????
Turn-in: (1) the answers to the questions on the exam (2) SPSS output Please answer the question using BLUE color: 1. 2. For this problem, you will use the satm.sav dataset. These data represent high school seniors’ responses to the SAT “Student Questionnaire” along with student’s SAT mathematics test score. The specific variables you will be using in your analysis are: sexcoded 1 = male, 2 = female sessocioeconomic status hsrankhigh school rank mathgrgrades in high school mathematics courses scigrgrades in high school science courses yrssciyears of science studied in high school yrsmathyears of mathematics studied in high school mthsciscself-concept of mathematical and scientific abilities writescself-concept of writing abilities leadscself-concept of leadership abilities You are to conduct an analysis that answers three questions: (1) are student background (sex, ses, hsrank), high school experiences (mathgr, scigr, yrssci, yrsmath), or self-concept (mthscisc, writesc, leadsc) more important in explaining the variance in SAT mathematics scores?, (2) considering all 10 independent variables, which have a unique relationship to SAT mathematics scores, and (3) after controlling for the other 9 independent variables, is there a significant difference between males and females in SAT mathematics scores? To answer these questions you must conduct a block-entry, hierarchical analysis of these data. With satm as the dependent variable, you should first enter the background characteristics as Model 1. Next, enter the high school experience variables in Model 2, and finally, add the self-concept measures in Model 3. Thus, Model 3 will contain all 10 independent variables. Treat any ordinal variables as continuous. Use the output from these analyses with α = .05 to answer the following questions: a) What proportion of variance in SAT scores is explained by the background characteristics alone? Is this variance explained significantly different from 0? Report the F string (i.e., F(df, df) = #, p = #) you used to determine this answer. (1 point) b) What proportion of variance in SAT scores is explained by the high school experiences over and beyond that explained by background characteristics? Is this variance explained significantly different from 0? Report the F string. (1 point) c) What proportion of variance in SAT scores is explained by the self-concept measures over and beyond that explained by the previous 2 blocks of variables? Is this variance explained significantly different from 0? Report the F string. (1 point) d) Which block of variables explains the greatest proportion of variance in SAT performance? (1 point) e) What is the total variance explained by the set of 10 independent variables? Is this variance explained significantly different from 0? Report the F string. (1 point) f) Does multicollinearity appear to be a problem in the analysis? Why or why not? (2 points) g) In the full model (Model 3), which of the independent variables have a significant unique relationship to SAT performance? (2 points) h) In Model 1, how do you interpret the coefficient for sex (be substantively specific)? (1 point) i) In Model 3, how do you interpret the coefficient for sex (again, be substantively specific)? (1 point) j) Some researchers argue that self-concept is more important than background and experiences in understanding mathematics performance. i. Is this true with respect to explaining variance? Why or why not? (2 points) ii. Is this true with respect to variables influencing SAT performance? Why or why not? (2 points)