In addition to asking whether ‘politics is for people like you’, the survey asked each respondent whether they consider themselves to be a ‘political person.’ These two questions both focus on people’s relationship with politics but the second shifts the focus from the general accessibility of the system to individual dispositions. The order of the two questions was randomised, meaning that half of the respondents saw the ‘people like you’ question first, whilst the other half saw the ‘political person’ question first. This may alter the answers that people give (by making them first think about themselves or the about the accessibility of the political system), and this possibility will be investigated separately.
The answers to whether people see themselves as political are presented in Figure 1 (above, using weighted data) and, as with whether ‘politics is for people like you’, almost two fifth of people (38.9%) pick the middle option (‘Somewhat’). However, people are less willing to say that they are political than they are to say that ‘politics is for people like you.’ Approaching two fifths (37.1%) say that they are not really (27.1%) or not at all (10.1%) a political person, whilst just under a quarter (24.0%) say that they are largely (19.1%) or completely (4.9%) a political person. This distribution of answers is almost the mirror image of those regarding whether ‘politics is for people like you.’ Thus, people again err toward the middle option on this question but, beyond that, are notably less likely to consider themselves political than they are to consider that politics is for people like them.
Here we can see hints of the difference between internal and external political efficacy. The former focuses on our own capacities to navigate and engage with the political system, simple or complex as it may be, whilst the latter focuses on whether the system itself is open and accessible to people. The similar but contrasting answers to the ‘political person’ and ‘politics is for people like you’ questions indicate, perhaps surprisingly and perhaps reasonably, that people’s external efficacy is slightly higher than their internal efficacy. However, there are other questions in the survey that focus more explicitly on both types of efficacy and it remains to be seen whether the answers to those relate to the answers to these two questions. This will be investigated subsequently, as will the relationships between a range of variables relating to political engagement.
| Variable names | poli_polpers1, poli_polpers2 |
| Number of cases | 1,405 |
| Number of categories | 6 |
| Categories to code as missing | 9 (‘Not asked’) |
| Cases to code as missing | 701-704 |
| Recoded variable name | pe_pp_cir |
| Number of cases | 1,405 |
| Number of categories | 5 |
| New and old categories | The two original questions have identical wordings and answer options, and were asked separately as part of a question order experiment. As such, they needed to be combined into a single variable containing all respondents’ answers. At the same time, the numerical values of the answers were inverted (by deducting the original answers values from 5) so that higher values equate to a greater sense that the respondent is a ‘political person.’ In light the above, the scale of the new variable runs from 0 (‘Not at all’) to 4 (‘Completely’). This is in contrast to the original two variables, which ran from 1 (‘Completely’) to 5 (‘Not at all’). The respondents with missing values on the first of the two original questions gave substantive answers on the second, and vice versa. Thus, the combination of the two variables into the new variable eliminates the ‘Not asked’ (9) category, and there are no missing values. |
Privilege and Participation 
3 thoughts on “Survey Variable: Political Person”