Discussion

WEAKNESSES

As we were conducting our study, we found some weaknesses. First, there was a nonresponse bias. We sent out 60 teacher surveys and only received 44 back. In order to minimize this bias we could have sent out more than 60 surveys to increase the sample size and hopefully get more than just 44 surveys back. However, there was no nonresponse bias with the students data because that came straight from a list and we were able to just randomly select 30 students. Also, we dealt with response bias in our sample. Four teachers did not answer the question correctly and only put the model of their car such as “Chevy Cruze” instead of the model year of that car. This forced us to eliminate these responses from the experiment which left us with less responses. Increasing the sample size would have produced a more accurate study, but besides doing that, there aren’t many other ways to reduce the influence of response bias. We also encountered no measurement bias in our study. We could have strengthened our study by stratifying based on the senior class or making sub groups on salaries the teachers make compared to what students make.

EXTRAPOLATION

We could extrapolate our data to all city schools in the Cuyahoga County area.  We feel that in some school systems our study would remain true, but there are many school systems that are in better shape (money wise) than North Olmsted.  Therefore, many students may have just as new of cars as the teachers.  North Olmsted High School has the perfect balance of middle class families that make this study well rounded.

SUGGESTIONS FOR FUTURE STUDY

Further work we could do for this study is to get the amount of income each teacher made with the year of their car.  We could also do the same for students, obviously asking if they had a job first, and compare their yearly income with the year of the car they drive.  To do this we could have run a chi-squared test to see if there is an association between the two.