Causal attribution of sex differences and attitudes toward women in STEM

Last registered on December 06, 2023

View Trial History

Pre-Trial

Trial Information

General Information

Title

Causal attribution of sex differences and attitudes toward women in STEM

RCT ID

AEARCTR-0012404

Initial registration date

December 04, 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

December 06, 2023, 7:19 AM EST

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

Locations

Country

Italy

Region

Primary Investigator

Name

Elena De Gioannis

Affiliation

Università degli Studi di Milano

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

In development

Start date

2023-11-24

End date

2023-12-11

Keywords

Gender

Additional Keywords

JEL code(s)

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Despite the significant increase in women working in traditionally masculine sectors, the presence of women in STEM (Science, Technology, Engineering and Math) fields, and thus in high-pay and high-status professions, is still scarce.
Previous experimental studies found that exposing individuals to theories of sex differences, attributing the cause of these differences to either biological or social factors, affects stereotype endorsement, self-stereotyping, interest in STEM and performance in a math-related task. However, the existing studies on the effect of essentialist beliefs on attitudes toward role models are too scarce to provide a clear picture of how the endorsement of fixed versus malleable theories of traits could mediate the efficacy of interventions using role models.
The study aims to test the effect of the endorsement of gender essentialist beliefs on attitudes toward and perceptions of women interested in STEM. It is hypotesized that people endorsing essentialist beliefs on gender would have worse attitudes toward women in STEM and would be more likely to see female counterstereotypical role models as exceptions (subtyping effect) compared to both people thinking that gender differences either do not exist or are due to social factors.
The hypotheses are tested using a randomized control trial with parallel design where essentialist beliefs are manipulated asking participants to read an article on gender (sex) differences in math abilities. Participants are recruited by a panel provider based in Italy and are paid for their participation. The experiment is conducted online. In study 1, 200 participants aged 16-20 are randomly assigned to an article either suggesting that gender differences do not exist, that exist but are due to social factors or that are due to biological differences. They are then asked to answer a questionnaire measuring attitudes toward women in science and gender stereotypes about science. In study 2, 150 participants aged 16-20 are randomly assigned to the same articles (biological differences or no differences), they are then exposed to counterstereotypical role models and asked their opinion on their gender typicality. Their gender bias is also measured.

External Link(s)

Registration Citation

Citation

De Gioannis, Elena. 2023. "Causal attribution of sex differences and attitudes toward women in STEM." AEA RCT Registry. December 06. https://doi.org/10.1257/rct.12404-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Participants are asked to read an article on gender (sex) differences in math abilities.

Intervention Start Date

2023-11-24

Intervention End Date

2023-12-11

Primary Outcomes

Primary Outcomes (end points)

Study 1:
- Attitudes toward women in science
- Endorsement of explicit gender-science stereotypes

Study 2
- perception of gender-typicality of stereotypical and counterstereotypical role models
- gender bias

Primary Outcomes (explanation)

Study 1:
- attitudes toward women in science: "Women possess characteristics which enable them to be successful in science careers" dimension from the "Women in science scale" (WISS, Erb & Smith, 1984)
- endorsement of gender-science stereotypes: instrument proposed by Schmader et al. (2004)

Study 2:
- perception of gender typicality: adaptation of the scales on self-perceived gender typicality (Krueger et al., 2003; Leaper & Van, 2008; Patterson, 2012)
- gender bias: adaptation of the instrument used to measure gender bias by De Gioannis et al. (2023)

Secondary Outcomes

Secondary Outcomes (end points)

Study 1:
- attitudes toward science

Secondary Outcomes (explanation)

Attitudes Toward Science Inventory in the modified version proposed by Tai et al. (2022)

Experimental Design

Participants are first asked to read an article from a newspaper describing gender (sex) differences in math abillities. This is aimed to manipulate their gender essentialist beliefs. They are randomly assigned to read an article that either suggests that these differences exist and are due to biological factors (T1), that they exist but are due to social factors (C1) or that they do not exist (C2). In study 2, participants are either assigned to T1 or to C2.

Participants are first asked a few questions on the articles to reinforce the manipulation. They are then asked to answer a questionnaire including sociodemographic information and the questions measuring the outcomes of interest.

Experimental Design Details

Randomization Method

Block randomization based on sociodemographic chacteristics, e.g. gender, residence.

Randomization Unit

Individual

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

Study 1: 200 individuals aged 16-20
Study 2: 150 individuals aged 16-20

Sample size: planned number of observations

Study 1: 200 individuals aged 16-20 Study 2: 150 individuals aged 16-20

Sample size (or number of clusters) by treatment arms

Study 1: 70 individuals for biological differences, 65 individuals for social differences, 65 individuals for no differences
Study 2: 75 individuals for biological differences, 75 individuals no differences

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Minimum detectable effect size: Study 1: small-to-medium effect size (f2 = 0.03) Study 2: medium effect size (f = 0.25)

Supporting Documents and Materials

IRB