Overview
For 1a, highlight or type your answer in addition to including the output. For 1b-1g, your output does not look correct. Did you set SPSS to analyze all cases? SPSS: Data-Split File-check Analyze all cases, do not create groups-OK 1b: SPSS: Analyze-Classify-Discriminant-Grouping Variable: X4, Define Range: Minimum 0 Maximum 1, continue, Independents: X6-X18, Selection Variable: Split130, Value = 0, continue, check use stepwise method-Statistics check Means, Univariate ANOVAs, Box’s M, Fisher’s, Unstandardized, continue-Method check Mahalanobis distance, in Display check Summary of steps and F for pairwise distances, continue-Classify for Prior Probabilities check compute from group sizes, for Display check Casewise results, Summary table, and Leave-one-out classification, continue-OK For 1c, you want to include the group statistics and tests of equality of group means 1c-1g the output is not correct. 2a-2c look good. For 2d and 2e, you need to split the file to compare groups SPSS: Data-Split File-check Compare groups with Groups based on Split130-OK 2d: Analyze-Classify-ROC Curve-Test Variable: PRE_1, State Variable, X4, Value of State Variable: 1, under Display check ROC Curve and With diagonal reference line-OK 2e: SPSS: Graphs-Chartbuilder-Scatter/Dot-under Gallery click Scatter Plot, move Normalized residual to Y-axis and ID to X-axis-OK Double-click on data, in Elements select Show Data Labels, Data Value Labels=ID 2f looks good. For 2g, the proportional chance value = (81/200)^2 + (119/200)^2 = 51.81%. With the 25% adjustment you get 64.76%.