For this assignment, you will measure 3 variables for at least 30 people and use descriptive and inferential statistics to analyze the results. Finally, you will write a report about the study and the results.
- Start by determining what variables you will measure. The following is a list of what you will need.
I. A Categorical Variable with two levels (e.g., A variable of sex with levels of male and female).
II. A Quantitative Variable that you believe would have a statistical relationship with your Categorical Variable (e.g., I may expect the variable of sex to be related to a quantitative measure of the fear of crime).
III. An additional quantitative variable that you expect would have a relationship with your previous variables (e.g., I may expect those reporting greater fear of crime to also report greater fear of terrorism and I may also expect females to report greater fear of terrorism and greater fear of crime compared to males because previous research has shown this).
Note you cannot use any of my examples as your variables. You need to think of new variables that would reasonably be related to each other. Note that I am not grading you on whether the results show a relationship, just that the variables have a logical connection. - Measure these variables for at least 30 people. You could use students on campus, friends/family, or just post the survey to social media. Be sure to consider your variables when deciding how to measure them and the sample you use. If I was measuring the variable of sex (male and female) I would want to make sure I can get a decent number of both male and female responses to make for a good comparison between them.
- Enter your data into Excel in the manner demonstrated in previous videos and lectures. You will turn in the Excel sheet along with a written report of your results. I will not be grading on the structure of the data, but I will expect it to be organized in a way that I can easily verify your results (i.e., be sure all your variables, raw data, and stats are clearly labeled). Note that students need to gather their own individual data and conduct their own analysis.
- Use descriptive statistics to explore the relationships between the variables. This includes using Cohen’s d to measure the relationships between your categorical variable and each of your quantitative variables. Note that if you do things correctly, you will have two Cohen’s d values. For example, imagine I used the variables I listed earlier (sex, fear of crime, and fear of terrorism) for my variables. I would have a d value showing the difference between males and females on fear of crime and a separate d value showing the difference between males and females on fear of terrorism. You will also calculate Pearson’s r for your quantitative variables (e.g., the correlation between the reported fear of crime and the reported fear of terrorism).
Finally, you will include one scatter plot and two column charts. You will have one column chart for each of the group differences you examined (e.g., I would have one column chart showing the difference between the means for males and females on fear of crime and another column chart showing the difference between the means for males and females on fear of terrorism). I would also have a scatter plot showing the relationship between the two quantitative variables (e.g., the correlation between the reported fear of crime and the reported fear of terrorism). Be sure that your figures are formatted in APA style as shown in lecture and previous assignments. - Use appropriate inferential statistics to evaluate the statistical significance of the relationships you examined. Note that you can use your textbook or the website I mentioned in the r redux lecture to find the significance of your Pearson’s r value. For the group differences, the test you will need to run is one we covered in the last assignment. I expect you to select the appropriate test, calculate it correctly, report the results in APA style (as shown in lecture), and describe the results in words. Note again that there are two group comparisons, so there would be two inferential tests to report for them.
- Write a report (2-page maximum, no lower limit) on the study and the results. Report what your variables were, how you measured the variables, the size of your sample, the method you used to gather the sample, and all of the statistics requested in these instructions along with verbal interpretations of the results (e.g. a Cohen’s d of .87 constitutes a strong relationship by Cohen’s guidelines). Be sure you include your figures at the end of the report and mention them in the text of the report.
An example structure I could follow for this assignment
- Start with an opening that introduces your population and the variables you were interested in. This could start as a general introduction that does not get into specifics, but just introduces the key variables you wanted to look at.
- You could then describe what your sample was like. How did you recruit participants? How large was your sample?
- You could then describe the specific ways you measured your variables. You should make it clear what your variables were and identify the levels for each variable (e.g., the variable of fear of crime could have levels between 1 to 10 with 10 being high fear).
- Next, report the results of the analysis for the statistical relationships (both descriptive and inferential). Note that this would include reporting means, standard deviations, d and r values, and the results of any Hull hypothesis test conducted. Be sure to include a written description of what the results indicate along with the relevant statistics. You should also report the results of each relationships one at a time (i.e., report all statistics for one relationship before moving on to another relationship).
- Turn in the report in a Word or PDF document (charts included in the report) and the Excel sheet in an Excel file (with raw data and your calculations) to the link on Blackboard by 11:59 PM on Wednesday, December 19th.
You do not need to report skew or kurtosis for any variables on this assignment. Also, be sure to use proper null hypothesis tests. Do not just reference intuitive null hypothesis testing as the basis for your decision because intuitive null hypothesis testing is an informal method of assessment that should not be reported in any official reports (it is just a tool to help you think about the elements of formal null hypothesis testing). You should also report results in the manner discussed in lecture. You will also need to describe your results in both numbers and words and do not report any statistics or figures beyond those requested. Be sure that you do not just include a table of your results. All I want are written paragraphs with the statistics reported in the paragraphs and the figures at the end. As with the last assignment, you should make your report as concise as possible.
Note. I mentioned during lecture that I was planning on having you make a correlation matrix but decided that a scatter plot would be a better choice. You can just ignore the brief comment in lecture about making a correla