Understanding Faculty Perspectives in Indian Higher Education

The study examines how faculty perceives affirmative action policies and related factors in the Indian higher education context.

This study investigates caste-based discrimination by professors in Indian higher education through a natural field experiment. Using a correspondence-style design, Reserved and Unreserved Caste students, recruited as interns, send standardized email requests to professors, asking them to complete an internship-related survey. Professors are randomly assigned to one of eight experimental conditions, varying by student caste, the complexity of the requested task, and the cost of non-response. In the high-complexity condition, the survey requires 8-10 minutes to complete, compared to 2–4 minutes in the low-complexity condition. To increase the cost of discriminatory behavior, the high-cost (Reward) treatment introduces a monetary incentive for survey completion, while the Control group offers no such incentive. Student credentials are held constant across all conditions. This design allows for a rigorous examination of whether caste-based discrimination exists, whether it intensifies with task complexity, and whether raising the cost of discrimination can mitigate bias in academic settings.

2025-05-26

2025-09-01

Professors' beliefs and actions

The experiment involves collecting professors' beliefs about higher education policies on DEI.

All participating professors receive a survey link embedded in the email sent by the student interns. The survey begins by assessing faculty attitudes and perceptions toward students from disadvantaged caste backgrounds. It then collects standardized demographic information and includes a question about how frequently they receive similar outreach from students, which helps evaluate the generalizability of the study's findings. Finally, professors are asked whether they would be open to participating in a follow-up qualitative discussion at a later date. This question serves to gauge their willingness to engage more deeply on the topic in a time-intensive, non-incentivized context.

Professors are randomly assigned to one of eight treatment arms in a 2 × 2 × 2 factorial design, manipulating caste identity, task complexity, and financial incentives. Randomization is conducted at the individual level in Stata, with stratification by gender, academic discipline, and geographic region.

All emails sent across treatment groups include standardized information about the student’s name, academic institution, internship details, and the supervising faculty member’s affiliation. They are sent using newly created email IDs for the two interns (RC and URC), each formatted uniformly to include only the student's first name, followed by six numeric digits and an abbreviation of their affiliated institution. This standardization, along with standardized quality signals included in the email, is intended to reduce informational asymmetries and minimize statistical discrimination, thereby helping to isolate potential taste-based biases in faculty responses. The core message in each email emphasizes that the survey is part of the intern’s academic requirement and aim to understand faculty perspectives in Indian higher education.

To avoid temporal response bias, emails are evenly distributed across all five weekdays and sent at the same time each day. A single follow-up reminder will be sent to non-respondents exactly two weeks after the initial email. Appendix Figure A1 provides the full email templates and the variations across treatment arms.

Randomization is conducted at the individual level in Stata, with stratification by gender, academic discipline, and geographic region.

Individual level

No

6000 (max 9400)

6000 (max 9400)

Divided equally across the treatment arms

2 percentage points (in response rate)