Tired of ads?
Join today and never see them again.
Ok, so by now we've worked our data hardcore in the exercises on the previous pages. We totaled raw numbers, calculated averages, and made pretty plots and histograms to show off the data. Now, we just need to make sure we understand what it says. Let's go back to our fictional survey on social networking, and quickly review our theme question and its data once more.
We asked 50 middle school boys and 50 middle school girls these questions:
Here's what we found*:
|Social Networking Data (% that answered yes to question)||Girls||Boys|
|"Friend" with Parent||66%||50%|
|Mean time spent on these sites||2.20 hr/day||1.01 hr/day|
|Time Spent Social Networking||Girls||Boys|
* This is not real data.
Based on all of this data, it is evident that middle school girls in San Francisco do spend more time on social networking websites than boys. Both their mean and medians are significantly higher. In fact, the girls' median is higher than the boys' upper quartile (Q3). In addition, girls are more likely to "friend" their parents and in turn, their parents are more likely to monitor them. Our study has not proven our hypotheses to be true, but it does strongly suggest that they are likely true.
Based on these results, we might want to ask why parents monitor their daughters more than their sons? Why do girls spend so much free time socializing? What are the boys doing with their free time?
Finally, we need to ask if there is anything else we should look at or more questions that we should ask. For example, we could extend our survey beyond San Francisco. We could also try to break the data down by age: do 13-year-olds spend more time on these social networking sites than 12-year-olds?
It's also important to ask why and how we got the numbers we did. Was the data we gathered correct; did students answer truthfully? Could some of the boys have under-reported their time spent on social networking sites? Did some students overestimate their time? These are all great questions to push our statistical experiment further. We bet you can come up with many more!