Gender Stereotypes and Education Gaps in the Economics Field (GSEGEF)

Definition: Summarizes information about the trial. To clarify the study’s objective, please include information on: the main outcome(s) , the intervention(s), the level of randomization , eligibility criterion, population of interest, the sample size , and treatment assignment mechanism .
Limit: 10.000 characters

The project aims to detect and contrast the influence of gender stereotypes in high-quantitative subjects with special attention to economics
These stereotypes, by undermining female confidence, negatively affect girls' enrolment in university and their career opportunities in this field.
In addition to causing unfair gender imbalances, stereotypes are the source of a significant loss of talent. The project is articulated into three components:
1) Experimentation and field research activities conducted with students in their last year of high school.
2) Laboratory activities with the same subjects as in point 1).
3) Extensive statistical and econometric research.
Students will be involved in a series of activities aimed at tackling gender stereotypes as barriers to equal opportunities in education and employment.
Activities include exposure to:
● Board game to raise awareness on gender stereotypes in the choice of professions [estimated time: 90 minutes].
● Short films on gender stereotypes screening and class discussion. [estimated time: 90 minutes]
● Interactive meeting with female professional role models and discussion on elements relevant for school-to-work transition [estimated time: 60 minutes]
Before and after treatment students are screened with questionnaires that will provide useful data to measure the impact of gender stereotypes in choice processes. Within the questionnaires, Implicit Association Tests (IAT) will be used in order to highlight implicit gender attitudes and gender stereotypes acting at an unconscious level.
Among the determinants of study choices, the potential impact of educational preparation in mathematics, employment and career expectations, and socio-economic characteristics of families will be assessed. The survey will also consider provincial data on employment rates, gender gaps in the labour market, size of enterprises, as well as institutional data on curricula offered by faculties, the percentage of women in the teaching staff, and rankings regarding the quality of teaching.
The eligibility criteria were applied to the classes and were based on the following guidelines: the participant must receive the maximum possible benefit from the treatment and the measurement of the potential outcome must be consistent and reasonable and meaningful. For these reasons we decided to include in our experiment only:
- fifth classes: excluding fourth or previous classes allows us both to offer the treatment to more aware pupils and at the same time capturing a more precise outcome regarding their future choice of university (as they are closer in time to making this choice).
- mixed or female-dominated classes: male-dominated classes were excluded as they would have benefited less from our intervention and would have been of less interest in the outcome estimation.
In addition, we excluded from the project those classes that required a reduction in the total number of project hours, as this would also have reduced both outcome and potential benefit.
The treatment assignment mechanism was based on ethical constraints imposed by institutes; the class treated were chosen by the high schools and we were blind on the selection methods, but we had put as many constraints as possible in order to reduce the differences between the groups: every class treated must have a control class that belongs to the same school, has the same course of study and have a similar students’ composition.
Moreover, at the end of intervention and data collection all classes (treated and not) are exposed to the entire educational offer of “Unimore” universities. The latter part is not considered as an element of the intervention, but rather as a way to raise the awareness of the pupils of the university offer in the area. So that they have all the possible elements to be enabled to evaluate and directly apply the benefits of the project in their choices.
The main outcomes we are aiming to measure are the increase in female enrolment intention in universities with a high quantitative content (mathematics, statistics, etc.), the reduction in gender stereotypes on quantitative domain, the increase in female students’ sense of belonging in quantitative domain.
Consistent with the objectives of Horizon Europe and the Smart Specialization Strategy of the Emilia Romagna Region, this study is highly innovative and is based on the cooperation of experts in economics, gender inequality and neuroscience.

External Link(s)

Registration Citation

Citation

Addabbo, Tindara et al. 2021. "Gender Stereotypes and Education Gaps in the Economics Field (GSEGEF)." AEA RCT Registry. November 28. https://doi.org/10.1257/rct.8014-1.0

Sponsors & Partners

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

Request Information

Experimental Details

Interventions

Intervention(s)

Before treatment, all students (treated and control) are screened with a questionnaire in wich the Implicit Association Test (IAT) it is also included.
Subsequently, In autumn 2021, students from the selected classes participate in the following activities:
- Board game to raise awareness on gender stereotypes in the choice of professions [estimated time: 90 minutes].
- Short films on gender stereotypes screening and class discussion. [estimated time: 90 minutes]
- Interactive meeting with female professional role models and discussion on elements relevant for school-to-work transition [estimated time: 60 minutes]
Finally, the same initial questions are asked again to all the pupils, moreover their actual enrolment at university is observed.

