Northwood’s AP Statistics and Journalism classes started a combined project a few weeks ago, which aimed to discern the Northwood community’s candidate preferences in the 2024 US Presidential election. The stats class was trying to minimize bias, so they decided the best way to do that was by randomly surveying 25% of each grade at Northwood and 25% of the faculty. A sample size too large lacks independence of the individual data points, whereas one too small doesn’t have a robust enough data set to draw conclusions from. This would be 56 total people surveyed; 45 responded which is just over 80%.
The data came with a margin of error of +/- 13%, which means that the researchers are 95% certain the percentages we found are within 13% of the true percentages. That being said, all of the data and trends that were recognized should be taken with a grain of salt. Despite the small sample size and large margin of error, we were able to find some statistically significant associations between cohort and political preference in a few cases.
In statistics, an association can only be concluded if the “p” value, which is the probability that the event in question happened by chance or random, is less than 0.05, and in the survey, there were only two cohorts that represented some relationship. The surveyed hockey players had a p-value of 0.018 in terms of voting for Donald Trump. If there was no political preference for the Hockey Cohort, there would be a 1.8% chance that we could have gotten data that is skewed by chance alone. There is also a similar association between NAS/independent students at Northwood and a preference for Kamala Harris. This doesn’t mean that every hockey player at Northwood is going to support Donald Trump. Association doesn’t imply causation, which again means that even though there is a trend, that doesn’t mean that someone who is an independent or participates in NAS will vote for Harris.
Snowsports, soccer and faculty showed no real statistical association, but faculty was close, we are 85% sure the faculty prefer Harris but need to be 95% confident to be able to make conclusions. You can click on this link to view all of the data from the survey. The images below also represent political preference by cohort.
The AP Statistics survey of the Northwood community revealed a statistically significant association between students’ hockey participation and their preference for Donald Trump.
They also concluded a statistically significant association between their participation in the NAS/independent cohort and their political preference for Kamala Harris.
While the raw data may lead someone without knowledge of statistics to conclude candidate preferences in other cohorts, researchers cannot conclude that those differences did not occur by chance.