Before applying inferential tools to the data, you should examine your data Visually to get a picture of the associations between SES and IQ. (a) (13} Use the Line Chart feature in the Graphs menu to obtain the plot of mean IQ versus SES of biological parents with the lines representing two levels of SES of adoptive parents. Paste the plot into your report. What do you conclude about the association of SES of biological parents and SES of adoptive parents with IQ? How strong is the interaction between adoptive parent category and biological parent category? Which SES has a stronger relationship with IQ? Explain referring to the chart. Obtain a clustered boxplot of the mean IQ with SES of biological parents on the category axis and clusters dened by SES of adoptive parents. Paste the plot into your report. What do you conclude about the association of each of the two factors (SES of biological parents and SES of adoptive parents) with IQ? Are your conclusions consistent with the conclusions reached in part (a)? Now you will obtain some numerical summaries of the data. (a) 0)) Use the Means procedure in the Compare Means menu (Statistics) to obtain the table of the marginal means of IQ for all combinations of SES (biological and adoptive parents). The two factors, SES of biological and SES of adoptive parents should be entered as two layers in the procedure. To obtain the summaries you will need to recode the two categories High, Low into numerical values I, 0. The recoding may also be useful in Question 6. Paste the output into your report. Use the table obtained in part (a) to estimate the difference in the mean IQ for children of high SES biological parents and children of low SES biological parents regardless of the adoptive parents' SES. Moreover, estimate the difference in the mean IQ for children of high SES adoptive parents and children of low SES adoptive parents regardless of the biological parents' SES. Which SES has a stronger relationship with IQ? Do the summaries confirm the ndings in Question 2a? Explain briey. In this part you will use the GLM (General Linear Model) in SPSS to make inferences about the association between socioeconomic status and IQ. (a) 0)) Use the Univariaie Procedure in the General Linear Model menu to first test for an interaction between SES of adoptive and biological parents. Using the result from this test, appropriately test the association between the mean IQ of the child and SES of his biological or adoptive parents. For each test, state the null hypothesis to be tested, report the p-value of the appropriate F-tests, and state your conclusions about the signicance of the associations at the 5% level. Paste the related output into your report. Use the value of type III sum of squares for the interaction term in the output obtained in part (a) and hand calculations to demonstrate how to obtain the value of the test statistic F in this case. What is the distribution of the test statistic under the null hypothesis? The results of the analysis in Question 4 are valid only if the appropriate assumptions are valid. The assumptions are that the data come from a normal population and all group variances are the same. (a) (b) Is there any evidence that the assumption of equal variance might be violated? Refer to the side-by-side boxplots obtained in part (b) of Question 2) to verify the assumption. Moreover, confirm your ndings by referring to the numerical values of the standard deviations provided in the output in Question 3 and the output of the Levene's test. Is the normality assumption reasonable? Comment referring to the boxplots
