men and women, BIPOC men and women, and White and BIPOC women.

When do we lose paras over the course of their employment? Looking across a 10 year span, we find that most of those who leave (move schools, switch roles, or exit the workforce entirely) do so w/i the first three years. By year 10, about 30% of the original cohort remain employed as paraeducators.

Excited to circulate another @caldercenter.bsky.social WP! Another study from my dissertation, w/ @roddy-theobald.bsky.social‬. Using data from WA, we track individual paraeducators to predict exit from the state education workforce. Ungated WP here: ‬ tinyurl.com/3wtfrm2m
🧵 below:

So how does this impact students? We use fixed effects to compare schools to themselves in years in which they have higher or lower turnover. We find that GET turnover matters for all students, and SET turnover matters most for students with disabilities:

Another reason, which we show in this WP, is that turnover rates among paraeducators and teachers, as well as among general and special ed staff are different. Special ed staff consistently turn over more than their gen ed counterparts, and paras consistently turn over more than teachers:

New @caldercenter.bsky.social WP! A study from my dissertation, w/ @roddy-theobald.bsky.social, @natejones.bsky.social‬, & @lizbettini.bsky.social. We find staff turnover matters for SWDs. Ungated WP & one pager:
• Working paper: tinyurl.com/j3rykkp9
• One-pager: tinyurl.com/bddtzxtp