30 ITEMS – INITIAL
Case Processing Summary
N %
Valid 30 100.0
Cases Excludeda 0 .0
Total 30 100.0
a. Listwise deletion based on all variables in the
procedure.
Reliability Statistics
Cronbach's N of Items
Alpha
.633 30
Item-Total Statistics
Scale Mean if Scale Variance Corrected Item- Cronbach's
Item Deleted if Item Deleted Total Correlation Alpha if Item
Deleted
1. 14.47 17.016 .014 .640
2. 15.07 16.823 .115 .630
3. 15.00 17.241 -.041 .643
4. 14.93 17.582 -.137 .652
5. 14.80 15.683 .339 .609
6. 14.80 15.683 .339 .609
7. 14.70 16.838 .041 .639
8. 14.43 16.944 .040 .637
9. 14.43 16.806 .079 .634
10. 14.73 16.271 .181 .625
11. 14.53 15.706 .351 .609
12. 14.63 16.447 .139 .629
13. 14.83 15.799 .316 .612
14. 14.70 15.666 .334 .609
15. 14.93 18.271 -.313 .667
16. 14.60 15.490 .390 .604
17. 14.67 14.368 .687 .571
18. 15.00 17.517 -.122 .649
19. 14.50 15.776 .345 .610
20. 14.50 15.500 .423 .602
21. 14.60 15.283 .446 .598
22. 14.50 16.810 .063 .636
23. 14.87 16.602 .112 .632
24. 14.43 16.116 .280 .617
25. 14.67 17.333 -.076 .650
26. 14.67 15.126 .478 .594
27. 14.70 15.872 .281 .615
28. 14.93 17.926 -.226 .660
29. 14.57 16.323 .178 .625
30. 14.60 15.421 .409 .602
ANOVA
Sum of Squares df Mean Square F Sig
Between People 16.693 29 .576
Between Items 30.427 29 1.049 4.962 .000
Within People Residual 177.840 841 .211
Total 208.267 870 .239
Total 224.960 899 .250
Grand Mean = .51
DELETED 6 ITEMS – 3, 4, 15, 18, 25, 28
Based on the results of your reliability test, here’s an expert interpretation and analysis of the
data:
🔍 Summary of Results
✅ 1. Cronbach’s Alpha:
Value: 0.633
Interpretation:
o Cronbach’s alpha measures the internal consistency or reliability of a
questionnaire.
o A rule of thumb for interpretation:
> 0.9 – Excellent
0.8 – 0.9 – Good
0.7 – 0.8 – Acceptable
0.6 – 0.7 – Questionable
0.5 – 0.6 – Poor
< 0.5 – Unacceptable
o Your value of 0.633 indicates questionable reliability. This means the items
are not consistently measuring the same construct.
📊 2. Item-Total Statistics
Several items have negative or very low corrected item-total correlations,
including:
o Item 3: -0.041
o Item 4: -0.137
o Item 15: -0.313
o Item 18: -0.122
o Item 25: -0.076
o Item 28: -0.226
Items with negative or very low correlations (< 0.2) suggest they do not align well with the
overall scale, and may be confusing, irrelevant, or measuring something different.
🔢 3. Effect of Deleting Items
Cronbach’s Alpha would increase if some problematic items were deleted:
o Removing Item 15 would increase alpha to 0.667
o Removing Item 28 would increase alpha to 0.660
o Removing Item 4 would increase alpha to 0.652
This implies that these items are likely reducing the reliability of your instrument.
📈 4. ANOVA Table
The F-test is significant (p = .000) for differences between items, meaning the items
are not performing uniformly — another indicator of inconsistency.
📌 Recommendations for Improvement
1. Revise or Remove Problematic Items:
o Review the content of items with negative/low correlations:
Items 3, 4, 15, 18, 25, and 28
o Check for:
Ambiguous wording
Double-barreled questions (asking two things at once)
Items not aligned with the main construct being measured
2. Simplify Language:
o Make sure items are clear and easily understandable, especially if respondents
vary in age, education, or language proficiency.
3. Ensure Unidimensionality:
o Make sure all items are measuring the same underlying construct. If not, you
may need to split the questionnaire into subscales.
4. Pilot Test Revised Version:
o After making changes, conduct another pilot (at least 30 respondents) and
rerun the reliability test.
5. Consider Expert Review:
o Have subject matter experts review the questionnaire to evaluate whether
items are relevant and appropriately phrased.
✅ Conclusion
Your questionnaire has moderate but questionable reliability. With a few item revisions
or removals, you can potentially improve its internal consistency. Focus especially on the
items that negatively affect Cronbach’s alpha.
With Deleted Items: 3,4,15,18,25,28
Case Processing Summary
N %
Valid 30 100.0
Cases Excludeda 0 .0
Total 30 100.0
a. Listwise deletion based on all variables in the
procedure.
