Epidemiology and Health Studies in Schools

Epidemiology and Health Studies in Schools

Epidemiology involves the study and analysis of the distribution, patterns and determinant of health and illness.

A central focus of the mission of Empirical Education is to evaluate the conditions that that support student well-being, recognizing that healthy bodies and minds are essential precursors to success in educational outcomes – from achievement, to educational aspiration and persistence in higher education, to career attainment, and ultimately quality of life.

Empirical’s work in Epidemiological and health-related research, led by Andrew Jaciw, follows two major strands: (1) Evaluation of school conditions and program effects on health outcomes, and (2) Development of methodology pertaining to clinical trials and other epidemiological study designs for evaluating causal impacts of school programs and school contexts on student and teacher outcomes.

Select works by Empirical addressing Epidemiological questions are organized by strands below:

Strand 1: Evaluation of school conditions and program effects on health outcomes

Research Brief: School Instructional Modalities in Oregon in Response to COVID-19 and Associations with Incidence of Illness Academic Outcomes

This work examines how instruction was delivered (hybrid, in-person, remote) over 34 weeks during the 2020/21 school year in Oregon. It disaggregates mode of delivery across schools depending on whether they report data through the National Center for Education Statistics (NCES). Continuing work will (1) disaggregate instructional modalities over time among schools with NCES data (i.e., by urbanicity, and school size), and (2) assess the discontinuities in trends and incidence of COVID-19 in schools that switch from hybrid to in-person.

How Are the Children?: A Randomized Experiment of the Effects of an Emotional-Support program on Student Resilience, School Attendance, and Disciplinary Referrals.

This study examined whether high school students who are provided with an advisory program with supports to strengthen skills including self-management, self-awareness and responsible decision-making demonstrated growth in these areas. It also examined whether the program impacted school attendance, number of disciplinary referrals, and school achievement.

Final Report of the i3 Evaluation of the Collaboration and Reflection to Enhance Atlanta Teacher Effectiveness (CREATE) Teacher Residency Program: A Quasi-Experiment in Georgia.

This study used Survival Analysis methods to examine the impact of a teacher residency program on retention in teaching. It also examined the effects of the program on teacher-reported levels of self-efficacy in teaching, commitment to teaching, stress management, resilience, and mindfulness. Jaciw, A. P., Wingard, A., Zacamy, J., Lin, L., & Lau, S. (2021). Final Report of the i3 Evaluation of the Collaboration and Reflection to Enhance Atlanta Teacher Effectiveness (CREATE) Teacher Residency Program: A Quasi-Experiment in Georgia. (Empirical Education Rep. No. Empirical_GSU-7031-FR1-2021-O.1). Empirical Education Inc.

Strand 2: Development of methodology in clinical trials and observational studies for evaluating causal relationships between exposures and outcomes.

Hold the Bets! Should Quasi-Experiments Be Preferred to True Experiments When Causal Generalization Is the Goal?

This study compared the situations when observational studies and quasi-experiments should be preferred to clinical trials when the goal is to generalize causal effects across study sites.

Jaciw, A. P. (2024). Hold the Bets! Should Quasi-Experiments Be Preferred to True Experiments When Causal Generalization Is the Goal? American Journal of Evaluation, 46(1), 90-127.

Introduction to Special Issue: Multi-armed Randomized Control Trials in Evaluation and Policy Analysis.

This work summarized the development of multi-armed clinical trials by several pioneering researchers, including experiments of the RAND Health Insurance Experiment (Newhouse) and the Health Profession Opportunity Grant (Peck)

Jaciw, A. P. (2023). Introduction to Special Issue: Multi-armed Randomized Control Trials in Evaluation and Policy Analysis. Evaluation Review, 47(1), 3-10.

Do social programs help some beneficiaries more than others? Evaluating the potential for comparison group designs to yield low-bias estimates of differential impact.

This work demonstrates how quasi-experimental observational studies can yield results that are low-bias and provide a suitable alternative to randomized trials in cases where the question of interest is whether an intervention is differentially effective across groups.

Jaciw, A. P. (2023). Do social programs help some beneficiaries more than others? Evaluating the potential for comparison group designs to yield low-bias estimates of differential impact. American Journal of Evaluation.

A Within-Study Approach to Evaluating the Role of Moderators of Impact in Limiting Generalizations from "Large to Small".

This work used a multisite experiment to empirically evaluate how well a grand-mean program effect generalizes to individual sites on average. The results demonstrated that generalization is hard to achieve especially when sites are multilevel, which introduces random sampling error across units at the intermediate level.

Jaciw, A. P., Unlu, F., & Nguyen, T. (2022). A Within-Study Approach to Evaluating the Role of Moderators of Impact in Limiting Generalizations from "Large to Small." American Journal of Evaluation, 43(1), 108–131."

Participant Sorting in Cluster Randomized Controlled Trials: Is it Happening and Why Does it Matter?

This work examined the extent to which natural mobility that happens over time in site-randomized trials compromises the accuracy of impact estimates. This occurs when studies last multiple years and select members within each site choose to participate in the trial in first place. The threat to validity is that during the trial “joiners” may enter into study confounding selection effects with program impacts.

Jaciw, A. P., Nguyen, T., Lin, L., & Zacamy, J. (2020). Participant Sorting in Cluster Randomized Controlled Trials: Is it Happening and Why Does it Matter? Poster presentation accepted for the annual spring conference of the Society for Research on Educational Effectiveness, Washington, DC

Using Multi-Armed Experiments to Test “Improvement Versions” of Programs: When Beneficence Matters

This work proposes a type of SMART ("sequential, multiple assignment, randomized trial") experimental design in which program improvement is studied simultaneously with program impact during the first phase of an experiment, and treatment and control members in the first phase are then re-randomized to continue with the preliminary condition or the improved version during he second phase of the trial.

Jaciw, A., & Nguyen, T. (2018, November). Using Multi-Armed Experiments to Test “Improvement Versions” of Programs: When Beneficence Matters. In E. Hedberg (Chair), Switching, Phasing-in and Improving: Multi-Armed Randomized Experiments with Multiple or Time-Contingent Components (Methods and Tools of Analysis). Panel presented at the annual fall research conference of the Association for Public Policy Analysis and Management (APPAM), Washington, DC.