SPSS ASSIGNMENT
Follow the steps below to complete your SPSS homework assignment:
1. From Blackboard, download the SPSS HW #3 data file. The data file is from a handwashing study that was conducted at University XYZ.
The researcher, Dr. Z, conducted the study to see if students' sex and age had an impact on how often students wash their hands and how many seconds they wash their hands (this information is important to know when entering the "role" of each variable in SPSS).
2. The data file is based from the survey questions below. Based from the survey questions, correctly enter the variables' label, values, measure, and role in the variable view in SPSS. For the sex variable, please give male a value of 1 and female a value of 2.
What is your sex?
Male
Female
What is your age?
(participants wrote down their age in the survey)
How often do you wash your hands after using the bathroom?
All of the time 5
Most of the time 4
Half of the time 3
Some of the time 2
None of the time 1
How long do you wash your hands with soap when washing your hands after using the bathroom?
(participants wrote down their handwashing time)
3. Create a "file information" output of your variables in SPSS.
4. Then, "clean" the data set by deleting all participants who did not answer one or more survey questions. Do this by right-clicking on the first variable.
5. Then, click on "Sort Ascending."
6. SPSS will arrange the data with the lowest values at the top and the highest values at the bottom. Missing values (the survey questions that were left blank by participants) will be at the very top and will be represented by a period.
7. Then, select the rows that have the missing values for the first variable by clicking the number 1, holding down the click, and moving your mouse down to the row that contains the last missing value
8. Then, delete those rows by pushing the delete button on your computer. This will automatically bring up the rows of data.
9. Then, repeat this for each variable so that there are absolutely no missing values throughout the entire data set.
Central tendency and dispersion
10. Then, use SPSS to determine each variable's correct measure of central tendency and dispersion. Do this by clicking on "Analyze" → "Descriptive Statistics" → "Frequencies."
11. Then, the box below will pop up. Click on the first variable and then click on the arrow.
12. This will move the variable over to the "Variable(s)" box.
13. Then, click on the "Statistics" button
14. Then, click on the correct measures of central tendency that are applicable to the variable's level of measurement (e.g., nominal, ordinal, scale).
15. Then, click on the correct measures of dispersion that are applicable to the variable's level of measurement (e.g., nominal, ordinal, scale) (hint: we can only look at levels of dispersion on scale variables). Only select std. deviation, minimum, and maximum. Then, click "Continue."
16. Then, click "OK." This will create an output that has two tables: (1) one table for the central tendency & dispersion (if applicable depending on its level of measurement (nominal, ordinal, scale) and (2) the other table for the frequencies.
17. Then, repeat the process for each of the variables. In order to do so, you will need to move the variable you just analyzed out of the variable box in SPSS.
Do this by clicking on "Analyze" → "Descriptive Statistics" → "Frequencies."
18. Then, the box below will pop up. Click on the variable that you had just analyzed previously and then click on the arrow.
This will move the variable back over to the box on the left.
19. Then, click on the next variable that you need to analyze and repeat the process for each variable.
Analytic testing for dependent variable normality
20. Now, we need to test if our dependent variable is approximately normally distributed. To do so, we have to test the dependent variable for each category of our independent variable.
21. To do this, click on "Analyze" → "Descriptive Statistics" → "Explore."
22. For this SPSS HW assignment, we are going to look at the variables of "sex" and "seconds of handwashing." Which of these is the dependent variable? Which of these is the independent variable?
23. Since "seconds of handwashing" is the dependent variable, click on it, and then click on the arrow that will move it over to the "Dependent List" box.
24. Since "sex" is the independent variable, click on it, and then click on the arrow that will move it over to the "factor list" box.
25. This will move the variables into their respective boxes. Then, click on "Plots."
26. Then, from the pop-up box, click here to de-select the "Stem-and-leaf" option (we do not want it). Then, click here to select the "Histogram" option. Then, click here to select the "Normality plots with tests" option.
27. Then, click "Continue."
28. Then, click "OK."
29. Then, this output will show up. We will have to hand calculate the z-value for skewness and kurtosis to determine just how skewed or kurtotic the dependent variables are for each category of the independent variable. If the z-value falls between -1.96 and +1.96, then the dependent variable is not skewed or kurtotic. To perform this calculation, you divide the skewness and kurtosis statistic by its standard error. In the example below, we will divide the following:
1.111 / .261 = 4.25. Yikes! This value is way beyond +1.96. This means we have some really skewed data.
30. Repeat this same calculation for kurtosis to determine just how kurtotic the dependent variable is for the female category of the independent variable. This is exactly the same as calculating skewness. If the z-value falls between -1.96 and +1.96, then the dependent variable is not kurtotic. To perform this calculation, you divide the kurtosis statistic by its standard error.
31. Next, let's determine if the data are normally distributed. We do this through the Shapiro-Wilk test, which is the next item below in your SPSS output. Please ignore the Kolmogorov-Smirnov test. If the value of "sig" is >.05, then the data are normally distributed. If the value of "sig" is ≤.05, then the data are not normally distributed.
What do you think? Is our data for SPSS HW #3 normally distributed?
Visual testing for dependent variable normality
32. Next, let's look at the histograms to visually see how the data are distributed. From the histograms, do the data look like normal curves?
33. Next, let's look at the "Normal Q-Q Plots." If there is a normal distribution, the dots should line up very tight against the line. What do you think? Are the dots tight against the line?
34. Ignore the "Detrended Normal Q-Q Plots." We will not need them for this course.
35. Now, let's look at the box plots. If the data are normally distributed, the box plot's whiskers will appear symmetrical. In the example below, there is only 1 set of whiskers for each box plot. This means our data are seriously skewed! If the data were normally distributed, there would be another whisker here and here.
36. Next, export your SPSS output to a Word document. Do this by clicking on the arrow symbol on the SPSS output view.
37. The box below will pop up. Browse where you want to save your file to and create a file name. Then, click "OK" to save the output as Word document.
Attachment:- SPSS Assignment.rar