Uitvoer van analyses (SPSS 6) voor het aalfeedback en Oriëntatie voorbeeld in hoofdstuk 7 (Herhaalde metingen) > ** Berekening van lineaire en kwadratische trendvariabele. Compute ylin = -.77678 * y + * y +.77678 *. Compute ykwad =.48489 * y - *.48489 * y +.48489 *. ** Basisgegevens. MEAS TABLES=y y ylin ykwad BY orient /CELLS MEA COUT STDDEV MI MAX. s Case Processing Summary Cases Included Excluded Total Percent Percent Percent y * orient,%,%,% y * orient,%,%,% * orient,%,%,% ylin * orient,%,%,% ykwad * orient,%,%,% Report orient Total Std. Deviation Minimum Maximum Std. Deviation Minimum Maximum Std. Deviation Minimum Maximum y,3859 3,86 6,5 7,5,3 3,6559 3,37 9,55,7 3,673 3,37 9,55 y 3,89 4,4634 4,8,4,877 4,36666 4,,,9548 4,443 4,8,4 8,4684 4,7546 9,74 8,58 8,584 3,94873,8 5,8 3,56 6,44,8 8,58 ylin 5,8,973 -,89, -,449,859-5,58,8,836 4,37643-5,58, ykwad,95,644-3,8 6,4 -,454,6743-4,77,98 -,745,4935-4,77 6,4 Page
** Zijn er verschillen tussen de drie metingen in de?. ** Zijn deze verschillen te beschrijven ** als een lineaire en/of kwadratische trend?. GLM y y /WSACTOR = 3 Polynomial /METHOD = SSTYPE(3) /PLOT = PROILE( ) /EMMEAS = TABLES(OVERALL) /EMMEAS = TABLES() /PRIT = DESCRIPTIVE ETASQ PARAMETER /CRITERIA = ALPHA(.5) /WSDESIG =. General Linear Model Measure:MEASURE_ y y 3 Dependent Variable Within-Subjects actors y y Descriptive Statistics,7,9548 Std. Deviation 3,673 4,443 3,56 6,44 Multivariate Tests b Effect Pillai's Trace Wilks' Lambda Hotelling's Trace Roy's Largest Root Value,9,87,49,49 4,7 a 4,7 a 4,7 a 4,7 a Hypothesis Error 5,8 E 5,8 E 5,8 E 5,8 E,8,8,8,8,9,9,9,9 a. Exact statistic b. Design: Intercept Within Subjects Design: Page
Mauchly's Test of Sphericity b Measure:MEASURE_ Epsilon a Within Approx. Greenho Subjects Mauchly's Chi- use- Huynh- Lower- Effect W Square Geisser eldt bound,66 7,77,78,74,5 Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to a. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of b. Design: Intercept Within Subjects Design: o an identity matrix. of Within-Subjects Effects table. Tests of Within-Subjects Effects Measure:MEASURE_ Sphericity Assumed Greenhouse-Geisser Huynh-eldt Lower-bound Error() Sphericity Assumed Greenhouse-Geisser Huynh-eldt Lower-bound Type III 3,379 3,379 3,379 3,379 496,876 496,876 496,876 496,876,455,483 8 85,874 87,488 59 Square 5,69 7,7 69,77 3,379,685 7,43 7,9 5,37 4,75 4,75 4,75 4,75,9,3,3,48,65,65,65,65 Tests of Within-Subjects Contrasts Measure:MEASURE_ Sour ce Linear Quadratic Type III 98,859 4,5 Square 98,859 4,5 5,6,77,7,397,8, Page 3
Tests of Within-Subjects Contrasts Measure:MEASURE_ Error() Linear Type III,E3 59 Square 9,53 Quadratic 366,84 59 6,8 Tests of Between-Subjects Effects Measure:MEASURE_ Transformed Variable:Average Intercept Type III Sum of 97,546 Square 97,546 7,769,9 Error 83,964 59 47,999 Dependent Variable y y Parameter Intercept Intercept B,7,955 Parameter Estimates Std. Error,49,573 t 8,64,595 95% Confidence Interval Lower Bound,88 Upper Bound 4, Intercept 3,56,853 5,865,8 5,3,8 Estimated Marginal s,893,59,933,896. Measure:MEASURE_ 95% Confidence Interval Std. Error Lower Bound Upper Bound,7,49,893,59,955,573,88 4, 3 3,56,853,8 5,3 Profile Plots Page 4
Estimated Marginal s of MEASURE_ 4, Estimated Marginal s 3,5 3,,5,,5 3 ** Correlaties tussen de drie metingen en de twee trendvariabelen. CORRELATIOS /VARIABLES=y y ylin ykwad /PRIT=TWOTAIL OSIG /MISSIG=PAIRWISE. Correlations y y y Pearson Correlation,65 ** (-tailed) Correlations,367 **,4 ylin ykwad -, -,,359,8 y Pearson Correlation (-tailed),65 **,63 **,34 **,8 -,435 **, **. Correlation is significant at the. level (-tailed). Page 5
