* For further details see http://www.jutze.com/research/2013-schult-et-al-differential-prediction/ * Project: Genderfairness * Purpose of this syntax: data management of the 1st data set for the "Reanalysen" publication * * written by Johannes Schult, johannes.schult@uni-konstanz.de * last updated 2012-03-02 * * Prepare data set courtesy of Benedikt Hell for re-analysis * * Open original data set (Masterdatei Kriteriendaten Hohenheim.sav). GET FILE='myhome\Masterdatei Kriteriendaten Hohenheim.sav'. DATASET NAME master WINDOW=FRONT. * Calculate mean college GPA. COMPUTE fgpaO = MEAN.1(n1_1011,n1_1021,n1_1022,n1_1031,n1_1033,n1_1042,n1_5010,n1_5020,n1_5050,n1_5060,n1_9951,n1_9952). EXECUTE. * Delete all cases with missing data on the variables of interest. FILTER OFF. USE ALL. SELECT IF (sex < 9999 AND std_gang < 9999 AND abi_note < 9999 AND fgpaO < 9999 AND v_stich < 9999 AND n_stich < 9999). EXECUTE. * Save standardized predictor variables. DESCRIPTIVES VARIABLES= abi_note v_stich n_stich /SAVE /STATISTICS=MEAN STDDEV MIN MAX. * Recode sex. RENAME VARIABLES (sex = female). IF(female = 1) female = 0. IF(female = 2) female = 1. VARIABLE LABELS female "sex (female)". VALUE LABELS female 0 "male" 1 "female" 2 "". EXECUTE. * Save standardized criterion variable. * GPAs are standardized within each field of study to control for differences in grading. SORT CASES BY std_gang. SPLIT FILE SEPARATE BY std_gang. DESCRIPTIVES VARIABLES=fgpaO /SAVE /STATISTICS=MEAN STDDEV MIN MAX. SPLIT FILE OFF. * Reverse GPAs (in Germany, low grades are good; after reversing, they are in line with American GPAs, where high grades are good). * Then rename variables (so that variable names are similar across samples) and label them in English. COMPUTE hsgpa = Zabi_note * (-1). COMPUTE fgpa = ZfgpaO * (-1). *MISSING VALUES n1_1011 n2_1011 n1_1021 n2_1021 n1_1022 n2_1022 n1_1031 n2_1031 n1_1033 n2_1033 n1_1042 n2_1042 n1_5010 n2_5010 n1_5020 n2_5020 n1_5050 n2_5050 n1_5060 n2_5060 n1_9951 n2_9951 n1_9952 n2_9952(5 TO 999). *COMPUTE fgpaO = MEAN.1(n1_1011,n2_1011,n1_1021,n2_1021,n1_1022,n2_1022,n1_1031,n2_1031,n1_1033,n2_1033,n1_1042, n2_1042,n1_5010,n2_5010,n1_5020,n2_5020,n1_5050,n2_5050,n1_5060,n2_5060,n1_9951,n2_9951,n1_9952,n2_9952). RENAME VARIABLES (std_gang Zv_stich Zn_stich abi_note = fos test_verbal test_numeric abi). COMPUTE vn = v_stich + n_stich. DESCRIPTIVES VARIABLES= vn /SAVE /STATISTICS=MEAN STDDEV MIN MAX. COMPUTE test_vn = Zvn. VARIABLE LABELS hsgpa "high school grade point average (Abinote, standardized and reversed)" fgpa "frechman grade point average (college grades, standardized and reversed)" fos "field of study" test_verbal "BIS verbal score standardized" test_numeric "BIS numeric score standardized" test_vn "BIS verbal+numeric score (standardized after addition)" abi "Abinote" fgpaO "college gpa (German original scoring)". EXECUTE. * Save data set under a new name (i.e., the name of this syntax file minus cr). SAVE OUTFILE='myhome\set1V1ask.sav' /KEEP abi fgpaO female fos hsgpa fgpa test_verbal test_numeric test_vn /COMPRESSED. * Close window. DATASET CLOSE master.