Gender Differences in Task Perceptions: Experimental Evidence

2024-01-25

2025-12-31

Our primary outcomes include:
• Task performance (both absolute and relative)
• Beliefs about performance (both absolute and relative)
• Task perceptions
• Social norms (or norm perceptions)
• Misperceptions (task misperceptions and norm misperceptions)

The baseline experiment consists of two parts:
Part 1: Task Performance
In Part 1, participants will undertake a series of real-effort tasks tailored to evaluate their task performance, beliefs, perceptions, confidence, and misperceptions regarding gender stereotypes. These tasks include Emotion, Encryption, Math, Rotation, and Ravens. Participants will engage in tasks assessing emotional intelligence, logical reasoning, mathematical problem-solving, spatial reasoning, and abstract reasoning skills.

Part 2: Belief Elicitation
In Part 2, all participants will respond to four Belief Questions related to each task, assessing their perceptions of performance (both absolute and relative) and gender differences (first- and second-order gender-task perceptions).

Survey
After completing Part 2, participants will respond to a survey include demographic questions (e.g., age, nationality, study level, etc.) and questions about their experience in the experiment.

The baseline experiment consists of two parts:
Part 1: Task Performance
In Part 1, participants will engage in a series of real-effort tasks designed to assess their task performance, beliefs, perceptions, confidence, and misperceptions related to gender stereotypes. The tasks include Emotion, Encryption, Math, Rotation, and Ravens.
• Emotion: Participants will complete tasks designed to evaluate their emotional intelligence and perception.
• Encryption: Participants will engage in tasks requiring logical reasoning and problem-solving skills related to code decryption.
• Math: Participants will complete mathematical problem-solving tasks.
• Rotation: Participants will perform spatial rotation tasks, assessing their spatial reasoning skills.
• Ravens: Participants will engage in abstract reasoning tasks, assessing their cognitive abilities.

Part 2: Belief Elicitation
In Part 2, all participants answer the following Belief Questions regarding each task:
Question 1: You have answered X questions in this task. How many questions do you think you answered correctly?
Question 2: There are 20 participants in your session who completed this task, including you.
If everyone is ranked from 1 to 20, 1 being the best performer, which rank do you think you will be, based on the number of questions you answered correctly?
Question 3: On average, do you think men or women answered more questions correctly in this task?
Question 4: We previously asked whether you think men or women answered more questions correctly in this task on average. We now ask you how you think other participants in today’s experiment responded to this question.
What do you think was the most common response among other participants?

After completing Part 2, subjects also asked unincentivized questions on how difficult they perceived each of the tasks to be. They then complete a survey including demographic questions (e.g., age, nationality, study level, etc.) and questions about their experience in the experiment.

Participants are paid for either Part 1 or Part 2 of the experiment, which is randomly determined for each participant at the end of the experiment. If paid for Part 1, participants will receive a piece rate for correct answers, with wrong answers not affecting their payoff. If paid for Part 2, participants will receive a flat rate for selected belief questions. Payment rates are balanced to ensure roughly equal payments for both stages, with participants also receiving a show-up fee.

Randomization is vital to mitigate order effects during task completion and belief expression.

In Part 1, participants are assigned random task sequences to complete the five tasks. Each participant is randomly assigned one of the following sequences:
1) Emotions, Encryption, Math, Rotation, Ravens;
2) Rotation, Math, Encryption, Ravens, Emotions;
3) Math, Emotions, Ravens, Encryption, Rotation;
4) Encryption, Ravens, Rotation, Emotions, Math;
5) Ravens, Rotation, Emotions, Math, Encryption.
These sequences are allocated randomly by a computer to ensure variability across participants and prevent biases from a fixed task order.

In Part 2, the options for perception and norm questions include:
1) Men perform better,
2) Women perform better, and
3) Men and women perform equally.
Each participant is randomly assigned an option order by the computer to minimize order effects. However, within each participant, the assigned order remains consistent for all belief questions related to task perceptions and social norms.

The randomization unit for this experiment is the individual participant. There is only one level of randomization, with participants assigned to task sequences individually.

No

120-130 participants

We are aiming for 120-130 participants, 60-65 for each gender

120-130 participants in total for 5 task sequences

Power calculations are included in the Pre-analysis Plan.