Correlations ylin ykwad Pearson Correlation (-tailed) Pearson Correlation (-tailed) Pearson Correlation (-tailed) y,367 **,4 -,,359 -,,8 y,63 **,34 **,8 -,435 **,,879 **,354 **,6 ylin,879 **,393 **, ykwad,354 **,6,393 **, **. Correlation is significant at the. level (-tailed). ** Zijn er verschillen tussen de drie metingen in de ** Zijn deze verschillen te beschrijven ** als een lineaire en/of kwadratische trend ** en lopen deze verschillen uiteen voor de beide orientatiegroepen?. GLM y y BY orient /WSACTOR = 3 Polynomial /contrast (orient) =simple /METHOD = SSTYPE(3) /PLOT = PROILE( *orient ) /EMMEAS = TABLES(OVERALL) /EMMEAS = TABLES(orient) /EMMEAS = TABLES() /EMMEAS = TABLES(orient*) /PRIT = DESCRIPTIVE ETASQ PARAMETER TEST(MMATRIX) HOMOGEEITY /CRITERIA = ALPHA(.5) /WSDESIG = /DESIG = orient. General Linear Model Measure:MEASURE_ y y 3 Dependent Variable Within-Subjects actors Page 6
Between-Subjects actors orient Descriptive Statistics y y ori Total Total Total,3859,3,7 3,89,877,9548 8,4684 8,584 3,56 Std. Deviation 3,86 3,6559 3,673 4,4634 4,36666 4,443 4,7546 3,94873 6,44 Box's M,447,958 6 4373,3,68 Effect Box's Test of Equality of Covariance Matrices a Tests the null hypothesis that the observed covariance matrices of the dependent variables are equal across groups. a. Design: Intercept + orient Within Subjects Design: Pillai's Trace Wilks' Lambda Hotelling's Trace Roy's Largest Root Value,343 9,784 a 9,784 a Hypothesis Error 57 * orient Pillai's Trace,759 8,99 57,759 a. Exact statistic b. Design: Intercept + orient Within Subjects Design:,56,744,343 9,784 a 9,784 a Multivariate Tests b 57 57 57,56,56,56,56 Page 7
Multivariate Tests b Effect * orient Wilks' Lambda Hotelling's Trace Roy's Largest Root Value,4 3,55 3,55 8,99 8,99 8,99 Hypothesis Error 57 57 57,759,759,759 a. Exact statistic b. Design: Intercept + orient Within Subjects Design: Mauchly's Test of Sphericity b Measure:MEASURE_ Epsilon a Within Approx. Greenh Subjects Mauchly's Chi- ouse- Huynh- Lower- Effect W Square Geisser eldt bound,998,3,95,998,5 Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to a. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of b. Design: Intercept + orient Within Subjects Design: o an identity matrix. of Within-Subjects Effects table. Tests of Within-Subjects Effects Measure:MEASURE_ Sphericity Assumed Greenhouse-Geisser Huynh-eldt Lower-bound * orient Sphericity Assumed Greenhouse-Geisser Type III 3,379 3,379 3,379 3,379 95,4 95,4,996,996 Square 5,69 5,783 5,69 3,379 457,76 458,59,3,3,3,3 9,3 9,3,,5,5,5,5,6,6 Page 8
Tests of Within-Subjects Effects Measure:MEASURE_ * orient Huynh-eldt Lower-bound Error() Sphericity Assumed Greenhouse-Geisser Huynh-eldt Lower-bound Type III 95,4 95,4 58,463 58,463 58,463 58,463 6 5,79 6 58 Square 457,76 95,4 5,3 5, 5,3,5 9,3 9,3,6,6 Tests of Within-Subjects Contrasts Measure:MEASURE_ Linear Quadratic * orient Linear Quadratic Error() Linear Quadratic Type III 98,859 4,5 83, 83, 97,733 83,7 58 58 Square 98,859 4,5 83, 83, 5,33 4,89 9,58,94 6,37 6,989,34,49,6,737,7 y y,53,89,766,53 58, 58 58 Levene's Test of Equality of Error Variances a,88 Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept + orient Within Subjects Design: Tests of Between-Subjects Effects Measure:MEASURE_ Transformed Variable:Average Intercept Type III 97,546 Square,97E4 758,,99 orient,533,533 5,64, Error,43 58 38,456 Page 9
Depe nden t Varia ble y y Paramet er Intercept [orient=] [orient=] Intercept [orient=] [orient=] Intercept [orient=] [orient=] B,36,88 8,584 Std. Error,58,8,798 t,748 5,76,758 Parameter Estimates 95% Confidence Interval Lower Bound,875,483 6,987 Upper Bound 3,97 3,693,8,88 -,65,8 -,79,43 -,9,99, a......,797,734,34,59,3 -,536 4,4,39 a......,666 9,884,8 8,7 7,66,43,57 a...... a. This parameter is set to zero because it is redundant. Transformation Coefficients (M Matrix) Measure:MEASURE_ Transformed Variable:AVERAGE y,577 y,577,577 Measure:MEASURE_ Depe nd y y Linear -,77 Quadratic,48 -,86,77,48 a. The contrasts for the within subjects factors are: : Polynomial contrast Custom Hypothesis Tests Average a Transformation Coefficients (M Matrix) Measure:MEASURE_ Transformed Variable:AVERAGE y,333 y,333,333 Page
Contrast Results (K Matrix) orient Simple Contrast a Level vs. Level Contrast Estimate Hypothesized Value Difference (Estimate - Hypothesized) Std. Error a. Reference category = 95% Confidence Interval for Difference Lower Bound Upper Bound Averaged MEASURE_ 3,656 3,656,94,86 5,57 Test Results Measure:MEASURE_ Transformed Variable:AVERAGE Contrast,5 Square,5 5,64, Error 743,477 58,89 Estimated Marginal s. Grand Transformation Coefficients (M Matrix) Depe nd y y Measure MEASURE_,333,333,333 Estimates Measure:MEASURE_ Std. Error,73,46 95% Confidence Interval Lower Bound Upper Bound,85 3,656. orient Transformation Coefficients (M Matrix) Measure Depe nd MEASURE_ y,333 y,333 Page
Transformation Coefficients (M Matrix) Measure Depe nd MEASURE_,333 Estimates Measure:MEASURE_ 95% Confidence Interval orien t Std. Error Lower Bound Upper Bound 4,559,654 3,5 5,867,93,654 9,594, 3. Measure:MEASURE_ Depe nd y y Measure:MEASURE_,7,955 Std. Error,4,567 3 Lower Bound Transformation Coefficients (M Matrix) 95% Confidence Interval,89,8 Upper Bound Estimates,53 4,9 3 3,56,564,397 4,656 4. orient * Measure:MEASURE_ Depe nd y y Measure:MEASURE_ 3 Transformation Coefficients (M Matrix) Estimates 95% Confidence Interval orien t Std. Error Lower Bound Upper Bound,386,58,5,547 Page
Measure:MEASURE_ orien t 3,8 3 8,468,36,88 3 8,584 Profile Plots Std. Error,8,798,58,8,798 Estimates 95% Confidence Interval Lower Bound Upper Bound,7 5,47 6,87,66,875 3,97,483 3,693 6,987,8 Estimated Marginal s of MEASURE_, orient 8, Estimated Marginal s 6, 4,,, 8, 3 ** Correlaties tussen de drie metingen en de twee trendvariabelen, ** apart voor de beide orientatiegroepen. SORT CASES BY orient. SPLIT ILE SEPARATE BY orient. CORRELATIOS Page 3
/VARIABLES=y y ylin ykwad /PRIT=TWOTAIL OSIG /MISSIG=PAIRWISE. Correlations orient = y Pearson Correlation (-tailed) y Pearson Correlation (-tailed) Pearson Correlation (-tailed) ylin Pearson Correlation (-tailed) ykwad Pearson Correlation (-tailed) Correlations a y y,57 **,63 **, **. Correlation is significant at the. level (-tailed). *. Correlation is significant at the.5 level (-tailed). a. orient = orient = ylin ykwad -,4,6,9,394,57 **,663 **,377 * -,6 **,,4,63 **,663 **,76 **,3,49 -,4,377 *,76 **,34,9,4,8,6 -,6 **,3,34,394,49,8 Correlations a y y ylin ykwad y Pearson Correlation,799 **,75 ** -,6 -,83 (-tailed),543,33 y Pearson Correlation,799 **,86 **, -,663 ** (-tailed),4 Pearson Correlation,75 **,86 **,57 ** -,76 **. Correlation is significant at the. level (-tailed). a. orient = Page 4
Correlations a ylin ykwad (-tailed) Pearson Correlation (-tailed) Pearson Correlation (-tailed) y -,6,543 -,83,33 y,,4 -,663 **,57 **, -,76,4 ylin, -,87,3 ykwad,4 -,87,3 **. Correlation is significant at the. level (-tailed). a. orient = > SPLIT ILE O. Page 5