error in SPSS output

#1
Hello, I am using ANOVA to compare the variable means of 3 groups (N-52), but keep receiving a warning error in SPSS output - "Post Hoc tests are not performed because at least one group has fewer than 2 cases. And also "Test of homogeneity of variances cannot be performed because the sum of caseweights is less than the number of groups". Could anyone help me solve this problem? Many thanks!
PS -I am using SPSS 16
 

Masteras

TS Contributor
#2
Hi bobrose, one of your groups has 1 observation only. You have to remove this group (or line). When you ask for anova, ask for the descriptive statistics also. You will see how many observations you have in each sample. One of these will have only 1. Remove that one and re-perform the anova.
 
#3
Hi Masteras,

Thank you for your reply.
I have 3 groups, (G1 N-23, G2 N=14, G3 N-15), and descriptive statistics show me the 3 groups, but only when performing ANOVA with homogeneity and post-hoc test that I get these errors. I was wondering if it is not a bug of the program, but I could not find descrition of it before.

Output In One-Way Anova

Descriptives
IIEF questionnaire
95% Confidence Interval for Mean
N Mean Std. Deviation Std. Error Lower Bound Upper Bound Minimum Maximum
21.9 1 64,00 . . . . 64 64
22.3 2 57,50 ,707 ,500 51,15 63,85 57 58
22.5 1 53,00 . . . . 53 53
22.6 1 52,00 . . . . 52 52
23.1 1 72,00 . . . . 72 72
23.3 1 71,00 . . . . 71 71
23.7 1 54,00 . . . . 54 54
24.7 1 66,00 . . . . 66 66
24.8 1 69,00 . . . . 69 69
25.4 1 69,00 . . . . 69 69
25.5 2 64,00 9,899 7,000 -24,94 152,94 57 71
26.5 1 59,00 . . . . 59 59
27.5 1 56,00 . . . . 56 56
27.6 1 69,00 . . . . 69 69
27.8 1 54,00 . . . . 54 54
28.4 1 64,00 . . . . 64 64
29.7 2 62,00 5,657 4,000 11,18 112,82 58 66
30.4 1 34,00 . . . . 34 34
30.5 1 61,00 . . . . 61 61
30.8 1 40,00 . . . . 40 40
30.9 1 70,00 . . . . 70 70
31.4 1 69,00 . . . . 69 69
32 1 61,00 . . . . 61 61
32.1 1 32,00 . . . . 32 32
33.3 1 28,00 . . . . 28 28
33.6 1 54,00 . . . . 54 54
34.1 1 55,00 . . . . 55 55
34.2 2 67,00 2,828 2,000 41,59 92,41 65 69
34.3 1 63,00 . . . . 63 63
34.6 1 53,00 . . . . 53 53
34.8 1 46,00 . . . . 46 46
35.4 1 42,00 . . . . 42 42
35.9 1 57,00 . . . . 57 57
36.8 1 63,00 . . . . 63 63
37.6 1 58,00 . . . . 58 58
37.8 1 56,00 . . . . 56 56
39 1 30,00 . . . . 30 30
39.1 1 40,00 . . . . 40 40
39.9 1 52,00 . . . . 52 52
40.9 1 19,00 . . . . 19 19
41 1 71,00 . . . . 71 71
42 1 61,00 . . . . 61 61
42.4 1 61,00 . . . . 61 61
44.5 1 67,00 . . . . 67 67
45 1 64,00 . . . . 64 64
49.8 1 46,00 . . . . 46 46
55.9 1 30,00 . . . . 30 30
Total 51 56,00 12,903 1,807 52,37 59,63 19 72

Warnings
Post hoc tests are not performed for IIEF questionnaire because at least one group has fewer than two cases.

Test of Homogeneity of Variances(b)
IIEF questionnaire
Levene Statistic df1 df2 Sig.
.a 3 . .
a. Groups with only one case are ignored in computing the test of homogeneity of variance for IIEF questionnaire.
b. Test of homogeneity of variances cannot be performed for IIEF questionnaire because the sum of caseweights is less than the number of groups.

ANOVA
IIEF questionnaire
Sum of Squares df Mean Square F Sig.
Between Groups 8185,500 46 177,946 5,139 ,060
Within Groups 138,500 4 34,625
Total 8324,000 50

Thanks.

Alberto
 
#6
Dear Masteras,
Would you please help me to find the homogeneity test for my data?
These are the result after I analyze it with one way Anova. I used SPSS 17. *attached

Thank you.
 

Masteras

TS Contributor
#7
The second table is the test. But you shouldn't do anova. Who told you this? If you do want to do anova how many observations are in each group? Give me more information.
 
#8
Hi, According to your letter in this post, but I still could not solve my problem." Post hoc tests are not performed for TNV because error term has zero degrees of freedom."
I wanted to send my data file so that if you could help me.
Please send me your email address so I can send the files to you.
 

Naren

New Member
#9
I'm doing "two-factor factorial design" on spss but having trouble with geting out puts

Post hoc tests are not performed for Date because error term has zero degrees of freedom.
Post hoc tests are not performed for SamplePoint because error term has zero degrees of freedom.
All absolute deviations are constant within each cell. Levene F statistics cannot be computed.

these error messages are appearing
what could be the reason