Weaknesses: There were some weaknesses and confounding variables in our study. One of the weaknesses was the presents of a non-response bias. Out of the 75 surveys we distributed, only 60 were returned, clearly indicating the non-response bias. One major weakness of our study was that our categories were too broad. With our original four possible categories for each question, we has multiple cell counts less than 5, so we had to continue combining our categories until there was no cell count less than five, which took until we had only 2 categories for each question.
Extrapolation: For our study, we could extrapolate to almost all of the students at High Schools in suburban areas that are similar in nature with North Olmsted because in our study we got a fairly diverse group of teenagers in our school and since our study involves technology that isn't available to everyone in all demographics, therefore we have to narrow or extrapolation down a little.
Suggestions for Future Work: If further work in our study was to be done the samopling size should be greatly increased to eliminate expected cell counts of less than 5. There should also be more available answer options when it comes to the survey question because our intervals covered too big of ranges, which made making a distinctive decision on whether GPA was affected by the amount of time spent on Facebook difficult.
