Causal Analysis in Education

2021/2022

Category 'Best Course for Broadening Horizons and Diversity of Knowledge and Skills'

Category 'Best Course for New Knowledge and Skills'

Type: Optional course (faculty)

Delivered by: Institute of Education

Where: Institute of Education

When: 1, 2 module

Open to: students of all HSE University campuses

Instructors: Andrey Zakharov

Language: English

ECTS credits: 4

Contact hours: 30

Full Syllabus

Abstract

What works is probably one of the crucial questions for educational research today. Policy makers want to know what initiatives can help to use educational resources more effectively. Schools and universities are looking for new ways of teaching that can increase students’ outcomes. Educational market is rapidly changing with coming of new technologies and regular users are looking for the evidence that the online course or other service they buy provides the result they need. In all the cases people rely on the promises or expectations for the effects. However to be reliable or unbiased such a promise should be based on the research done with a specific design known as causal analysis. This course has two major goals. First, it introduces contemporary theory of causal analysis and different research designs aimed at estimating effects. By the end of the course students will be aware of how to do a randomized experiment and most widely used quasi-experiments. They will understand strengths and weaknesses of different research designs and get some experience in application of quasi-experimental methods in practice while working with real life cases. Second, this course introduces contemporary studies, that investigate important educational issues. Among them, for instance, are the effects of different policies and resources distribution on educational outcomes. Students will learn about the effects of education attained on adults’ income and career path. We will also focus on the studies that investigate effects of various educational technologies. Altogether, this course will help students to better understand what works in education and how to estimate effects of different interventions.

Learning Objectives

This course will introduce theory of causal effects analysis as well as experimental and quasi-experimental research designs.
Students will get familiar with contemporary studies that investigate what works in education.
Students will get experience in estimation of causal effects of different educational resources, technologies or school programs.

Expected Learning Outcomes

Students are familiar with the Neyman–Rubin causal model and its assumptions
Students can define a good instrument, test the instrumental variable assumptions, perform 2SLS analysis, explain it's weakness and strength, and give interpretation to the results
Students can do first difference, difference-in-difference, and regression discontinuity analysis with sharp design. They can give correct interpretations to the results
Students can estimate a sample size for an experiment and make randomization correctly
Students can estimate the propensity score, and do causal analysis with different methods - optimal matching, greedy (nearest neighbor, caliper) matching, kernel matching, stratification. Students can test the assumptions and interpret the results correctly
Students can explain the concept of propensity score and are familiar with stratification and matching technics
Students can explain the concept of the instrumental variable and the assumptions of a good instrument
Students can explain the discontinuity design and the cases when it is applicable
Students can explain why randomized experiment is the "golden standard" of causal analysis. They can design an experiment and prepare research program

Course Contents

Introduction to Causal Analysis
Randomized Experiment
Instrumental Variables
First difference. Difference-in-difference. Regression Discontinuity
Matching

Assessment Elements

Questions related to the articles assigned to read
Students write short answers to five questions related to the paper assigned to read.
Data analysis and description
Student presents the code and the text with the short description of the research question, data used, analysis, results, and their interpretation.
Presentations of the articles
Intermediate Control Project
Students present their RCT project
Final Control Project

Interim Assessment

2021/2022 2nd module
0.15 * Intermediate Control Project + 0.22 * Questions related to the articles assigned to read + 0.15 * Presentations of the articles + 0.2 * Final Control Project + 0.28 * Data analysis and description

Bibliography

Recommended Core Bibliography

Methods matter : improving causal inference in educational and social science research, Murnane, R. J., 2011
Peter Steiner. (2010). S. Guo & M.W. Fraser (2010). Propensity Score Analysis: Statistical Methods and Applications. Psychometrika, (4), 775. https://doi.org/10.1007/s11336-010-9170-8
Роберт, И. R в действии. Анализ и визуализация данных в программе R : руководство / И. Роберт, Кабаков ; перевод с английского Полины А. Волковой. — Москва : ДМК Пресс, 2014. — 588 с. — ISBN 978-5-97060-077-1. — Текст : электронный // Лань : электронно-библиотечная система. — URL: https://e.lanbook.com/book/58703 (дата обращения: 00.00.0000). — Режим доступа: для авториз. пользователей.

Recommended Additional Bibliography

Angrist, J. D., & Lavy, V. (1999). Using Maimonides’ Rule to Estimate the Effect of Class Size on Scholastic Achievement. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.8AB10B37
Dee, T. S., & National Bureau of Economic Research, C. M. (2003). Are There Civic Returns to Education? NBER Working Paper Series.
Experimental and quasi-experimental designs for generalized causal inference, Shadish, W. R., 2002
Fairlie, R. W., Loyalka, P., Rozelle, S., & Ma, Y. (2020). Isolating the “Tech” from EdTech: Experimental Evidence on Computer Assisted Learning in China. IZA Discussion Papers.
Imbens, G. W., & Rubin, D. B. (2015). Causal Inference for Statistics, Social, and Biomedical Sciences. Cambridge University Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsrep&AN=edsrep.b.cup.cbooks.9780521885881
Mina Johnson-Glenerg, Mina C. Johnson-Glenberg, Colleen Megowan-Romanowicz, David Birchfield, & Caroline Savio-Ramos. (2016). Effects of embodied learning and digital platform on the retention of physics content: Centripetal force. Frontiers in Psychology, 7. https://doi.org/10.3389/fpsyg.2016.01819
Mistree, D., Loyalka, P., Fairlie, R., Bhuradia, A., Angrish, M., Lin, J., Karoshi, A., Yen, S. J., Mistri, J., & Bayat, V. (2021). Instructional interventions for improving COVID-19 knowledge, attitudes, behaviors: Evidence from a large-scale RCT in India.
Mostly harmless econometrics : an empiricist's companion, Angrist, J. D., 2009
Rachel Baker, Eric Bettinger, Brian Jacob, & Ioana Marinescu. (2017). The Effect of Labor Market Information on Community College Students’ Major Choice. NBER Working Papers.
Rajeev H Dehejia, & Sadek Wahba. (1999). Causal Effects in Nonexperimental Studies: Reevaluating the Evaluation of Training Programs. Http://Www.Uh.Edu/%7Eadkugler/Dehejia%26Wahba_JASA.Pdf.

Course Syllabus