Scribd
Upload
202 views5 pages

Test Retest Reliability

Test-retest reliability was examined for composite scores and single items on a self-efficacy instrument administered twice over a 2 week period. For composite scores of 9 students, the mean scores were similar at Time 1 and Time 2 and the correlation was very high (r = .958). For a single item administered to the same 9 students, the mean scores were also similar at both time points and the correlation was strong (r = .825), indicating the item provided stable scores over time. In both cases, the instrument demonstrated good test-retest reliability.

Uploaded by

Irina Franghiu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
202 views5 pages

Test Retest Reliability

Test-retest reliability was examined for composite scores and single items on a self-efficacy instrument administered twice over a 2 week period. For composite scores of 9 students, the mean scores were similar at Time 1 and Time 2 and the correlation was very high (r = .958). For a single item administered to the same 9 students, the mean scores were also similar at both time points and the correlation was strong (r = .825), indicating the item provided stable scores over time. In both cases, the instrument demonstrated good test-retest reliability.

Uploaded by

Irina Franghiu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Test-retest Reliability (draft)

1. Composite scores

Example of items measuring self-efficacy toward dissertation process.

Administer instrument to group of students once, let time elapse, say 2 weeks,
readminister second time. Scores must be linked, so have some way to identify and match
scors for students.

Form composite scores, then correlate scores. Example of composite scores belw
Student, Time1, Time2
a 5.35 5.24
b 6.33 5.89
c 4.28 5.00
d 4.18 3.90
e 3.33 3.50
f 5.89 5.26
g 1.29 2.33
h 6.35 6.12
i 4.25 4.89

In SPSS:
Run correlation

Analyze -> Correlate - > Bivariate

Popup window for correlation,

Move scores from times 1 and 2 to Variables box, click options, choose M and SD:

Click continue then OK to run:

SPSS results

Descriptive Statistics

Mean Std. Deviation N


Time1 4.5833 1.62979 9
Time2 4.6811 1.21745 9
Correlations

Time1 Time2
Time1 Pearson
1 .958(**)
Correlation
Sig. (2-tailed) .000
N 9 9
Time2 Pearson
.958(**) 1
Correlation
Sig. (2-tailed) .000
N 9 9
** Correlation is significant at the 0.01 level (2-tailed).

Note means are similar (use correlated samples t-test to formally test whether means are
similar), and r is strong and positive (r = .958).

2. For Single Item, such as Item 5 here:

Does item 5 provide stable scores over time? Administer to group twice, then correlate
scores in same manner as above.

Example Data
Student Time1 Time2
a 5.00 5.00
b 6.00 5.00
c 4.00 5.00
d 4.00 3.00
e 3.00 4.00
f 5.00 4.00
g 1.00 2.00
h 7.00 6.00
i 4.00 5.00

In SPSS

Results of correlation:

Descriptive Statistics

Mean Std. Deviation N


item5_time1 4.3333 1.73205 9
item5_time2 4.3333 1.22474 9

Correlations
item5_time1 item5_time2
item5_time1 Pearson
1 .825(**)
Correlation
Sig. (2-tailed) .006
N 9 9
item5_time2 Pearson
.825(**) 1
Correlation
Sig. (2-tailed) .006
N 9 9
** Correlation is significant at the 0.01 level (2-tailed).

Similar means, strong correlation at r = .825 looks like item 5 provides consistent scores (stable scores) over
time.

You might also like