Weaknesses

Weaknesses to our study could be not getting responses from people who didn’t answer their door because they thought we were soliciting something or maybe they thought we wanted something other than a simple two question survey. The time we went out to gather our data could be a confounding variable. The times in which we went out was about 4-7 on two different occasions. Most people get home from work at around 5-6 so this means we might have missed some houses we initially selected because no one was home when we went to their house. This means that we may have excluded people who work later hours in the day.

Extrapolate

We decided that we could extrapolate our data and study to the rest of suburban Northeastern Ohio cities on the matters of voter registration between males and females. We realize there maybe exceptions within certain communities with religious or urban settings so that is why we decided that we would not be able to extrapolate this study to all of Ohio or the rest of the country. We feel comfortable extrapolating our results to other  suburban Northeastern Ohio cities due to the fact that studies which we found on some surrounding neighborhoods produced similar results. Our extrapolation to other surrounding Northeastern Ohio cities was justified by the results we found on similar studies but it was confounded to just that based on the differences that we realized existed in other areas of Ohio and the rest of the country.

Suggestions

With concerns to further work in this topic, a census would be a way to find the true proportion of registered males to registered females and see if a two proportion z-test shows a significant difference between the two data sets. But if a census can not be obtained accurately because of response bias, knowing the actual proportion of males to females is going to be hard to predict or test.