### SPSS Code Fragment Repeated measures ANOVA in SPSS

* Example 1.
* Table 1, page 264.
DATA LIST LIST / subject cond1 cond2 cond3.
BEGIN DATA.
1 100  90 130
2  90 100 100
3 110 110 109
4 100  90 109
5 100 100 130
END DATA.

GLM cond1 cond2 cond3
/WSFACTOR = conditn 3.

The GLM command produces 3 of the results shown on Table 1 on page 264.
1. Anova with uncorrected df: F(2,8) = 4.73, p = 0.044, shown in red in the table immediately below.
2. Anova with Huynh-Feldt corrected df, F(1.59, 6.36) = 4.73, shown in pink in the table immediately below
3. MANOVA (Wilks's Lambda), F(2,3) = 2.40, p=0.238, shown in blue in the second table below.

Tests of Within-Subjects Effects Measure: MEASURE_1
Source Type III Sum of Squares df Mean Square F Sig.
CONDITN Sphericity Assumed 928.533 2 464.267 4.725 .044
Greenhouse-Geisser 928.533 1.269 731.910 4.725 .077
Huynh-Feldt 928.533 1.590 583.935 4.725 .060
Lower-bound 928.533 1.000 928.533 4.725 .095
Error(CONDITN) Sphericity Assumed 786.133 8 98.267

Greenhouse-Geisser 786.133 5.075 154.916

Huynh-Feldt 786.133 6.361 123.596

Lower-bound 786.133 4.000 196.533


Multivariate Tests(b)
Effect Value F Hypothesis df Error df Sig.
CONDUIT Pillai's Trace .615 2.401(a) 2.000 3.000 .238
Wilks' Lambda .385 2.401(a) 2.000 3.000 .238
Hotelling's Trace 1.601 2.401(a) 2.000 3.000 .238
Roy's Largest Root 1.601 2.401(a) 2.000 3.000 .238
a Exact statistic
b Design: Intercept
Within Subjects Design: CONDITN

* Example 2.
* Table 3, page 268.
DATA LIST LIST / subject c1t1 c1t2 c1t3  c2t1 c2t2 c2t3  c3t1 c3t2 c3t3.
BEGIN DATA.
1  8  9  8   8  9  7  10  9 10
2  9 10  9  10  9 13   8  9  9
3  8  7  7  12  7  9  10  9  7
4  6  8  9   8 10 10  12  9 10
5  7  6  7  11 12  8   8 11  9
END DATA.

GLM c1t1 c1t2 c1t3  c2t1 c2t2 c2t3  c3t1 c3t2 c3t3
/WSFACTOR = cond 3 trial 3.

The GLM command produces two of the results shown on Table 3 on page 268.
1. Anova with Huynh-Feldt corrected df, F(2,8) = 4.02, shown in pink in the table immediately below
2. MANOVA (Wilks's Lambda), F(2,3) = 3.30, p=0.175, shown in blue in the second table below.

Tests of Within-Subjects Effects
Measure: MEASURE_1

Source Type III Sum of Squares df Mean Square F Sig.
COND Sphericity Assumed 24.844 2 12.422 4.022 .062
Greenhouse-Geisser 24.844 1.976 12.570 4.022 .063
Huynh-Feldt 24.844 2.000 12.422 4.022 .062
Lower-bound 24.844 1.000 24.844 4.022 .115
Error(COND) Sphericity Assumed 24.711 8 3.089

Greenhouse-Geisser 24.711 7.906 3.126

Huynh-Feldt 24.711 8.000 3.089

Lower-bound 24.711 4.000 6.178

TRIAL Sphericity Assumed .311 2 .156 .063 .940
Greenhouse-Geisser .311 1.783 .174 .063 .924
Huynh-Feldt .311 2.000 .156 .063 .940
Lower-bound .311 1.000 .311 .063 .815
Error(TRIAL) Sphericity Assumed 19.911 8 2.489

Greenhouse-Geisser 19.911 7.134 2.791

Huynh-Feldt 19.911 8.000 2.489

Lower-bound 19.911 4.000 4.978

COND * TRIAL Sphericity Assumed 1.689 4 .422 .191 .940
Greenhouse-Geisser 1.689 2.601 .649 .191 .878
Huynh-Feldt 1.689 4.000 .422 .191 .940
Lower-bound 1.689 1.000 1.689 .191 .685
Error(COND*TRIAL) Sphericity Assumed 35.422 16 2.214

Greenhouse-Geisser 35.422 10.402 3.405

Huynh-Feldt 35.422 16.000 2.214

Lower-bound 35.422 4.000 8.856


Multivariate Tests(b)
Effect Value F Hypothesis df Error df Sig.
COND Pillai's Trace .688 3.301(a) 2.000 3.000 .175
Wilks' Lambda .312 3.301(a) 2.000 3.000 .175
Hotelling's Trace 2.201 3.301(a) 2.000 3.000 .175
Roy's Largest Root 2.201 3.301(a) 2.000 3.000 .175
TRIAL Pillai's Trace .028 .043(a) 2.000 3.000 .959
Wilks' Lambda .972 .043(a) 2.000 3.000 .959
Hotelling's Trace .028 .043(a) 2.000 3.000 .959
Roy's Largest Root .028 .043(a) 2.000 3.000 .959
COND * TRIAL Pillai's Trace .348 .133(a) 4.000 1.000 .948
Wilks' Lambda .652 .133(a) 4.000 1.000 .948
Hotelling's Trace .533 .133(a) 4.000 1.000 .948
Roy's Largest Root .533 .133(a) 4.000 1.000 .948
a Exact statistic
b Design: Intercept
Within Subjects Design: COND+TRIAL+COND*TRIAL

* Example 3.
* Table 4, page 269.
DATA LIST LIST / subject cond1 cond2 cond3 .
BEGIN DATA.
1 8.333  8.000  9.667
2 9.333 10.667  8.667
3 7.333  9.333  8.667
4 7.667  9.333 10.333
5 6.667 10.333  9.333
END DATA.

GLM cond1 cond2 cond3
/WSFACTOR cond 3.

The GLM command produces two of the results shown on Table 4 on page 269.
1. Anova with Huynh-Feldt corrected df, F(2,8) = 4.02, shown in pink in the table immediately below
2. MANOVA (Wilks's Lambda), F(2,3) = 3.31, p=0.174, shown in blue in the second table below.

Tests of Within-Subjects Effects
Measure: MEASURE_1
Source Type III Sum of Squares df Mean Square F Sig.
COND Sphericity Assumed 8.282 2 4.141 4.023 .062
Greenhouse-Geisser 8.282 1.976 4.191 4.023 .063
Huynh-Feldt 8.282 2.000 4.141 4.023 .062
Lower-bound 8.282 1.000 8.282 4.023 .115
Error(COND) Sphericity Assumed 8.234 8 1.029

Greenhouse-Geisser 8.234 7.905 1.042

Huynh-Feldt 8.234 8.000 1.029

Lower-bound 8.234 4.000 2.058



Multivariate Tests(b)

Effect Value F Hypothesis df Error df Sig.
COND Pillai's Trace .688 3.304(a) 2.000 3.000 .174
Wilks' Lambda .312 3.304(a) 2.000 3.000 .174
Hotelling's Trace 2.203 3.304(a) 2.000 3.000 .174
Roy's Largest Root 2.203 3.304(a) 2.000 3.000 .174
a Exact statistic
b Design: Intercept
Within Subjects Design: COND

The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California.