EXERCISES

1. Meyer and Schvaneveldt (1971) performed one of the first lexical decision experiments
in the history of psychology. In their experiment, subjects were asked to decide as quick as
possible in each trial whether two strings of letters (simultaneously presented on a screen)
were existing words or not. Three types of string pairs (3 conditions) were shown equally
often: (1) pairs of related existing words, (2) pairs of unrelated existing words, and (3) pairs
of which only one of the strings was an existing word. Per trial the reaction time in milli-
seconds (1 second is 1000 milli-seconds) was recorded. The dependent variable was the
mean reaction time over trials per condition. The expectation of the researchers was that
the mean reaction time over subjects was the shortest in the related words condition and
the longest in the only one word condition. The data of six subjects are given in the
following table.

Conditions
1 2 3
1 745 764 774
2 777 786 788
3 734 733 763
Subjects
4 779 801 797
5 756 786 785
6 721 732 740

Enter the data of the six subjects in SPSS.

a. Make a table containing means and standard deviations per condition.
Descriptive Statistics
Std.
Mean Deviation N
related 752.0000 23.25511 6
unrelated 767.0000 29.21643 6
strong 774.5000 20.59854 6

b. Perform a repeated-measures analysis.

Multivariate Testsa
Hypothesis Partial Eta
Effect Value F df Error df Sig. Squared
b
pair_typ Pillai's Trace .919 22.713 2.000 4.000 .007 .919
e Wilks' Lambda .081 22.713 b
2.000 4.000 .007 .919
b
Hotelling's 11.357 22.713 2.000 4.000 .007 .919
Trace
Roy's Largest 11.357 22.713b 2.000 4.000 .007 .919
Root
 c. Decide on the basis of Mauchly’s test of sphericity which output should be used to test
the expectation of the researchers. What should be done otherwise?

Mauchly's Test of Sphericitya

Measure: MEASURE_1
Within Epsilonb
Subjects Mauchly's Approx. Chi- Greenhouse-
Effect W Square df Sig. Geisser Huynh-Feldt Lower-bound
pair_type .758 1.108 2 .575 .805 1.000 .500

Is not significant
Sphericity cannot be rejected p>0.05

Tests of Within-Subjects Effects

Measure: MEASURE_1
Type III Sum Mean Partial Eta
Source of Squares df Square F Sig. Squared
pair_type Sphericity Assumed 1575.000 2 787.500 14.432 .001 .743
Greenhouse-Geisser 1575.000 1.610 978.061 14.432 .003 .743
Huynh-Feldt 1575.000 2.000 787.500 14.432 .001 .743
Lower-bound 1575.000 1.000 1575.000 14.432 .013 .743
Error(pair_type) Sphericity Assumed 545.667 10 54.567
Greenhouse-Geisser 545.667 8.052 67.771
Huynh-Feldt 545.667 10.000 54.567
Lower-bound 545.667 5.000 109.133

d. Formulate the null hypothesis of the repeated-measures analysis.

The null hypothesis of no interaction can be rejected

e. What can be concluded (significance and relevance)? Don’t forget to perform post hoc
tests.
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Type III Sum of
Source Squares df Mean Square F Sig. Partial Eta Squared
Intercept 10520284.500 1 10520284.500 6153.773 <.001 .999
Error 8547.833 5 1709.567
2. A researcher is interested in comparing the speed with which words differing in emotional
value are learned. Each of 8 students studied the same list of 60 different words. Twenty
words were judged to be of negative emotional value, 20 neutral, and 20 positive. Within
each of these categories, 5 words were presented once, 5 were presented twice, 5 three
times, and 5 four times. The list of words was presented at the rate of two seconds per
word, and after the study period, the students were given 5 minutes to recall the words. For
each student, the number of words from each of the 12 treatment combinations that were
recalled on the test, are given below.

Within-factor 1 1 2 3
Within-factor 2 1 2 3 4 1 2 3 4 1 2 3 4
1 1 1 2 2 1 2 2 3 1 2 2 3
2 2 2 3 4 2 3 4 5 3 3 3 5
3 2 3 3 3 3 4 3 4 2 3 3 4
4 0 1 0 2 1 2 2 3 2 2 2 3
Student
5 1 1 2 2 1 1 2 3 1 1 2 2
6 1 2 3 3 2 3 3 5 2 3 4 4
7 1 1 2 3 2 1 2 3 1 1 2 3
8 1 1 2 3 1 2 2 3 2 2 3 4

Enter the data of the eight subjects in SPSS and save the file under an appropriate name (note,
the format of the data-file is 8 rows and 12 columns).
a. Give new names to the two within-subject factors.
b. Formulate the null hypotheses.
Null hypotheses:
c. Perform a repeated measures analysis.
d. Decide on the basis of Mauchly’s tests of sphericity which output should be used. What
should be done otherwise?

