Capstone Research Project

Examining the relationship between a feminist self-identity, degree of belief that gender inequality exists, and political participation.

This self-directed research project highlights the relationship between an individual who adopts a feminist label and their poltical participation. I hypothesized that those who identify as feminist would participate more than non-feminists and secondly, that the relationship between a feminist identity and politcal participaton was mediated by the degree to which an individual believes that gender inequality exists. Using theoretical modelling and hypothesis testing with OLS and logistic regressions, I found that those who identify as feminist are more poltically active, but that this relationship is not in fact mediated by degree of belief.

...
Theoretical Model

The diagram to the left reflects the theoretical model with which I arrived at my two hypotheses. Through my research, I found evidence that belonging to group or collective identity, like a feminist self-identity, works to socialize individuals to participate in politics at greater rates. In a time when identity politics are quite prevalent, I believe that determining what exactly is driving people to participate, beyond just the self-identity or label, is very important. Thus, I theorized that perhaps this socialization works through changes in the degree to which one believes that gender inequality exists in the U.S. today such that those who identify as feminist are socialized to have a higher degree of belief compared to non-femininsts. Therefore, I predicted that the degree to which one believes in the existence of inequality mediates the relationship between a feminist self-identity and political participation.

Data and Methods

For this project, I used data from the 2012 Outlook on Life Surveys which sought to capture general social and poltical attitudes among U.S. adults. My main predictor variables were feminist self-identification, measured with a binary indicator (0-1), as well as degree of belief that gender inequality exists, measured on a scale of 1-5 with 1 being a low degree of belief and 5 being a high degree of belief. I had two response variables to capture political participation. The first one was a measure of explicit participation, such as helping in a voter registration drive, giving money to a campaign, participating in a political party, etc. This was a count variable ranging from 0 (no participation) to 7 (maximum participation). Secondly, I also wanted to capture an individual's political activism through their involvement in activities like attending protests, signing petitions, or attending neighborhood marches, etc. This was also measured using a count variable ranging from 0 (no participation) to 4 (maximum participation).

Results

Variables Model 1 Model 2 Model 3 Model 4
Feminist 0.40*** (0.11) 0.37*** (0.11) 0.24*** (0.07) 0.23*** (0.07)
Degree of Belief that Gender Inequality Exists 0.04 (0.04) 0.04 (0.02)
Data Source: 2012 Outlook on Life Surveys
Notes: (Standard Error) *** p<0.0001, ** p< 0.001, * p<0.01, ‘.’ p<0.05

By running a few sets of OLS regression models, I was able to test my hypotheses. First (not shown), I regressed degree of belief that gender inequality exists on a feminist self-identity in order to see if identifying as femininst correlated with a higher degree of belief. I found this to be the case- individuals who identified as feminist believed to a higher degree that gender inequality exists (b = 0.29 p < 0.0001). This regression allowed me to establish that I was correct in expecting feminists to have a higher degree of belief compared to non-feminists. I controlled for gender, age, race, region, income, education, and liberalism.

The table above shows the results of models that regressed political participation on firstly, a feminist self-identification, and secondly, on the degree of belief that gender inequality exists. Political participation was measured in two ways, with models 1 and 2 representing explicit participation and models 3 and 4 representing political activism. In models 1 and 3, I do find support for my first hypothesis- self-identifying as feminist appears to correlate with higher explicit and activist political participation. Looking at models 2 and 4 which is the mediation analysis, I do not find support for my second hypothesis. If degree of belief was a mediator in this relationship, we would expect for it to be statistically significant at predicting higher participation and we would also expect that the 'effect' of a feminist label would shrink; the belief would be the true driver of action. However, this is not the case. Degree of belief was not statistically significant and the coefficient on the feminist variable shrank only very marginally. Across all four models, I controlled for gender, age, race, region, income, education, and liberalism.

Reflection

After running my analyses, I was extremely surprised that the degree to which one believes that gender inequality exists did not have a bigger influence on the relationship in question. Because of this, I ran subsequent analyses to see if perhaps the opposite pathway to what I had theorized was true- that there is a relationship between degree of belief in poltical participation and that a feminist self-identity mediated this relationship. To test this, I first ran a logistic regression to see if a higher degree of belief was predictive of identifying as feminist, and found that this was true. As an individual's degree of belief increased, the log-odds of identifying as feminist increased by about 33%. However, when I regressed political participation on degree of belief, belief was again not statistically significantly predictive of higher participation. When I added in the feminist variable, it was again statistically significant. Thus, it seems that the feminist identification certainly has an effect on participation, but it is not driven by the extent to which an individual believes that gender inequality exists.

In terms of implications for future research, this project gets the ball rolling in relation to establishing what is really motivating people to act. While I discovered that degree of belief was not a mechanism underlying the relationship between a feminist self-identity and higher political participation, I do believe some other process is at work. Alternatively, I think a possible mechanism could be the extent to which one believes that change is possible and could truly occur. If one belongs to a collective identity that centers itself on rectifying a social inequality, then they must believe to some extent that their actions can make a difference. Thus, if this belief is heightened through aligning oneself with a particular identity, then this could underly the relationship between identities and increased poltical participation. I think that determining the process that is really motivating people to exercise their political rights is very important and worthy of future research.