NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Roommates and Peer Effects in the United States
Initial registration date
March 24, 2017
March 24, 2017 3:38 PM EDT
Other Primary Investigator(s)
Additional Trial Information
The paper analyzes how peer effects relate to grade point average and decisions to join social groups such as fraternities. The results of the experiment indicate that residential peer effects are absent from major life decisions such as choice of college major. However, peer effects in GPA occur at the individual room level. Peer effects regarding fraternity membership occur both at the room level, whereas peer effects in fraternity membership occur both at the room level and the entire dorm level.
The intervention consists of measuring peer effects on observational data and regressing own outcomes and behavior on peer outcomes and behavior. The researcher indicates several main difficulties with this type of experiment as individuals generally self-select into neighborhoods, groups or roommate pairs, blurring the line between selection effect and peer effect.
The data comes from Dartmouth's database of students and includes a full history of housing/dorm assignments and term-by-term academic performance. Pretreatment characteristics include SAT scores, high school class rank, public versus private high school, home state and an academic index created by the admissions office. The data used are for the graduating classes of 1997 and 1998.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
GPA, major choice, membership in a fraternity
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
The experiment is based on Dartmouth College freshmen who are randomly assigned to dorms and roommates, implying that a roommate's background variables are uncorrelated with own background characteristics.
As a natural experiment, the results indicate that the coefficient on roommate GPA is 0.12 and significant. However, this freshman roommate effect on GPA disappears by senior year. Own senior year GPA is not correlated with freshman year roommates' senior year GPA. Additionally, the roommate's pre-enrollment intention to graduate with honors has a positive and statistically significant effect on GPA. Regarding the major choice, the data show that randomly assigned roommates have no effect on major. And regarding membership in a fraternity, if a freshman roommate joins a fraternity then he is 8% more likely to do so and 27% of the roommate pairs join the same house. Thus, roommate peer effects are important influences in freshman year GPA and in decisions to join social organizations. Roommate effects are not important in determining choice of major.
Experimental Design Details
Randomization done by housing system
Was the treatment clustered?
Sample size: planned number of clusters
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
1589 students in total
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Approval Date
IRB Approval Number
Details not available
Post Trial Information
Is the intervention completed?
Intervention Completion Date
December 31, 1998, 12:00 AM +00:00
Is data collection complete?
Data Collection Completion Date
December 31, 1998, 12:00 AM +00:00
Final Sample Size: Number of Clusters (Unit of Randomization)
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
Final Sample Size (or Number of Clusters) by Treatment Arms
1589 individuals total
Reports, Papers & Other Materials
This paper uses a unique data set to measure peer effects among college roommates. Freshman year roommates and dormmates are randomly assigned at Dartmouth College. I find that peers have an impact on grade point average and on decisions to join social groups such as fraternities. Residential peer effects are markedly absent in other major life decisions such as choice of college major. Peer effects in GPA occur at the individual room level, whereas peer effects in fraternity membership occur both at the room level and the entire dorm level. Overall, the data provide strong evidence for the existence of peer effects in student outcomes.
Sacerdote, Bruce. 2001. "Peer Effects with Random Assignment: Results for Dartmouth Roommates." Quarterly Journal of Economics 116(2): 681-704.
REPORTS & OTHER MATERIALS