Predicting Political Involvement through Demographics, Overall Involvement, and Political Interest

Age, political fascination, frequency of searching out political information, political interest, and membership in non-political clubs had significant collinear relationships and multiple regression weights, as seen in table 1 and 2. These variables then are a safe bet when recruiting political participants, regardless of what kind of participation is needed.

Political perception, political preference, time allotted for club memberships, and religious service attendance had no significant relationship with political involvement, as seen in table 1 and 2. Religious service attendance is the most surprising out of these results, as many previous studies had found it to be a significant predictor of political involvement. Analyses including other religious involvement factors, such as donating money or affiliation, combined into a scale could be a better predictor of political involvement. Religious affiliation and involvement has wider range of definitions, the same way political involvement does. Increasing the scope of this predictor may account for more of the underlying influences of different kinds of political involvement. Or, religious activity could only be significant in predicting active measures of political involvement, such as volunteering for a campaign, rather than passive measures such as frequency of political knowledge acquisition that are included in this scale of political involvement. 

Gender, strength of political feelings, and club membership contribution all had linear relationships with political involvement, but didn’t contribute to the full model until all other variables were held constant, as seen in tables 1 and 2. They were probably all too collinear with the other variables in the full model to have individual significant contributions. Club membership is the most surprising, because simply being involved in one area should influence political involvement later on, according to previous research. Maybe the same issue arose here, as did with the Pew study previously mentioned; an alternative explanation was needed to understand why lower SES individuals weren’t involved politically on social media, even when social media access was held constant. They reasoned that the desire, or example, to get involved just wasn’t there. Perhaps in this circumstance, perhaps the same reasoning is appropriate. Individuals could just be involved in other ways, and not have any desire to get involved politically too.

Partisan strength had a linear relationship but didn’t uniquely contribute to the full model, or its’ reduced model, as seen in tables 1 and 2. These results are surprising, since previous research has demonstrated many times that strongly associating with one party leads to higher involvement rates. The question in the survey “how strongly partisan are you?” is very face valid. Previous research has had a problem addressing the underlying reasoning that participants use when they answer questions this face valid. Snell, 2010, addressed this dilemma by allowing participants to explain themselves. Many said something to the effect of “I am affiliated with one party, but don’t agree with them every single time,” or “I agree with this party often but I wouldn’t identify as one of them,” suggesting that even though participants may actually be strongly partisan, they don’t want to identify that way (even though they vote Republican, and align with those party values, they don’t want to call themselves a Republican or have any attachment to the party). Older people are typically more concerned with party unity than younger people, and are maybe therefore more likely to identify as “very strongly partisan” on a scale. Perhaps, partisan strength is only associated with one kind of political involvement; previous research has found that older and younger people have different amounts of resources (time and money) that affect their ability to be involved in different ways politically. Older people donate more money, because they have more of it, and younger people donate more time, because they have more of it.

None of the models worked as well as the full model in predicting political involvement. But, political interest came close and did better than the other two reduced models. Having certain characteristics, and being involved, can help predict political involvement and are possibly factors that influence involvement, but without any political interest, that potential may not really matter. When identifying people that are going to be involved, in many different kinds of ways, most of these variables are useful. However, if the characteristics that this person has only fall into one reduced model, then they may not be the best candidate in comparison to someone else with characteristics in all of the reduced models.

 

 

Index

Introduction

Methods

Results

Discussion

Conclusion

References

Table 1

Table 2

Full Report