Browsing by Subject "Value-added models"

Now showing 1 - 2 of 2

Are value-added models for high-stakes teacher accountability arbitrary and capricious?
(2012-05) Melhem, Leila Melanie; Cantu, Norma V., 1954-; Osborne, Cynthia Anne, 1969-
Value-added models are complex statistical formulas that aim to isolate the effect a teacher has on student learning. States and districts across the nation are adopting laws and policies that will evaluate teachers, in part, using the results provided by value-added models. In many states and districts, these evaluations will be used to inform high-stakes decisions about teacher salary and retention. However, value-added models are imperfect tools for assessing teacher effectiveness, and many scholars have argued that they are not appropriate for use in high-stakes decisions. This Article provides a brief history of the use of value-added models in public education and summarizes the major criticisms of using value-added models. In this context, the Article analyzes and evaluates the extent to which substantive due process claims brought by teachers adversely affected by the results of value-added models will be successful. The Article concludes that while the system as a whole is rationally related to the objective of improving the overall effectiveness of the teaching workforce, in certain cases, individual teachers will be able to successfully claim that the results of their value-added model led to a termination that was arbitrary and capricious. Finally, the paper offers some recommendations to states and school districts on how to implement an evaluation system using value-added models to avoid substantive due process violations.
The teachers Texas needs : factors affecting recruitment, retention, and quality
(2023-08-04) Rhodes, Annelies; Marder, Michael P., 1960-; Marshall, Jill Ann; González-Howard, María; Reyes, Pedro
The U.S. education system does not have enough qualified teachers to serve all its students and this is especially problematic in STEM (Science, Technology, Engineering, and Math) subject areas. This dissertation is comprised of three separate quantitative research studies as three separate papers that offer analyses of research questions about policies designed to increase and/or improve the teacher workforce. The goal of this dissertation is to examine potential factors affecting recruitment, retention, and quality of teachers in Texas and discuss possible implications of these trends. The study in Chapter 1 examines how minimum grade point average (GPA) admissions requirements might influence who may enter Educator Preparation Programs (EPPs) and whether these admissions requirements meaningfully predict teaching outcomes. The analyses revealed that Black and Hispanic candidates, male candidates, and candidates in STEM majors are more likely to be excluded from EPPs at higher GPA thresholds. The analyses also demonstrated that GPA was not a good predictor of teaching outcomes, specifically in terms of teaching in a Title 1 school, finding employment as a teacher within one year of program completion, as well as within two years of program completion, and remaining in teaching for at least five years. Specifically, GPA is potentially acting as a barrier to teachers more likely to teach in high need schools. These findings suggest that the minimal predictive use of GPA in determining academic success must be balanced against the potential harm caused by the unintended consequence of excluding teaching candidates of color and STEM majors disproportionately. The study in Chapter 2 examines how elementary and secondary student outcomes in Texas depend upon the policies that regulate production of new teachers and two certification routes that are available to prospective teachers: Standard or Traditional certification and Alternative certification pathways. This study uses value-added models to estimate changes in student test scores in varying grade levels and test subjects based upon teacher preparation pathway. The results showed that there was not one significant estimate in any model where students were not learning more than expected from teachers from Standard and Not For-Profit pathways compared to Alternatively prepared teachers or teachers from For-Profit EPPs. The findings of this study do not suggest that only certain pathways should be used for teacher certification. The premise behind the emergence of alternative pathways holds merit – there should be fewer barriers to the teacher workforce for qualified individuals. These findings suggest that teacher training prior to entering the classroom is important and makes a difference in student achievement (as measured by student test scores). The study in Chapter 3 is written as a white paper that focuses on teacher attrition from 2012 through 2021 in Texas, including any changes to patterns during the COVID-19 pandemic. Specifically, this study examines how the likelihood of teachers staying in teaching has changed over time, utilizing logistic regression models with the outcome variable of leaving the classroom the subsequent year compared to staying. The results of this study show that while there were some shifts in the likelihood of teachers staying after the pandemic started, most patterns were consistent with patterns from the past decade. However, some longitudinal trends could be of potential concern when examining teacher attrition and retention in Texas, including teacher quality. When looking at teacher quality over time (as measured by students’ standardized test scores), Texas has been seeing a decrease in quality scores for teachers that are staying and an increase in scores for teachers that are leaving