Mauchly's Test of Sphericitya

Measure: MEASURE_1
Epsilonb
Approx. Greenhouse-
Within Subjects Effect Mauchly's W Chi-Square df Sig. Geisser Huynh-Feldt Lower-bound
emotional_Value .750 1.726 2 .422 .800 1.000 .500
repetiton .696 2.075 5 .841 .790 1.000 .333
emotional_Value * .029 16.857 20 .749 .591 1.000 .167
repetiton

Tests of Within-Subjects Contrasts

Measure: MEASURE_1
Type III Sum Partial Eta
Source emotional_Value repetiton of Squares df Mean Square F Sig. Squared
emotional_Value Linear 6.250 1 6.250 10.937 .013 .610
Quadratic 2.083 1 2.083 10.938 .013 .610
Error(emotional_Va Linear 4.000 7 .571
lue) Quadratic 1.333 7 .190
repetiton Linear 40.833 1 40.833 131.923 <.001 .950
Quadrati 1.042 1 1.042 3.723 .095 .347
c
Cubic .208 1 .208 .593 .466 .078
Error(repetiton) Linear 2.167 7 .310
Quadrati 1.958 7 .280
c
Cubic 2.458 7 .351
emotional_Value * Linear Linear .012 1 .012 .200 .668 .028
repetiton Quadrati .063 1 .063 2.333 .170 .250
c
Cubic .050 1 .050 .350 .573 .048
Quadratic Linear .104 1 .104 .504 .501 .067
Quadrati .021 1 .021 .084 .780 .012
c
Cubic .417 1 .417 3.365 .109 .325
Error(emotional_Va Linear Linear .437 7 .062
lue*repetiton) Quadrati .187 7 .027
c
Cubic 1.000 7 .143
Quadratic Linear 1.446 7 .207
Quadrati 1.729 7 .247
c
Cubic .867 7 .124

Pairwise Comparisons
Measure: MEASURE_1
Mean 95% Confidence Interval for
emotional (I) (J) Difference Std. Differenceb
_Value repetiton repetiton (I-J) Error Sig.b Lower Bound Upper Bound
1 1 2 -.375 .183 .478 -1.040 .290
3 -1.000* .189 .007 -1.687 -.313
*
4 -1.625 .183 <.001 -2.290 -.960
2 1 .375 .183 .478 -.290 1.040
3 -.625 .263 .295 -1.581 .331
4 -1.250* .250 .009 -2.159 -.341
*
3 1 1.000 .189 .007 .313 1.687
2 .625 .263 .295 -.331 1.581
4 -.625 .263 .295 -1.581 .331
4 1 1.625* .183 <.001 .960 2.290
 2 1.250* .250 .009 .341 2.159
3 .625 .263 .295 -.331 1.581
2 1 2 -.625 .263 .295 -1.581 .331
*
3 -.875 .227 .037 -1.699 -.051
*
4 -2.000 .267 <.001 -2.972 -1.028
2 1 .625 .263 .295 -.331 1.581
3 -.250 .250 1.000 -1.159 .659
*
4 -1.375 .263 .007 -2.331 -.419
*
3 1 .875 .227 .037 .051 1.699
2 .250 .250 1.000 -.659 1.159
4 -1.125* .125 <.001 -1.579 -.671
*
4 1 2.000 .267 <.001 1.028 2.972
*
2 1.375 .263 .007 .419 2.331
*
3 1.125 .125 <.001 .671 1.579
3 1 2 -.375 .183 .478 -1.040 .290
*
3 -.875 .227 .037 -1.699 -.051
*
4 -1.750 .164 <.001 -2.345 -1.155
2 1 .375 .183 .478 -.290 1.040
3 -.500 .189 .199 -1.187 .187
*
4 -1.375 .183 <.001 -2.040 -.710
*
3 1 .875 .227 .037 .051 1.699
2 .500 .189 .199 -.187 1.187
4 -.875* .227 .037 -1.699 -.051
*
4 1 1.750 .164 <.001 1.155 2.345
*
2 1.375 .183 <.001 .710 2.040
*
3 .875 .227 .037 .051 1.699

e. Give results and conclusions.

3. Five respondents in each of three occupational groups (belly dancers, politicians,

administrators) participated in four leisure activities (reading, dancing, watching TV, and
skiing) and, during each, were asked to rate their satisfaction on a 10-point scale. The data
are given below. Enter the data in SPSS and save the file under an appropriate name.