Intervention Start Date

2021-11-25

Intervention End Date

2022-03-31

Primary Outcomes

Primary Outcomes (end points)

- increase in female enrolment intention in universities with a high quantitative content (mathematics, statistics, etc.)
- reduction in gender stereotypes on quantitative domain
- increase in female students’ sense of belonging in quantitative domain

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

1) lower gender stereotypical beliefs [implicit and explicit measures]
2) greater female orientation towards economic subjects

Secondary Outcomes (explanation)

1) IMPLICIT MEASURES: A standardized difference score in IAT is calculated for each participant, indicating in which condition (compatible vs. incompatible) participants were faster. A difference score of 0 indicates no difference in speeds; a positive score indicates that one was faster in the compatible block; and a negative score indicates that one was faster in the incompatible block.
NOTES:
- Compatible condition is verifying by associating male to economic in the IAT exercise
- Incompatible condition is verifying by associating female to economic in the IAT exercise
If the difference in the pupil's score between the two is significant he has a stereotypical conception of the economist.

Experimental Design

The project involves 22 fifth grade classes of high schools.
The classes belong to high schools that showed interest in the project after dropping all candidates who:
- had a high percentage of males in their classes and/or
- requested a remodelling or reduction of the intervention we had designed and/or
- were not willing to accept our constraints on the assignment of the factual and counterfactual groups.

Randomisation took place within subgroups created considering the institute and the various curricula so that factual and counterfactual were as similar as possible.
In order to avoid interference in the outcome, none of the participants is aware that it is involved in a project aimed at deconstructing gender stereotypes. They believe that they are participating in a generic orientation project.
Before treatment, all students in the 22 classes are screened with a questionnaire. The questionnaire includes questions on demographics, economic situation, skills, economic propensity, plans for a future academic and university career etc. It also includes the Implicit Association Test (IAT) to highlight implicit gender attitudes and stereotypes towards economics that operate at an unconscious level.
Subsequently, starting from november 2021, students from the 11 selected classes participate in the following activities:
- Board game to raise awareness on gender stereotypes in the choice of professions [estimated time: 90 minutes].
- Short films on gender stereotypes screening and class discussion. [estimated time: 90 minutes]
- Interactive meeting with female professional role models and discussion on elements relevant for school-to-work transition [estimated time: 60 minutes]
Finally, the same initial questions are asked again to all the pupils of the 22 classes, moreover their intention of enrolment at university is observed in order to see if there are any differences between the factual and the counterfactual groups and to detect if the intervention had any eventual impact.

Experimental Design Details

Randomization Method

due to ethical constraints, the classes covered were chosen by the high schools and and researchers are blind to this step

Randomization Unit

Classes

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

22 classes: 11 treated and 11 controls

Sample size: planned number of observations

The number of observations was set guided by trying to maximise it within the constraints of the available resources (that allow us to treat 11 classes for an estimated total of about 220 pupils) and - counterfactual included - consist in nearly 22 classes and 440 pupils. Of course, a statistical power analysis was performed to verify that the minimum required number of students was reached. We did a sample size estimation, by utilizing GPower software. We set the expected effect size to 0,1 by following Cohen's (1988) criteria for the ANOVA: Repeated measures’ F-test that considered it as a small effect size. With an alpha = 0,05 and power = 0,80, the projected sample size needed with this effect size (GPower 3.1) is approximately N = 200 students for this between-within group comparison. Thus, we collected the largest sample possible given the resource (nearly 440 ) and we considered it as more than adequate for get an accurate estimate.

Sample size (or number of clusters) by treatment arms

11 control and 11 treated

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

At individual level only, a statistical power analysis was performed for sample size estimation, by utilizing GPower software. We set the expected effect size to 0,1 by following Cohen's (1988) criteria for the ANOVA: Repeated measures’ F-test that considered it as a small effect size. With an alpha = 0,05 and power = 0,80, the projected sample size needed with this effect size (GPower 3.1) is approximately N = 200 students for this between-within group comparison. But, as the students are clustered in classes we also took into account the rule of thumb to calculate the minimum number of 22 clusters.

Supporting Documents and Materials

IRB