What is Validity?
Validity refers to the degree to which an instrument measures what it is intended to measure. A valid questionnaire truly captures the concept under study.
Types of Validity
1. Face Validity (صدق المظهر)
Does the questionnaire look like it measures what it should? This is assessed by having experts review the instrument.
How to establish: Send your questionnaire to 5-10 experts (professors, practitioners) and ask them to evaluate each item's clarity and relevance.
2. Content Validity (صدق المحتوى)
Does the questionnaire cover all aspects of the concept? Are all dimensions adequately represented?
How to establish: Compare your items against the literature review to ensure comprehensive coverage.
3. Construct Validity (صدق البناء)
Does the questionnaire accurately measure the theoretical construct? Measured using factor analysis in SPSS.
Types:
- Convergent validity: Items measuring the same construct are correlated
- Discriminant validity: Items measuring different constructs are not highly correlated
4. Criterion Validity (صدق المحك)
How well does the instrument correlate with an established measure of the same concept?
What is Reliability?
Reliability refers to the consistency and stability of measurement. If you administer the same questionnaire twice, do you get similar results?
Types of Reliability
1. Internal Consistency (Cronbach's Alpha)
The most commonly used measure. Measures how well items within a scale are correlated with each other.
| Alpha Value | Interpretation |
|---|---|
| α ≥ 0.90 | Excellent |
| 0.80 ≤ α < 0.90 | Good |
| 0.70 ≤ α < 0.80 | Acceptable |
| 0.60 ≤ α < 0.70 | Questionable |
| α < 0.60 | Unacceptable |
2. Test-Retest Reliability
Administer the questionnaire twice to the same group with a time interval (usually 2-4 weeks). Calculate the correlation between the two sets of scores.
3. Split-Half Reliability
Divide the questionnaire into two halves (odd vs. even items) and correlate the scores.
Pilot Study
Before the main study, conduct a pilot study (دراسة استطلاعية):
- Select 30-50 participants (not from the final sample)
- Administer the questionnaire
- Calculate Cronbach's Alpha for each dimension and overall
- Identify and revise problematic items
- Check "Alpha if item deleted" — remove items that improve alpha when deleted
Important: In SPSS, you can calculate Cronbach's Alpha via: Analyze → Scale → Reliability Analysis. We will practice this in detail in the SPSS lessons.