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.