*following creates z scores for the ydacl statedp traitdp and rads vars. *specifically adding the /SAVE subcommand to descriptives will create z. *scores for whatever variables are in the command. DESCRIPTIVES VARIABLES=ydacl statedp traitdp rads /SAVE /STATISTICS=MEAN STDDEV MIN MAX. Descriptives Descriptive Statistics ydacl statedp state ion traitdp trait ion rads reynolds adol scale Valid N (listwise) N Minimum Maximum Mean Std. Deviation 80.00.00 6.0775.0797.00.00.890.668 0.00.00.95.658 90.00.97.9897.580 70 *Following just runs descriptives on the resulting variables to check to. * be sure that we have ended up with variables with mean = 0 and sd =. DESCRIPTIVES VARIABLES=zydacl zstatedp ztraitdp zrads /STATISTICS=MEAN STDDEV MIN MAX skew kurt. Descriptives Descriptive Statistics Zydacl Zscore(ydacl) Zstatedp Zscore: state ion Ztraitdp Zscore: trait ion Zrads Zscore: reynolds adol scale Valid N (listwise) N Minimum Maximum Mean Std. Skewness Kurtosis Statistic Statistic Statistic Statistic DStatistic i ti Statistic Std. Error Statistic Std. Error 80 -.559.79.0000000.0000000.896.066.00. -.077.650.0000000.0000000.88.067.96. 0 -.5.70.0000000.0000000.775.068.5.6 90 -.7087.05.0000000.0000000.55.068 -.06.6 70 *Following calculates the mean for the 6 items in question 5, provided that. *at least valid values are present. Page
compute socsup=mean.(q5a to q5f). *Following gives me a look at the resulting variable I asked for percentiles. *for 5 50 and 75 which I will use to construct four approximately equally sized. *groups for my social support variable. EXAMINE VARIABLES=socsup /PERCENTILES(5,50,75) /STATISTICS = descriptives /plot = none /MISSING LISTWISE /NOTOTAL. Explore Case Processing Summary socsup social support average Q5 w at least valid Cases Valid Missing Total N Percent N Percent N Percent 8 9.% 06 7.6% 89 00.0% Descriptives socsup social support average Q5 w at least valid Mean 95% Confidence Interval for Mean Lower Bound Upper Bound Statistic Std. Error.60.0767.5.97 5% Trimmed Mean Median Variance Std. Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis.5.667.00.678.00.00.00.8.65.068 -..7 Page
Percentiles Weighted Average(Definition ) Tukey's Hinges socsup social support average Q5 w at least valid socsup social support average Q5 w at least valid Percentiles 5 50 75.8.667.6667.8.667.6667 *First ran this syntax to create groups. *resulted in groups that were a little too unbalanced. *changed slightly and ran block of commands that follows. Do if socsup le.8. + compute socsupgp =. else if socsup le.66666. + compute socsupgp =. else if socsup le.666666. + compute socsupgp =. else if socsup gt.666666. + compute socsupgp =. end if. *This block created more nearly equal groups. Do if socsup le.8. + compute socsupgp =. else if socsup le.667. + compute socsupgp =. else if socsup le.6667. + compute socsupgp =. else if socsup gt.6667. + compute socsupgp =. end if. *runs frequencies to look at distribution across the groups. freq var = socsupgp. Frequencies Page
Statistics socsupgp social support group N Valid Missing 8 06 socsupgp social support group Valid Missing Total.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Total System Frequency Percent Valid Percent 66 6. 8.5 8.5 79 0..7 50. 58 5.8 7.9 78. 80 0..8 00.0 8 9. 00.0 06 7.6 89 00.0 Cumulative Percent *now label the new variables. Variable labels socsup 'social support average Q5 w at least valid' /socsupgp 'social support group'. Value labels socsupgp 'highest support le.8' 'second highest le.667' 'second lowest le.6667' 'lowest support gt.6667'. *runs the mixed analysis within subjects factor is the three different. * ion measures and the within factor is the four level social. * support group factor we just created. GLM zydacl ztraitdp zstatedp zrads BY socsupgp /WSFACTOR = /METHOD = SSTYPE() /POSTHOC = socsupgp ( SNK QREGW ) /PLOT = PROFILE( socsupgp ) /EMMEANS = TABLES() COMPARE ADJ(SIDAK) /EMMEANS = TABLES(socsupgp) COMPARE ADJ(SIDAK) /EMMEANS = TABLES(socsupgp*) Page
