Poster #145 - Estimating Compensating Wage Differentials for Special Education Teachers over Thirty Years

• In Event: Friday Poster Reception

Abstract

The demand for special education teachers has been steadily increasing, yet there is a significant shortage, particularly in schools with a high percentage of low-income and minority students (Billingsley & Bettini, 2019; Boyd et al., 2003; Hanushek, Kain, & Rivkin, 2004; Mason-Williams et al., 2019; Peyton et al., 2021). According to compensating wage differential models, qualitative job characteristics such as safety, working conditions, and the proportion of nonwhite students can influence teachers' decisions to work at certain schools (Boyd et al., 2003; Chambers, 1981; Goldhaber et al., 2010; Levinson, 1988; Martin, 2010; Rosen, 1974). Poor working conditions are a major driver of teacher turnover and can significantly hinder the effectiveness of special education instruction (Albrecht et al., 2009; Bettini et al., 2016; Johnson, Kraft, & Papay, 2012; Mason-Williams et al., 2019; Simon & Johnson, 2015). 

Our study focuses on the following three research questions: (1) How does the current salary structure compensate special education teachers working in schools with higher proportions of economically disadvantaged, minority, special education, or limited English proficient students (LEP)? How does this compensating wage differential evolve over time? (2) To what extent do student demographics such as higher proportions of low-income, minority, special education, or LEP, reflect the broader challenges in working conditions, including student tardiness, absenteeism, and lack of parental involvement? (3) How do compensating wage differentials for all teachers, including special education teachers, vary based on working conditions as identified through the principal component analysis from Research Question 2? 

We use a pooled cross-sectional dataset from the Schools and Staffing Surveys (SASS) and the National Teacher and Principal Survey (NTPS) that covers the years from 1990 to 2018. The dataset includes teacher compensation, as well as teacher characteristics (age, gender, race, years of experience, union membership, highest degree, and status of certificate or license) and working conditions. The working conditions we examine include teaching load (e.g., the number of students enrolled in a class, the number of hours spent teaching, the number of classes taught) and seriousness of work environment issues (e.g., drug abuse, absenteeism, tardiness). Additionally, we utilize the Common Core of Data (CCD) that includes school-level variables such as the percentages of students eligible for free lunch and non-white students, urbanicity, and school type (elementary, secondary, combined, charter). 

We first employed hedonic regression to estimate the conventional compensating wage differentials based on working conditions as measured by student demographics (% minority, % special ed, % economically disadvantaged, % LEP). We use principal component analysis to capture working conditions, including student tardiness, absenteeism, and class cutting. We find that working conditions as measured by student demographics can only explain about 10% of the variations in these classroom-behavior-based working conditions. Incorporating this broad working conditions measure into our hedonic regression; we find that special education teachers in our sample are not compensated for working in schools with more challenges. Furthermore, our examination of the teacher labor market over time reveals that special education teachers consistently receive no additional compensation across different survey waves.

Author

• Jieon Shim, Texas State University
  Presenting Author