7.2 The Study: Social Identity
7.2.1 Study Information
7.2.1.1 Description
People desire positive social identities. A positive social identity is achieved when the social groups to which one belongs (i.e., the “ingroups”) are perceived as favourable to the social groups to which they do not belong (i.e., the “outgroups”). Such positive social identities are most often fostered by favouring ingroup members, rather than by devaluing outgroups. As such, people tend to feel positively towards those they see as part of their ingroup.
The desire for a positive social identity can lead to discrimination when positive feelings are reserved for one’s ingroup. However it can also promote acceptance of diversity. Preserving positive social identities requires accepting diversity within one’s social groups. As such, broadly defining one’s ingroup can foster a greater acceptance of diversity. Therefore, encouraging the adoption of broad social groups has been proposed as a means of combatting discrimination.
Of course changing how people define their ingroup is difficult. It can occur when one changes social groups, such as when one moves to a new community, makes new friends, or changes political parties. More often, however, broadening one’s ingroup is accomplished by strengthening ties with existing groups. That is, rather than changing the social groups to which they belong, people are more likely to change which social groups they see as important to their social identities.
This study will examine one mechanism by which people determine the importance of social groups to their social identities: perceived social group threat. Research has demonstrated that people strengthen ties to social groups when they perceive those groups as threatened in some way. This was evident in the aftermath of the 9/11 attacks on the United States. Many US citizens felt a threat to their American social identity. This led to a dramatic outpouring of displays of patriotism, such as flying the US flag on one’s property.
This study will investigate a similar effect using a Canadian sample. Although several studies have demonstrated this effect, there is value in replicating these findings. Much of this research has been conducted using US samples, and although Canada and the US are similar in many ways, their political climates are distinct. Additionally, many of the foundational studies demonstrating this effect were conducted in the early 2000s and before; since then, the political climate in both Canada and the US has changed. For example, political convictions are more rigid than they were. These factors could mean that the effects observed among Americans in these foundational studies will be different than what can be expected among Canadians today.
7.2.2 Design Plan
7.2.2.1 Study Design
This study will employ an experimental design adapted from experiment three of Davies et al. (2008). Participants will read a brief newspaper article describing either a foreign or domestic event that could be perceived as threatening to Canadians. They will then complete a measure of national identity. Participants will be recruited at in-person public events throughout the Okanagan during the summer. Research assistants will operate tables at events throughout the summer. For completing the study, participants will be offered $5.00 food vouchers redeemable at participating vendors at the event. Recruitment will continue until sufficient data are obtained or until the summer’s end.
Individuals must be at least 18-years-old and Canadian citizens to be eligible to participate.
7.2.2.2 Measures
National identity will be measured with items adapted3 from Davies et al. (2008). They will be measured on a 7-point Likert-type scale ranging from 1 (not at all) to 7 (completely).
- Do you identify with being Canadian?
- Is being Canadian important to you?
- Are you proud to be a Canadian?
- Do you think of yourself as a Canadian?
Participant scores will be averaged across these four items to create a composite measure of national identity.
7.2.2.3 Analytic Strategy
A directional Welch’s t test will be conducted with the alternative hypothesis that national identity is higher in the foreign threat condition. Significance will be inferred using the traditional \(\alpha\) = .05.
7.2.2.4 Sample Size and Power Analysis
We plan to recruit 254 participants divided evenly into the two experimental conditions. This will provide 95% power to detect an effect if the effect size in the population is d = 0.414.
Davies et al. (2008) reported mean national identities of 5.57 and 4.99 for participants who read about foreign and domestic threats, respectively. This amounts to a raw difference of 0.58 on the scale of 1 to 7. Because the SDs for national identity were not reported, we cannot know the standardized effect size for this effect. We therefore based our effect size calculation on an assumed pooled SD of 1.4. Under this assumption, a raw effect of 0.58 translates to d = 0.414.