Week 3 Assignment Worksheet:
Analyze Data Using Bivariate Correlation, Regression, and T-Tests
I. Correlation and regression analysis.
Please open the file name personality.sav. In this exercise, you will focus on the relationship of two variables, beckdep (a measure of depression), and emcontot (measure of emotional control – (CECS) Courtauld Emotional Control Scale). These are two instruments used in psychology, the Beck Depression Inventory, which is used to measure characteristics related to depression, and the CECS which is used to measure subjective control of feelings of depression, anger, and anxiety in uncomfortable situations; this is a self-description instrument. Please read the instructions below.
Scenario
A researcher is interested in assessing the relationship between these 2 instruments, with the goal of using the emcontot scores as a predictor of the beckdep scores.
1. In SPSS, calculate the Pearson’s Correlation between beckdep and emcontot.
a. What are the null and alternative hypotheses?
b. Present a scatterplot with beckdep scores in the Y-axis, and emcontot in the X-axis.
c. What type of relationship is depicted in the scatterplot, positive or negative?
d. Conduct the correlation analysis using the correlate menu in SPSS.
e. Report the results of the correlation analysis in the APA style.
f. What was the amount of variance shared by the two variables?
g. What is your decision regarding the null hypothesis based on the results of the correlation analysis?
2. In SPSS, develop a regression model between the variables emcontot and beckdep.
a. Which of these variables is the dependent (predicted) variable?
b. Conduct the regression analysis using the regression menu in SPSS.
c. What is the mean and standard deviation for emcontot and beckdep?
d. What is the R-value? Is the value similar to the correlation coefficient conducted above? If so, why do you think is similar?
e. What is the value of the R square? What does that value tell you about the amount of variance predicted by emcomtot on beckdep?
f. Was the model significant? How could you determine this?
g. Report the results of the regression in the APA format.
h. Develop the regression equation for the predicted variable
i. Using the Regression equation developed above, please estimate the Beck Depression Inventory score for an individual who score 2.50 in the emcontot (measure of emotional control – (CECS) Courtauld Emotional Control Scale).
II. T-test analyses
Scenario
The same researcher is now interested in examining if differences in the Beck Depression Inventory score exist between individuals who are married, and not married. In order to determine if differences exist, the researcher will use the variable name marital as the independent variable, and beckdep as the dependent variable. Once the researcher examined the variable marital, he realizes that the variable included 6 different categories. Therefore, the variable needs to be recoded in two categories, single (including separated, divorce, or widowed), and married; because the independent samples t-test is used to compare differences between two groups.
3. In SPSS conduct an independent samples t-test
a. Your first task is to recode the variable marital into two categories, single and married.
i. From the SPSS top menu, complete the following:
1. Go to transform.
2. Click on Recode into a different variable.
3. Highlight the variable marital and enter it in the numeric variable – output variable window.
4. In the output variable window under name, write new_marital, and under label, marital 2, then click change.
5. Then, click old and new values.
6. Enter the following values;
a. 1 old value, and 1 new value, then click add.
b. Continues entering in old and new values 3-1, 4-1, 5-2, 2-2, and 9-9, remember to click add after each entry.
7. After entering all of the values, click on continue, and the newly created variable will appear at the end of all of the other variable in the data view.
b. Conduct the independent samples t-test, including the assumptions tests.
c. There are three main assumptions that need to be satisfied before using the independent-samples t-test for testing differences between the genders. Use SPSS to generate the output needed to test the assumptions. Please discuss each one and explain whether each has been met using SPSS output as needed to include the Shapiro-Wilk test for normality (used when the sample size is less than 50), histograms, and the Levene’s test. Keep in mind if the population value is unknown, it is permissible to infer from sample values. Regardless of sample size, test whether these assumptions are met.
d. What are the null and alternative hypotheses?
e. What is the mean of the Beck Depression Inventory for the single and married groups?
f. What is the value of t?
g. What is the associated probability?
h. Report the results in APA format.
i. What might be concluded from this hypothetical study? (Hint: The decision about the null hypothesis.)