This is a three part lab. You do not need to include SPSS print outs. 

Part I

Access the data file "HovlandSearslynch.sav" on Blackboard.  File includes the following variables:

• year
• price for one "unit" of cotton
• number of lynchings

This is the data from the famous Hovland & Sears study on displaced aggression and the frustration-aggression theory. Cotton was the primary source of income in the south during this time period. When the price of cotton was low, the plantation owners and other business owners in town would experience frustration because of the poor economy.  The frustration theory argues that frustration always leads to aggression.  In this instance, however, there was no obvious person to aggress against.  So, Hovland and Sears argued that the need to aggress was displaced onto an easy target. In this case, the "easy" targets were Blacks, accused of flimsy and/or unsubstantiated claims, who were then lynched, ostensibly as punishment.

1.  Use the data provided to test, first, the basic hypothesis that the price of cotton is correlated with the number of lynchings.  Report Pearson's r and the significance level in APA style, and comment on whether the correlation supports Hovland & Sears' hypothesis or not.

2.  Next, examine the data more carefully.  Look at the frequency tables for all three variables and examine the means and standard deviations. Then put all three variables into a bivariate correlation analysis and examine the results. The correlations between year and each of the other two variables suggest a problem with Hovland & Sears' hypothesis and the data supporting it.  In your lab write-up, write out what that problem is.  Run a multiple regression analysis to explore the relevant data.  Report your findings in APA style, and include commentary on how these findings change our interpretation of the correlation from #1.

To report a regression analysis:

• First the “overall equation” as significant or non-significant; R= .xx, F(x ,xxx) = x.xx, p = .xx.
• Then report the Betas and significance for each predictor; Beta = .xx, t = x.xx, p = .xx.

Part II

Access the datafile "moderatorstress.sav" on Blackboard.  File includes the following variables:

• self-reported stress (higher numbers mean more stress)
• score on a stressful life events scale (higher numbers mean more stressful events)
• a categorical variable measuring relationship "status" (0 = does not live with romantic partner; 1 = lives with romantic partner)

3.  The hypothesis for this data set was that as the number of stressful events in a person's life increased, self-reported stress would increase.  However, the researchers also hypothesized that living with one's romantic partner would serve as a protective factor and would sever the link between stressful events and feelings of stress. Use the data provided to test the hypothesis. Report the statistical results in APA style, and comment on whether or not the researchers' hypothesis was supported. Be sure to frame their hypothesis in terms of the vocabulary you learned during our last lecture on types of correlational relationships.

Part III

Access the datafile "regressmediation.sav" from Blackboard.  File includes the following variables:

• sex
• age
• optimism - 6 items; e.g., “In uncertain times I usually expect the best;” 1-5 Likert scale - high = more optimism
• perceived stress - 5 items; e.g., "To what extent have you felt active irritable, nervous in last week;" 1 (very slightly/not at all) – 5 (extremely)
• life satisfaction - 5 items; e.g., “The conditions of my life are excellent.” 1 – 7 Likert - high = more satisfied
• perceived control of internal states - 18 items; a measure of the degree to which people feel they have control of their emotions, thoughts, physical reactions; 1-5 Likert - high = more control
  

4. Although the optimism, stress and control scales have already been created for you, the life satisfaction scale has not.  First check in the internal consistency of the items. Report the alpha and indicate whether or not it is adequate. Assuming it is adequate, create the life satisfaction scale. (None of the 5 items needs to be recoded.)

5. The researchers predicted that stress would "mediate" the relationship between perceived control and life satisfaction.  Explain what this hypothesis means about the relationships amongst these three variables.  Write out the three variables and include arrows as appropriate to illustrate the hypothesis.

6. Run a series of regression analyses (linear and multiple) to test the mediational hypothesis from #5. Run each analysis, and using the below as a model, write a paragraph that could be in a Results section. To do this, you will need to figure out what X, Z, and Y stand for below, and you will need to have all of your regression results available.

Our mediational hypothesis (was/was not) supported. The predictor variable (X) was significantly related to both the proposed mediator (Z; R=.xx, F(df, df) = X.xx, p = .xx) and the outcome variable (Y); R=.xx, F(df, df) = X.xx, p = .xx. Additionally, Z was significantly related to Y; R=.xx, F(df, df) = X.xx, p = .xx . To test for mediation, we conducted a (name analysis) and entered X and Z as predictor variables and Y as the outcome variable. The overall equation was significant; R=.xx, F(df, df) = X.xx, p = .xx . Z’s relationship with Y remained significant even while controlling for X; Beta = .xx; t = x.xx, p = .xx. Most importantly, the relationship between X and Y was weaker in this analysis (Beta = .xx; t = x.xx, p = .xx) compared to the direct relationship (Beta = ). These results suggest (partial/full) mediation (see Figure 1).

7.  Use Word to create a figure illustrating your results. Follow the model below:

Figure 1. Model testing hypothesis that stress mediates the relationship between perceived control and life satisfaction. * p <.05. ** p<.01. ** p<.001.