Optional Extra Credit Opportunity
1) Think of a topic you want to research and write up (1) your research question, and (2) a brief questionnaire (maybe 5 or 6 questions) which includes at least two nominal variables and at least two ratio variables. For example, suppose you want to study the factors that affected people’s score on Midterm #1. In that case you might want to include questions such as:
• Do you work? ___ Yes, full time ___ Yes, part time ___ No
• Gender: ___ Male ___ Female
• How many hours did you study for Midterm #1 the week before the exam? _____
• How many hours of sleep did you get the night before the exam? ___
• What was your midterm score? ___
You get the idea. Your topic could be based on something at work, something from another class, something about a hobby, or whatever. It’s up to you. NOTE: you MUST get your research question and questionnaire approved by me before you start…check with me no later than Wednesday, May 2, or you can’t do the project. E-mail communication is fine, or just stop by to see me. NO EXCEPTIONS. This time frame will give you plenty of time to get the project done
2) Collect data from 20 or 30 of your classmates, friends or family members. You can either use a paper/pencil approach, or else I and the graduate assistants can show you how to do online surveys using Qualtrics or Google Docs. I just want you to have enough data to play with! Feel free to arrive to class early and ask your classmates to fill out your survey. I don’t expect you to use probabilistic techniques to select your sample, but think about how you would have done the study if you’d had the time and money to do it right. You will have to include a methodology section in your final report, indicating what the “right” way to sample would have been. You should have your data collected by Wednesday, May 23 so that you have enough time to work on your analysis.
3) Enter the data into Excel. Each person should have one row. You should have one column for each variable. For the above questionnaire the spreadsheet might look like this:
Person Work Gender* Study Sleep Score
1 No Male 6 2 90
2 No Female 5.5 3 80
3 Part time Male 1 4 50
4) Run some descriptive statistics (proportions, means, standard deviations) and graphs. In other words, show me that you have mastered chapters 1 – 3. Be sure you use the proper type of analysis depending on the type of data you have (so don’t try to find averages and standard deviations for nominal variables like gender or work)!
5) Conduct at least 2 hypothesis tests or confidence intervals using the data. For example:
a) In the past, people got an average of 6 hours of sleep the night before the exam, but I think that may be decreasing due to the complex lives we now lead. Is it true that sleep time has decreased?
b) Is there a relationship between the number of hours of study time and the midterm score?
c) Find the 95% confidence interval for the proportion of students who got a passing score on the exam.
You get the idea….you might want to check your analysis plan with me no later than Wednesday, May 30 so that I can re-focus you if you are going the wrong direction.
6) Write up a brief report of your findings. Pretend you are writing it for an employer who knows nothing about statistics…it must be in business format and must include the following:
• An introductory paragraph indicating the topic (research question) you are studying. In other words, state the reason why you are doing this study – other than for the extra credit points, of course.
• A paragraph about the methodology used and the methodology you SHOULD have used if you’d had the time and money to do it right. Be sure to identify the target population, best sampling technique to use (be specific), best mode of delivery of the survey (phone, text, e-mail, mail, etc.). I want details! Just saying “I would use a stratified sampling technique” isn’t enough…what are your strata, what list(s) will you use to draw from, would it have been best to do a phone survey or mail survey, etc.
• Statistical analyses (presentation of all graphs, calculations, hypothesis tests using the 6 step procedure discussed in class, etc.) – this will take several pages.
• A final section which should present an overview of your interpretation of the findings. This would be like a “briefing” you would give an employer.
NOTE: there is no page limit/requirement, but typically it takes around 5 – 10 pages to do what I’ve asked for in your report.