Search for EdWorkingPapers here by author, title, or keywords.
Methodology, measurement and data
Employers may favor applicants who played college sports if athletics participation contributes to leadership, conscientiousness, discipline, and other traits that are desirable for labor-market productivity. We conduct a resume audit to estimate the causal effect of listing collegiate athletics on employer callbacks and test for subgroup effects by ethnicity, gender, and sport type. We applied to more than 450 jobs on a large, well-known job board. For each job listing we submitted two fictitious resumes, one of which was randomly assigned to include collegiate varsity athletics. Overall, listing a college sport does not produce a statistically significant change in the likelihood of receiving a callback or interview request. However, among non-white applicants, athletes are 3.2 percentage points less likely to receive an interview request (p = .04) relative to non-athletes. We find no statistically significant differences among males or females.
We consider the case in which the number of seats in a program is limited, such as a job training program or a supplemental tutoring program, and explore the implications that peer effects have for which individuals should be assigned to the limited seats. In the frequently-studied case in which all applicants are assigned to a group, the average outcome is not changed by shuffling the group assignments if the peer effect is linear in the average composition of peers. However, when there are fewer seats than applicants, the presence of linear-in-means peer effects can dramatically influence the optimal choice of who gets to participate. We illustrate how peer effects impact optimal seat assignment, first under a general social planner utility function and then from both an efficiency and an equity perspective. We next use data from a recent job training RCT to provide the first evidence of large peer effects in the context of job training for disadvantaged adults. Finally, we combine the two results to show that the program's effectiveness varies greatly depending on whether the assignment choices account for or ignore peer effects.
Recent interest to promote and support replication efforts assume that there is well-established methodological guidance for designing and implementing these studies. However, no such consensus exists in the methodology literature. This article addresses these challenges by describing design-based approaches for planning systematic replication studies. Our general approach is derived from the Causal Replication Framework (CRF), which formalizes the assumptions under which replication success can be expected. The assumptions may be understood broadly as replication design requirements and individual study design requirements. Replication failure occurs when one or more CRF assumptions are violated. In design-based approaches to replication, CRF assumptions are systematically tested to evaluate the replicability of effects, as well as to identify sources of effect variation when replication failure is observed. In direct replication designs, replication failure is evidence of bias or incorrect reporting in individual study estimates, while in conceptual replication designs, replication failure occurs because of effect variation due to differences in treatments, outcomes, settings, and participant characteristics. The paper demonstrates how multiple research designs may be combined in systematic replication studies, as well as how diagnostic measures may be used to assess the extent to which CRF assumptions are met in field settings.
Valid and reliable measurements of teaching quality facilitate school-level decision-making and policies pertaining to teachers. Using nearly 1,000 word-to-word transcriptions of 4th- and 5th-grade English language arts classes, we apply novel text-as-data methods to develop automated measures of teaching to complement classroom observations traditionally done by human raters. This approach is free of rater bias and enables the detection of three instructional factors that are well aligned with commonly used observation protocols: classroom management, interactive instruction, and teacher-centered instruction. The teacher-centered instruction factor is a consistent negative predictor of value-added scores, even after controlling for teachers’ average classroom observation scores. The interactive instruction factor predicts positive value-added scores. Our results suggest that the text-as-data approach has the potential to enhance existing classroom observation systems through collecting far more data on teaching with a lower cost, higher speed, and the detection of multifaceted classroom practices.
There is no national consensus on how school districts calculate high school achievement disparities between students who experience homelessness and those who do not. Using administrative student-level data from a mid-sized public school district in the Southern United States, we show that commonly used ways of defining which students are considered homeless can yield markedly different estimates of the homelessness-housed student high school graduation gap. The key distinctions among homelessness definitions relate to how to classify homeless students who become housed and how to consider students who transfer out of the district or drop out of school. Eliminating housing insecurity-related achievement disparities necessitates understanding the link between homelessness and educational achievement; how districts quantify homelessness affects measured gaps.
Four agencies have estimated progress in U. S. student achievement by administering waves of psychometrically linked tests in math and reading to nationally representative samples of students for selected periods over the past half century. Observed agency effects are attributed to differences in purpose, test design and sampling frame. Nonetheless, evidence across surveys from eleven million observations shows achievement gains consistent with the Flynn hypothesis that intelligence is rising. Gains are less steep in reading than math, though math increments may be diminishing. Greater progress is observed for students who are younger, non-white, and from low socio-economic backgrounds. Results are consistent with causal explanations that emphasize early-in-life improvements in nutrition, health care, and protection from contagious diseases and environmental risks.
We show that fade out biases value-added estimates at the teacher-level. To do so, we use administrative data from North Carolina and show that teachers' value-added depend on the quality of the teacher that preceded them. Value-added estimators that control for fade out feature no such teacher-level bias. Under a benchmark policy that releases teachers in the bottom five percent of the value-added distribution, fifteen percent of teachers released using traditional techniques are not released once fade out is accounted for. Our results highlight the importance of incorporating dynamic features of education production into the estimation of teacher quality.
International assessments are important to benchmark the quality of education across countries. However, on low-stakes tests, students’ incentives to invest their maximum effort may be minimal. Research stresses that ignoring students’ effort when interpreting results from low-stakes assessments can lead to biased interpretations of test performance across groups of examinees. We use data from the Programme for International Student Assessment (PISA), a low-stakes test, to analyze the extent to which student effort helps to explain test scores heterogeneity across countries and by gender groups. Our results highlight the importance of accounting for differences in student effort to understand cross-country heterogeneity in performance and variations in gender achievement gaps across nations. We find that, once we account for differential student effort across gender groups, the estimated gender achievement gap in math and science could be up to 12 and 6 times wider, respectively, and up to 49 percent narrower in reading, in favor of boys. In math and science, the gap widens in most countries, even among some of the top 20 most gender-equal countries. Altogether, our effort measures on average explain between 36 and 40 percent of the cross-country variation in test scores.
School district consolidation is one of the most widespread education reforms of the last century, but surprisingly little research has directly investigated its effectiveness. To examine the impact of consolidation on student achievement, this study takes advantage of a policy that requires the consolidation of all Arkansas school districts with enrollment of fewer than 350 students for two consecutive school years. Using a regression discontinuity model, we find that consolidation has either null or small positive impacts on student achievement in math and English Language Arts (ELA). We do not find evidence that consolidation in Arkansas results in positive economies of scale, either by reducing overall cost or allowing for a greater share of resources to be spent in the classroom.
Researchers often include covariates when they analyze the results of randomized controlled trials (RCTs), valuing the increased precision of the estimates over the potential of inducing small-sample bias when doing so. In this paper, we develop a sufficient condition which ensures that the inclusion of covariates does not cause small-sample bias in the effect estimates. Using this result as a building block, we develop a novel approach that uses machine learning techniques to reduce the variance of the average treatment effect estimates while guaranteeing that the effect estimates remain unbiased. The framework also highlights how researchers can use data from outside the study sample to improve the precision of the treatment effect estimate by using the auxiliary data to better model the relationship between the covariates and the outcomes. We conclude with a simulation, which highlights the value of using the proposed approach.