Reliability Statistics
Cronbach's N of Items
Alpha
.751 24
Item-Total Statistics
Scale Mean if Scale Variance Corrected Item- Cronbach's
Item Deleted if Item Deleted Total Correlation Alpha if Item
Deleted
1. 12.73 19.375 .015 .759
2. 13.33 19.264 .083 .753
5. 13.07 18.064 .313 .741
6. 13.07 17.720 .398 .735
7. 12.97 19.137 .054 .758
8. 12.70 19.114 .090 .754
9. 12.70 18.838 .164 .750
10. 13.00 18.552 .189 .749
11. 12.80 17.614 .446 .732
12. 12.90 18.990 .089 .756
13. 13.10 17.748 .399 .735
14. 12.97 17.826 .361 .737
16. 12.87 17.706 .401 .735
17. 12.93 16.754 .632 .718
19. 12.77 17.840 .401 .735
20. 12.77 17.840 .401 .735
21. 12.87 17.085 .559 .723
22. 12.77 18.944 .117 .753
23. 13.13 19.016 .093 .755
24. 12.70 18.079 .375 .737
26. 12.93 17.789 .372 .737
27. 12.97 18.171 .278 .743
29. 12.83 18.489 .215 .747
30. 12.87 17.568 .436 .732
ANOVA
Sum of Squares df Mean Square F Sig
Between People 23.728 29 .818
Between Items 17.444 23 .758 3.716 .000
Within People Residual 136.139 667 .204
Total 153.583 690 .223
Total 177.311 719 .247
Grand Mean = .56
🔍 Interpretation and Discussion of Reliability Test Results
✅ 1. Improved Cronbach’s Alpha
Result: 0.751
Interpretation:
o This value indicates that your revised questionnaire now has acceptable
reliability.
o Compared to the previous alpha of 0.633, the reliability significantly
improved after removing six low-performing items.
o A Cronbach’s alpha between 0.7 and 0.8 is generally considered acceptable,
especially for exploratory research or newly developed instruments.
✅ Conclusion: The internal consistency of your instrument is now acceptable, which means
the retained items are more aligned in measuring the same construct.
📊 2. Item-Total Statistics
Corrected Item-Total Correlation:
o Most items now show moderate positive correlations with the total scale.
o Notable items with strong correlations:
Item 17: 0.632
Item 21: 0.559
o Some items still show low correlations (below 0.2):
Item 1: 0.015
Item 2: 0.083
Item 7: 0.054
Item 8: 0.090
Item 12: 0.089
Item 22: 0.117
Item 23: 0.093
⚠️Insight: These low-correlating items may still not be strongly related to the central
concept of your questionnaire. However, since overall reliability remains acceptable, their
impact is limited — but you may consider further revision or replacement if higher
reliability is desired.
📈 3. Cronbach's Alpha if Item Deleted
No item, if deleted, would significantly increase Cronbach’s alpha.
Example:
o Deleting Item 1 would increase alpha only slightly to 0.759.
o Other items, if removed, would lower or keep alpha within the same range.
✅ This confirms that the current set of 24 items is relatively balanced, and most items now
contribute positively to reliability.
🧮 4. ANOVA Table
F = 3.716, Sig. = 0.000
Interpretation:
o There are statistically significant differences between items, meaning not all
items are interpreted equally by respondents.
o However, in the context of a multi-item questionnaire, this is not necessarily
negative — items can vary in difficulty or emphasis but still measure the same
construct.
📌 Summary
Component Before (30 items) After (24 items) Interpretation
✅ Improved
Cronbach's Alpha 0.633 (Questionable) 0.751 (Acceptable)
reliability
Several with Still a few with low
Problematic ⚠️Some items can
negative/low correlation but no negative
Items be revised
correlations values
Remove or revise weak Continue refining low-
Recommendation 👍 Good progress
items correlation items
💡 Recommendations for Further Improvement
1. Review Low-Performing Items:
o Items with correlation < 0.2 (e.g., Items 1, 2, 7, 8, 12, 22, 23) may need:
Rewording for clarity
Alignment with main construct
Removal if found irrelevant or confusing
2. Conduct Validity Testing:
o Consider testing for construct validity (e.g., factor analysis) to ensure items
measure the intended dimensions.
3. Pilot with Larger Sample:
o Test the updated questionnaire with a larger and more diverse group to see
if reliability holds across different populations.
4. Add Reverse-Coded Items (if needed):
o Including well-designed reverse items may help identify response bias and
strengthen internal consistency.
20 Items Test – Deleting Items 3,4,15,18,25,28,
8,12,22,23
Case Processing Summary
N %
Cases Valid 30 100.0
Excludeda 0 .0
Total 30 100.0
a. Listwise deletion based on all variables in the
procedure.
Reliability Statistics
Cronbach's N of Items
Alpha
.771 20
Item-Total Statistics
Scale Mean if Scale Variance Corrected Item- Cronbach's
Item Deleted if Item Deleted Total Correlation Alpha if Item
Deleted
1. 10.37 16.861 .055 .779
2. 10.97 16.930 .076 .775
5. 10.70 15.803 .306 .763
6. 10.70 15.528 .379 .758
7. 10.60 16.662 .083 .779
9. 10.33 16.299 .225 .768
10. 10.63 16.585 .102 .777
11. 10.43 15.220 .485 .751
13. 10.73 15.375 .429 .754
14. 10.60 15.628 .343 .761
16. 10.50 15.500 .387 .757
17. 10.57 14.461 .660 .737
19. 10.40 15.283 .484 .751
20. 10.40 15.697 .366 .759
21. 10.50 14.948 .537 .746
24. 10.33 15.885 .349 .760
26. 10.57 15.633 .343 .761
27. 10.60 15.903 .272 .766
29. 10.47 16.051 .248 .767
30. 10.50 15.155 .480 .751
ANOVA
Sum of Squares df Mean Square F Sig
Between People 25.035 29 .863
Between Items 14.018 19 .738 3.725 .000
Within People Residual 109.132 551 .198
Total 123.150 570 .216
Total 148.185 599 .247
Grand Mean = .55