Activity
Case No. Read Dance TV Ski
1 7 10 6 5
2 8 9 5 7
Belly dancers 3 5 10 5 8
4 6 10 6 8
5 7 8 7 9
6 4 4 4 4
7 6 4 5 3
Politicians 8 5 5 5 6
9 6 6 6 7
 10 4 5 6 5
11 3 1 1 2
12 5 3 1 5
Administrators 13 4 2 2 5
14 7 1 2 4
15 6 3 3 3

Perform a mixed repeated measures analysis, and write down the results and
conclusions.

Descriptive Statistics

Descriptive Statistics
Std.
Mean Deviation N
reading 5.5333 1.40746 15
dancing 5.4000 3.26890 15
watching 4.2667 1.98086 15
skiing 5.4000 2.06328 15

Multivariate Test

Multivariate Testsa
Partial Eta
Effect Value F Hypothesis df Error df Sig. Squared
b
Hobby Pillai's Trace .468 3.522 3.000 12.000 .049 .468
b
Wilks' Lambda .532 3.522 3.000 12.000 .049 .468
b
Hotelling's Trace .880 3.522 3.000 12.000 .049 .468
Roy's Largest Root .880 3.522b 3.000 12.000 .049 .468

Mauchly’s test

Mauchly's Test of Sphericitya

Measure: MEASURE_1
Epsilonb
Within Subjects Mauchly's Approx. Chi- Greenhouse-
Effect W Square df Sig. Geisser Huynh-Feldt Lower-bound
Hobby .643 5.612 5 .347 .756 .910 .333

4. Open the file Ritalin.sav. The data are discussed on the lecture slides.
 a. Explain how the variables in the SPSS file are related to the variables discussed on the
lecture slides.
b. Perform a repeated measures analysis (General Linear Model – Repeated Measures).
Make sure to give an appropriate name to the Within-subject factor.

Mauchly's Test of Sphericitya

Measure: MEASURE_1
Epsilonb
Within Subjects Mauchly's Approx. Chi- Greenhouse-
Effect W Square df Sig. Geisser Huynh-Feldt Lower-bound
factor1 .427 17.893 2 <.001 .636 .657 .500

Tests of Within-Subjects Effects

Measure: MEASURE_1
Type III Sum Mean
Source of Squares df Square F Sig.
factor1 Sphericity 1121.671 2 560.836 3.527 .038
Assumed
Greenhouse- 1121.671 1.271 882.453 3.527 .062
Geisser
Huynh-Feldt 1121.671 1.314 853.722 3.527 .060
Lower-bound 1121.671 1.000 1121.671 3.527 .074
Error(facto Sphericity 6996.892 44 159.020
r1) Assumed
Greenhouse- 6996.892 27.964 250.212
Geisser
Huynh-Feldt 6996.892 28.905 242.066
Lower-bound 6996.892 22.000 318.041

Multivariate Testsa
Hypothesis
Effect Value F df Error df Sig.
factor1 Pillai's Trace .361 5.924b 2.000 21.000 .009
Wilks' Lambda .639 5.924b 2.000 21.000 .009
Hotelling's Trace .564 5.924b 2.000 21.000 .009
Roy's Largest .564 5.924b 2.000 21.000 .009
Root

c. What assumption needs to be tested first? Which test is used for this? What is the
conclusion?
d. Once you have decided which repeated measures analysis you must use (i.e., the
multivariate test, or the repeated measures ANOVA), check the right table to answer the
question whether dose has an effect on cognitive functioning. What is your conclusion?
 e. Also obtain the pairwise comparisons (under Options, ask to display means). What do
you conclude about the three doses?

5. Open the file BeerGogglesLighting.sav.

The data are explained on page 590 (Field) in Task 5.
a. What are the three null hypotheses tested in a two-way repeated measures analysis
with the factors alcohol consumption and lighting and the dependent variable
attractiveness?
No effect of gender (Reject) (significant)
No effect of product (reject) +
No effect of imajery (reject) +
No interaction between gender and product (Reject) +
No interaction between gender and imajery (Reject) +
No interaction between product and imajery (Reject) +
No interaction between gender, imajery, and product (rejec) +

b. Run the two-way repeated measures analysis and discuss the results (don’t forget the
assumption of sphericity).
c. Run the analysis again and activate PLOTS to obtain the interaction plot with alcohol
consumption on the x-axis. Describe the interaction.

6. Open the file MixedAttitude.sav.

The data are explained in Chapter 15 and on page 621 (Field).
a. Formulate all null-hypotheses.
b. Give the descriptive statistics (means and standard deviations) for males and females.
c. Perform a mixed repeated measures analysis and test all null-hypotheses. Provide all
significance values and conclusions.
d. Run the analysis again and activate PLOTS. Produce separate plots for males and
females with drink on the x-axis and separate lines for the types of imagery. Discuss the
results.