Beliefs, Ability and College Major Success

How to cite

Rury, Derek; Carrell, Scott. 2019. "Beliefs, Ability and College Major Success." AEA RCT Registry. AEARCTR-0003432. https://doi.org/10.1257/rct.3432-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Derek Rury* , Scott Carrell

* Primary investigator

Registry ID
AEARCTR-0003432
DOI
https://doi.org/10.1257/rct.3432-1.1
First registered
Last updated
Status
Completed
Trial dates
2018-10-02 – 2019-10-16
Countries
United States of America (California)
Registry record
http://www.socialscienceregistry.org/trials/3432

Abstract

Research in the economics of education has demonstrated the students often lack important information about the characteristics of different majors when in college. This field experiment aims to assess whether students may benefit from information about the distribution of academic ability in their preferred field of study by providing them with information about the average test score or high school GPA of recent graduates in their stated preferred major. Several outcomes of interest will be studied, including course taking behavior, grades in those courses, college major choice and graduation. Particular attention will be paid to differential effects based on gender as well as how accurate students' beliefs are about the distribution of beliefs.

Intervention

I present average standardized test of high school GPA information, calculated by major, to college students in introductory STEM or STEM-related fields to measure the impact of this information on important college outcomes, such as courses-taken, grades, college major(s) and minor and finally graduation. After asking students several baseline questions about their beliefs, students reveal their top two choices of major along with which academic measure (eg. SAT math) is most representative of ability in that field. I then ask students what they think is the average score, of the selected measure, of recent graduates with that major. Students are then randomly assigned the actual average score of that measure for that major.

Experimental design

I present average standardized test of high school GPA information, calculated by major, to college students in introductory STEM or STEM-related fields to measure the impact of this information on important college outcomes, such as courses-taken, grades, college major(s) and minor and finally graduation.

Design details

Randomization method
The intervention is conducted through a survey software called "Qualtrics". The software has a way to randomize who gets treatment and who gets control. I ensured that treatment was stratified across all first and second choices of major, so that each first choice and each second choice has an equal number of treatment and control units.
Randomization unit
Student level
Clusters
1,000 students
Observations
1,000 students
Arms
about 350 students were in treatment and 650 students were in control (this is unbalanced because only students who whose own score was above the average selected score for that major were eligible for treatment)
Primary outcomes
Courses-taken (dropped), grades, major(s) and minor selected and graduation
Secondary outcomes
I will look at how the baseline answers, as well as the treatment effect varies by several important demographic characteristics, such as gender, socio-economic status, first generation college student, high school success. I will also test for the existence of various biases on college major decisions, specifically "confirmation bias" in which students place more weight on information that confirms their beliefs and less weight on information that contradicts their beliefs

Cited by

No citation data available for this registration yet.

Keywords