Do after school jobs affect grades?

1. Design an appropriate study or experiment for the above question, free of bias, randomized, that will produce useful data.

2.Make clear summaries - graphical, numerical, and verbal.

3.Produce a sophisticated statistical analysis.

4.Reach statistically justifiable conclusions about the question "Do after school jobs affect grades?"

5.note a complete report.

Written Report Example Too Many Ads Is the proportion of advertisements in teen magazines greater than the proportion of advertisements in news magazines? While the advertisements in those magazines are quite different, there is reason to believe that there are more ads aimed at teens. The goal of this study was to determine whether the proportion of advertisements in these magazines was significantly different and whether there is a greater proportion of advertisements in teen magazines. Two samples were taken out of the population of the October, November, and December issues of Seventeen and the population of the first three issues of those months of Time magazine. Pages in the samples were randomly picked from the entire population of pages in those magazines. Only the pages that had at least a half page or more of advertisements were counted. Out of the 105 pages samples from Seventeen 61 pages were advertisements and 103 pages sampled from Time 47 were ads. A large sample, one sided significance test for the difference in two proportions was used to examine whether the difference is statistically significant. The assumptions for this test were fulfilled. The two samples were independently selected simple random samples and the sample sizes are large (greater than 30). A significance level was set before the test was performed at .05. The p value for the test is .036, which is statistically significant at the .05 level. The sample was found to provide strong evidence to reject the null hypothesis and the proportion of advertisements in Seventeen is the same as the proportion of advertisements in Time. A 95% confidence interval was calculated to estimate the true population difference between the two proportions. The interval, (-.0101, .2594), says that if this study were repeated 100 times and 95% confidence intervals were made for each sample then about 95 of the 100 confidence intervals would contain the true population difference in the two proportions. The confidence interval for this study estimates that the proportion of ads in Seventeen is anywhere between 1% less and 26% greater than the proportion of advertisements in Time. This interval is very wide due to the small sample sizes. Both the significance test and confidence interval would be more accurate with larger samples. There are a few weaknesses in this study. Time magazine is a weekly issue while Seventeen is only printed every month. This may have been a confounding variable in the study. To try to eliminate this bias the three issues from Time arrived at newsstands at the beginning of the month, the same time the monthly Seventeen were out. The size of the advertisements also may have affected the results. While most advertisements in Seventeen were full page ads most ads in Time were less than a page. The low power and the wide confidence interval both suggest that one cannot entirely rule out the null hypothesis as incorrect even though the sample is statistically significant at the .05 level. In conclusion, the sample is not practically significant. There is evidence to support the alternate hypothesis that the proportion of advertisements in Seventeen is greater than the proportion of advertisements in time, but the difference is small in magnitude