/PRINT = DESCRIPTIVE ETASQ /CRITERIA = ALPHA(.05) /WSDESIGN = /DESIGN = socsupgp. General Linear Model Within-Subjects Factors Dependent Variable Zydacl Ztraitdp Zstatedp Zrads Between-Subjects Factors socsupgp social support group.00.00.00.00 Value Label highest support le. 8 second highest le. 667 second lowest le. 6667 lowest support gt. 6667 N 6 7 5 7 Page 5
Descriptive Statistics Zydacl Zscore(ydacl) Ztraitdp Zscore: trait ion Zstatedp Zscore: state ion Zrads Zscore: reynolds adol scale socsupgp social.00 highest t support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Total.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Total.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Total.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Total Mean Std. Deviation N -.850.786670 6 -.785085.85 7.07.95558 5.5997.700 7 -.007970.995755 59 -.56.70868 6 -.700.805907 7.09868.98966 5.70.0906 7 -.0.99555 59 -.509757.756055 6 -.95.86659 7.067.956096 5.686.06896 7 -.0697.00899 59 -.5956.7887689 6 -.75.8877 7.8760.888905 5.69807.00907 7 -.0087098.997685 59 Page 6
Effect * socsupgp a. Exact statistic Pillai's Trace Wilks' Lambda Hotelling's Trace Roy's Largest Root Pillai's Trace Wilks' Lambda Hotelling's Trace Roy's Largest Root Multivariate Tests c b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept+socsupgp Within Subjects Design: Partial Eta Value F Hypothesis df Error df Sig. Squared.000. a.000 5.000.95.000.000. a.000 5.000.95.000.000. a.000 5.000.95.000.000. a.000 5.000.95.000.00.7 9.000 765.000.000.00.970.95 9.000 09.6.000.00.0. 9.000 755.000.000.00.06 0.9 b.000 55.000.000.05 Mauchly's Test of Sphericity b Epsilon a Within Subjects Effect Mauchly's W Approx. Chi-Square df Sig. Greenhouse -Geisser Huynh-Feldt Lower-bound.70 9.55 5.000.89.8. Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to an identity matrix. a. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table. b. Design: Intercept+socsupgp Within Subjects Design: Page 7
Tests of Within-Subjects Effects Source * socsupgp Error() Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound Type III Sum Partial Eta of Squares df Mean Square F Sig. Squared.6.0.6.9.000.6.56.05.6.890.000.6.67.05.6.89.000.6.000.6.6.688.000.99 9.660 6.6.000.05.99 7.67.08 6.6.000.05.99 7.00.09 6.6.000.05.99.000.980 6.6.000.05 98.959 765.6 98.959 08.70.9 98.959 095.580.7 98.959 55.000.78 Tests of Within-Subjects Contrasts Source * socsupgp Error() Linear Quadratic Cubic Linear Quadratic Cubic Linear Quadratic Cubic Type III Sum Partial Eta of Squares df Mean Square F Sig. Squared.058.058.85.667.000.009.009.08.85.000.059.059.67.606.000 8.85.98 9.6.000.0.79.760.068.07.007.8.7 5.77.00.0 9.559 55. 0.79 55.8 78.608 55. Page 8
Tests of Between-Subjects Effects Transformed Variable: Average Source Intercept socsupgp Error Type III Sum Partial Eta of Squares df Mean Square F Sig. Squared.96.96.97..00 95.6 08.9 5.5.000.0 09.56 55.6 Estimated Marginal Means. Estimates 95% Confidence Interval Mean Std. Error Lower Bound Upper Bound.07.06 -.05.069.0.05 -.05.07.08.06 -.0.068.00.05 -.00.079 Page 9
