Peer Influence on Political Preference Post: More Data Analysis

The variables describing activity and political preference are all ordinal variables. The values are able to be placed in a meaningful order, however the values cannot be used to make conclusions based on a consistent degree of variation. Because of this, I found the best way to analyze the data was through chi-squared tests. I am using SPSS Statistics for my analysis.

Initial Analysis
To analyze the sample, it is important to understand their climate, or in this particular research, the political climate. The 2016 Presidential election tremendously increased political activity among citizens of all ages. Political activity is much more than attending a march or campaign event; it can include posting on social media to voice your opinion, contacting your representatives, and voting. I listed many examples of political activity to avoid any discrepancies regarding the conceptual definition. According to other studies, nearly 2/3 of social media users use the platforms to engage with others regarding politics or share their opinions. I certainly noticed an increase in political social media posts as well as discussion of politics in 2016 and 2017 versus 2014 and 2015, when the older portion of the sample began school.
I split up my sample into two groups. One is comprised of subjects from the social classes of 2017, 2018, and 2019, while the other is classes of 2020 and 2021. I predict that the older group will be more influenced by their peers, while the younger group will show as more influenced by their families.
I began by comparing the pre-college political activity of the two groups. As I expected, the younger group was slightly more active. Their time prior to College includes 2016 and 2017, two years of high political activity. 75% of the whole sample considered themselves either “not active” or “sparingly active”, which is not surprising for incoming college freshman.
Also as expected, fewer students considered themselves “not active” or “sparingly active” when asked about their current political activity. Only about 15% of students across all social classes decreased their political activity from entering college until now.

Now that we know students at the College are more political active, how are they being active?

Engagement in political discussion can be indicative of knowledge of current events, interest in political figures, and desire to share beliefs with others. As many of us are aware, William & Mary students are very politically engaged! Nearly 45% of subjects either often or always discuss politics with their peers. Many that described themselves as sparingly or somewhat active often discuss politics with peers.
Similarly, 50% often discuss politics with their family, and 12.5% always do. Surprisingly, almost 3/4 of the subjects said they never (48%) or rarely (25%) discuss politics on social media or attend political events. Exactly 2/3 of subjects never contact their elected representatives.

I noticed that 16.7% of subjects reported they never voted in elections. Since I remembered this was similar to the portion of non-U.S. citizens in the samples, I ran a chi-squared test to check. I was extremely surprised to find the same number of U.S. and non-U.S. subjects never voted in elections, but 66.7% of non-U.S. subjects found themselves in this category. To further investigate this, I also performed a chi-squared test with the voter registration age and voting frequency. All subjects who answered that they never or rarely voted were not registered to vote.

I conclude that greatest political activity is derived from voting and political discussion among family and friends, which may lead to additional activity with political groups.

Well, unfortunately the overall political preference was different from my prediction. More subjects from classes of 2020 and 2021 than 2017-2019 considered themselves either extremely liberal or moderately liberal. I expected the younger portion of the sample to trend more liberal than the older portion.
Overall, one third of subjects deem themselves extremely liberal, while only 16.7% say they are either moderately or extremely conservative. 20.8% consider themselves moderate. This is different from national average for young voters. In 2016, the Brookings Institute found that young voters were 37% liberal, 38% moderate, and 26% conservative. As expected, the median William & Mary student are more liberal than their youth population counterpart.

While my hypothesis that younger students at the College will be conservative than older students because they are more influenced by their parents was not proven correct, the data still produced insight into the political climate at the College. I’ll touch more on that in my final post.