Table 2.
Description of the main study variables in study 1. Note: each factual minority status coded as 0—majority, 1—minority. The distribution of perceived minority status as a function of factual minority status across countries is displayed in the electronic supplementary material, table S3.)
| variable | example item | response format and categorization |
|---|---|---|
| susceptibility to misinformationa | ‘the new 5G network may be making us more susceptible to the virus.’ (five items; USA: α = 0.89; UK: α = 0.87; Poland: α = 0.83; Germany: α = 0.80) | 6-point Likert scale from 1 (very unreliable) to 6 (very reliable) |
| self-perceived minority status | ‘to what extent do you consider yourself part of a minority group based on your… [ethnicity, religion, political views, sexual orientation, gender identity, social class, region of living]’ (seven single items in total) | 7-point Likert scale from 1 (strongly disagree) to 7 (strongly agree) |
| factual minority status (each 1 item) | ||
| ethnicity | ‘please indicate your ethnic origin’ | coded as 1 (ethnic minority) or 0 (ethnic majority) |
| religion/atheist | ‘do you belong to a religious denomination? If yes, which one?’ | two dummy variablesb, one coded as 1 (religious minority) or 0 (religious majority). The other coded as 1 (atheist/non-religious) or 0 (religious majority) |
| political orientation | ‘please indicate below to what extent you consider yourself as liberal or conservative in terms of political attitudes’ | 11-point Likert scale from 0 (very left-wing/liberal) to 10 (very right-wing/conservative). Two dummy variablesc coded as 1 (political minority) or 0 (political majority) |
| sexual orientation | ‘what is your sexual orientation?’ | coded as 1 (non-heterosexuals) or 0 (heterosexuals) |
| gender identity | ‘what is your gender?’ | coded as 1 (women/non-binary/other) or 0 (men) |
| social classd | ‘what is your highest level of education?’ and ‘what is your annual income?’ | coded as 1 (low social class) or 0 (not low social class) |
| region of residence | ‘what best describes your area of living?’ | coded as 1 (rural residents) or 0 (urban residents) |
| attention checke | ‘it is important that you pay attention to the survey. Please select ‘somewhat agree’ | two items interspersed within other measures on 5- and 7-point Likert scales |
aAdopted from Roozenbeek et al. [17]. The susceptibility to misinformation scale was computed by averaging the six false statements (see the electronic supplementary material, table S1 for all items). One false statement was dropped to improve measurement invariance and achieve metric invariance (see the electronic supplementary material, S1 and table S2) for the remaining five-item scale.
bBy simultaneously introducing both dummy variables to the model, we were able to test the effect of belonging to a religious minority group or being non-religious relative to belonging to the religious majority group.
cBy entering both dummies into the models we compared the effect of belonging to the left-wing and the right-wing minority compared to the political midsection.
dParticipants reporting an income of less than 66% of the median income who also had not completed high school were coded as the minority group.
eParticipants who failed both attention checks were excluded from the final sample.