Pairwise Comparisons (I) (J) Based on estimated marginal means a. Adjustment for multiple comparisons: Sidak. Mean 95% Confidence Interval for Difference Difference a (I-J) Std. Error Sig. a Lower Bound Upper Bound -.006.0.000 -.068.055.000.0.000 -.059.058 -.0.0.997 -.077.05.006.0.000 -.055.068.006.06.999 -.07.09 -.006.06.999 -.07.05.000.0.000 -.058.059 -.006.06.999 -.09.07 -.0.09.990 -.06.09.0.0.997 -.05.077.006.06.999 -.05.07.0.09.990 -.09.06 Pillai's trace Wilks' lambda Hotelling's trace Roy's largest root Multivariate Tests Partial Eta Value F Hypothesis df Error df Sig. Squared.000. a.000 5.000.95.000.000. a.000 5.000.95.000.000. a.000 5.000.95.000.000. a.000 5.000.95.000 Each F tests the multivariate effect of. These tests are based on the linearly independent pairwise comparisons among the estimated marginal means. a. Exact statistic. social support group Page 0
Estimates social support group.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 95% Confidence Interval Mean Std. Error Lower Bound Upper Bound -.5.0 -.59 -. -.7.08 -.66 -.079.098.0.06.8.676.08.58.769 Page
Pairwise Comparisons (I) social support group.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 (J) social support group.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667.00 highest support le.8.00 second lowest le.6667.00 lowest support gt.6667.00 highest support le.8.00 second highest le.667.00 lowest support gt.6667.00 highest support le.8.00 second highest le.667.00 second lowest le.6667 Based on estimated marginal means *. The mean difference is significant at the.05 level. a. Adjustment for multiple comparisons: Sidak. Mean 95% Confidence Interval for Difference Difference a (I-J) Std. Error Sig. a Lower Bound Upper Bound -.*.06.000 -.507 -.75 -.6*.059.000 -.767 -.57 -.89*.06.000 -.55 -.0.*.06.000.75.507 -.7*.06.000 -.8 -.0 -.88*.067.000 -.06 -.67.6*.059.000.57.767.7*.06.000.0.8 -.577*.06.000 -.7 -.0.89*.06.000.0.55.88*.067.000.67.06.577*.06.000.0.7 Page
Univariate Tests Sum of Squares df Mean Square F Sig. Partial Eta Squared Contrast.6 77. 5.5.000.0 Error 77.8 55.66 The F tests the effect of social support group. This test is based on the linearly independent pairwise comparisons among the estimated marginal means.. social support group * social support group.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 95% Confidence Interval Mean Std. Error Lower Bound Upper Bound -.85.09 -.8 -.90 -.56.06 -.655 -.7 -.509.08 -.60 -.6 -.595.06 -.686 -.50 -.79.056 -.89 -.068 -.7.05 -.79 -.069 -..055 -. -.07 -..05 -.9 -.08.0.050 -.06...07.09.0..08.06.05.8.07.06..599.056.89.70.7.05.67.87.68.055.575.79.698.05.59.80 Post Hoc Tests socsupgp social support group Homogeneous Subsets Page
MEASURE_ Student-Newman-Keuls a,b Ryan-Einot-Gabriel- Welsch Range c,b social support group.00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Sig..00 highest support le.8.00 second highest le.667.00 second lowest le.6667.00 lowest support gt.6667 Sig. Means for groups in homogeneous subsets are displayed. Based on Type III Sum of Squares The error term is Mean Square(Error) =.66. a. Uses Harmonic Mean Sample Size = 08.9. b. Alpha =.05. Subset N 6 -.56 7 -.765 5.09855 7.67587 6 -.56.000.000.000.000 7 -.765 5.09855 7.67587.000.000.000.000 c. Critical values are not monotonic for these data. Substitutions have been made to ensure monotonicity. Type I error is therefore smaller. Profile Plots Page
Estimated Marginal Means of MEASURE_ 0.80000 0.60000 Estimated Marginal Means 0.0000 0.0000 0.00000-0.0000-0.0000-0.60000 highest support le.8 second highest le.667 second lowest le.6667 social support group lowest support gt.6667 Page 5