Do Time-Constraints Matter? How, Why, and for Whom?

Last registered on April 26, 2023


Trial Information

General Information

Do Time-Constraints Matter? How, Why, and for Whom?
Initial registration date
January 02, 2023

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
January 03, 2023, 5:32 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
April 26, 2023, 12:30 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.



Primary Investigator

Penn State University

Other Primary Investigator(s)

PI Affiliation
Penn State University
PI Affiliation
Bilkent University

Additional Trial Information

On going
Start date
End date
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
We study test-taking behavior in time-constrained exams and describe time's role in shaping the distribution of students' performance outcomes. We do this by leveraging a policy change in Turkey which increased the college entrance exam time. Supporting administrative school-level data with the experimental data we collected, we show that having more test time is significantly associated with less wrong answers and less skipping patterns in a multiple-choice test where negative marking exists. Moreover, this association varies by gender and ability which we proxy for using background data. Female students are less likely to gain from an extra minute than male students and above median students are less likely to gain from an extra minute than below median students. Besides the performance changes, we also find that more test-time reduces the standard error and thus leads convergence of scores to each other. These data patterns motivates us to construct a model of test-taking behavior in time-constrained exam settings to firstly quantify the gains and losses from the policy change and secondly provide alternative exam policies with counterfactual exercises, interacting test-time with other test settings.
External Link(s)

Registration Citation

Akyol, Pelin , Kala Krishna and Esma Ozer. 2023. "Do Time-Constraints Matter? How, Why, and for Whom?." AEA RCT Registry. April 26.
Experimental Details


Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
We are interested in learning how time-constraints impact (i) probability of getting correct, (ii) probability of skipping in a multiple-choice exam set up.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The sample is composed of 12th grade students in a high school. There is one treatment-arm which is the (more) time limit in a multiple choice test.The sample is randomly grouped into two: time-relaxed and time-constrained.
In the first wave of experiment: students take a multiple choice (with 5 choices) test with four subjects (Turkish, Math, Science, Social Science) where negative marking applies. The booklets given to students are identical, i.e., the question difficulty/order doesn't differ across randomization units. Time-constrained group are given 90 minutes for 100 questions while time-relaxed group are given 170 minutes for 100 questions. The time-limit is designed such that the time-relaxed group is only ability-constrained. Proctoring is provided by randomly assigned school teachers in each of 5 classrooms. Students are incentivized with money per their ranking in their group. They can earn money between $0-30.
In the second wave of the experiment, all the settings stay same but the negative marking. We remove negative marking to see the changes in our primary outcomes with respect to time-constraints.
Experimental Design Details
Randomization Method
Randomization is done using computer with random number generator.
Randomization Unit
Randomization unit is individiual in a given school.
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
One school with planned 80-100 students.
Sample size: planned number of observations
80-100 students.
Sample size (or number of clusters) by treatment arms
40-50 students control, 40-50 students treatment.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Institutional Review Boards (IRBs)

IRB Name
Pennsylvania State University
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials