SPSS 11.0 Example of a 2 x 2 Between-Subjects Factorial BROWNE PSYCH

Document Sample
SPSS 11.0 Example of a 2 x 2 Between-Subjects Factorial BROWNE PSYCH Powered By Docstoc
					             SPSS 11.0 Example of a 2 x 2 Between-Subjects Factorial              BROWNE            PSYCH 3500

1.   Define the independent (IV) and dependent (DV) variables. In the Variable view field be sure to label all variables and also
     put in all the appropriate value labels too. Ex: Type in sex as variable label then put 1 = males, 2 = females for the value
     labels. Also make sure to change the measure to Nominal for the IV’s.




2.   Be sure to type in the VARIABLE NAME of the DV (recall) and specify the measurement level as SCALE.

3.   After all variables are defined, enter the values for both IVs and the DV.
4.   To run the factorial ANOVA, select ANALYZE, GENERAL LINEAR MODEL, and UNIVARIATE.




5.   At the Univariate window, highlight and move the IVs and DVs to the appropriate boxes. Click on OPTIONS.




                                                                   Independent variables
                                                                   (factors) go here
6.      Select DESCRIPTIVE STATISTICS, EFFECT SIZE,HOMOGENEITY OF VARIANCE. Click CONTINUE, then OK.




7. Next Click on PLOTS to get plots of the means. Move one of the IVs to the horizontal box and the other to the separate lines box.
Then click continie
8. Next, click OK and the ANOVA will run.
2 x 2 factorial ANOVA. Analysis, graphs, and results.

Univariate Analysis of Variance
            Between-Subjects Factors

                          Value Label         N
  interview     1.00      no interview              20
  type          2.00      structured
                                                    20
                          interview
  sex           1.00      males                     20                                                   Main effect Means and SD’s
                2.00      females                   20                                                   for Group (no interview &
                                                                                                         structured interview)
                              Descriptive Statistics

  Dependent Variable: words recalled
  interview type           sex            Mean           Std. Deviation         N
  no interview             males          32.6000              1.26491              10
                           females        26.7000              3.74314              10
                           Total          29.6500              4.06882              20
  structured interview     males          30.5000              3.92287              10
                           females        30.4000              2.79682              10                           Main effect Means and
                           Total          30.4500              3.31623              20                           SD’s for Sex
  Total                    males          31.5500              3.03445              20
                           females        28.5500              3.73427              20
                           Total          30.0500              3.68608              40

                                              a
   Levene's Test of Equality of Error Variances

  Dependent Variable: words recalled
                                                                                         Tests to see if our cell variances are equal. We want
        F          df1            df2              Sig.                                  non significance here. If non significant then we can
        2.607             3              36          .067                                say that the variances do not differ significantly, they
  Tests the null hypothesis that the error variance of the                               are homogenous.
  dependent variable is equal across groups.
    a. Design: Intercept+GROUP+SEX+GROUP * SEX
                                                                                                                        Main Effect Analyses
                                        Tests of Between-Subjects Effects

  Dependent Variable: words recalled
                         Type III Sum                                                                     Partial Eta
  Source                  of Squares          df          Mean Square         F               Sig.         Squared
  Corrected Model             180.500a              3          60.167         6.199             .002             .341
  Intercept                36120.100                1       36120.100      3721.590             .000             .990
                                                                                                                             Significant Main effect
  GROUP                         6.400               1           6.400          .659             .422             .018
  SEX                          90.000               1          90.000         9.273             .004             .205
  GROUP * SEX                  84.100               1          84.100         8.665             .006             .194
  Error                       349.400              36           9.706
  Total                    36650.000               40
  Corrected Total             529.900              39
     a. R Squared = .341 (Adjusted R Squared = .286)
                                                                                                                   Significant Interaction

                                                         Interaction Analysis
Profile Plots
                                   Estimated Marginal Means of words recalled
                              33


                              32
                                                                                                          Here’s a quick graph of the
                                                                                                          interaction to help you interpret it.
                              31
   Estimated Marginal Means




                              30


                              29


                              28
                                                                                 sex

                              27                                                       males

                              26                                                       females
                       no interview                                   structured interview


                                   interview type



                                                                                             Results

                                   An alpha level of .05 was used for all statistical test and r was calculated as the effect size (Rosenthal,

1991).

                                   A 2 x 2 (Group x Sex) between-subjects factorial ANOVA was conducted on number of items recalled.

There was a significant main effect for sex, F(1, 36) = 9.27, p < .05 (r = .45). Males (M = 31.55, SD =3.03)

recalled more words overall than females (M = 28.55, SD = 3.73). There was not a significant main effect for

group, F(1, 36) = .659, p > .05. Those participants who did not have an interview (M = 29.65, SD = .4.07) did

not recall significantly more words than those participants in the structured interview group (M = 30.45, SD =

3.31). There was also a significant interaction between sex and group, F(1, 36) = 8.66, p < .05 (r = .44). Males

recalled more items in the No interview condition (M = 32.6, SD = 1.26) than the structured interview condition

(M = 30.50, SD = 3.93), whereas the females recalled more items in the structured interview condition (M =

30.40, SD = 2.79) than in the no interview condition (M = 26.70, SD = 3.74).