What’s the Best Predictor of Skidmore Population’s Probability of Having the Seasonal Flu?

Courtesy of The Telegraph By Mohammed Almashhadani, '15

 

Despite efforts on behalf of the Skidmore Health Services, which provides flu shots as well as information on prevention, Skidmore community members inevitably get the flu each year, usually between January and February. This observation could be associated with the high number of people who do not perhaps take the flu shot, do not exercise often, do not wash their hands regularly every day, or engage in other unhealthy habits. Those left unaffected by the virus are left wondering to which factors they can attribute their success.

A study has been conducted on campus this semester to track flu cases. The survey, created with Survey Monkey, was shared on Facebook, emailed, and posted on Skidmore all-class announcements. It asked the members their gender, age, hours of weekly exercise, hours of weekly socialization, number of times they washed their hands daily, whether they got the flu shot in 2014, and whether they had a chronic disease. All of these factors became the explanatory variables. The response variable was whether they got the seasonal flu in 2015 or not. Ninety Skidmore community members responded to the survey.

Prior to using statistical methods to determine the best predictor of the response variable, certain groups of respondents were assumed to be at a higher risk for the flu: those who had not gotten the flu shot in 2014, did not wash their hands regularly, and those who spent many hours socializing with others.

According to the statistics program Rstudio, those who have chronic diseases are more likely to report having the seasonal flu in 2015. These results suggest that although the presence of a chronic disease does not guarantee the contraction of the flu, it increases the likelihood of getting the virus. The results are considered statistically significant.

This study does not provide enough evidence that hand washing and lack of vaccination increase the likelihood of actually getting the flu. A greater response pool could potentially have altered the findings.

Because the survey was open to all Skidmore community members, sample bias was avoided. However, because the survey was distributed heavily on Facebook, the demographic of respondents may not be completely random.

The results of this data study conclude that those with the highest probability of getting the flu in 2015 were those suffering from chronic disease. Other factors cannot be linked to the flu given the data collected; a greater number of responses would be necessary to support them. If more people took the survey or answers to the questions were different, other factors could have been statistically significant.