Test-retest reliability of an instrument is established by which method?
Free Applied Statistics in Healthcare practice question on Reliability and validity for BSN and Post-RN nursing students. Choose your answer, then open the answer and the explanation.
ACorrelating the instrument's scores with those of an established gold-standard instrument
BAdministering the instrument to the same people on two occasions and correlating the scores
CComparing the instrument's items with the content of the construct it is meant to measure
DHaving two independent raters score the same set of responses and comparing their scores
Show answer and explanation
✓ Correct answerB. Administering the instrument to the same people on two occasions and correlating the scores
Explanation
Reliability is about consistency of measurement, and test-retest reliability is the form that checks stability over time. You give the same instrument to the same group on two separate occasions and then correlate the two sets of scores; a high correlation means the instrument gives stable results when nothing has changed. The time gap matters: too short and people remember their answers, too long and real change in the attribute can lower the correlation.
Why each option is right or wrong
✗A. Correlating the instrument's scores with those of an established gold-standard instrument — Correlating scores with a gold-standard instrument assesses criterion validity, which is about accuracy against a reference standard.
✓B. Administering the instrument to the same people on two occasions and correlating the scores — Giving the same instrument twice to the same people and correlating the two sets of scores is the standard method for testing stability over time.
✗C. Comparing the instrument's items with the content of the construct it is meant to measure — Matching items to the construct's content is content validity, which asks whether the tool covers the domain, not whether it is consistent.
✗D. Having two independent raters score the same set of responses and comparing their scores — Comparing scores from two independent raters assesses inter-rater reliability, which concerns agreement between observers rather than stability over time.
Type of reliability/validity
What it checks
How it is tested
Test-retest reliability
Stability over time
Same people tested twice; scores correlated
Inter-rater reliability
Agreement between observers
Two raters score the same responses
Internal consistency
Consistency across items
Cronbach's alpha
Content validity
Coverage of the construct
Expert review of items
Criterion validity
Accuracy against a standard
Correlation with a gold-standard tool
EXAM HINT Test-retest = two tests, one person-set, then correlate: it checks stability, not accuracy.
اردو وضاحت
Test-retest reliability کیا ہے؟
Reliability کا مطلب ہے کہ کوئی instrument جب بار بار استعمال ہو تو ایک جیسا نتیجہ دے۔ Test-retest reliability یہ دیکھتی ہے کہ وقت گزرنے کے بعد بھی وہی instrument وہی نتیجہ دیتا ہے یا نہیں۔
یہ کس طریقے سے معلوم کی جاتی ہے؟
ایک ہی گروپ کو وہی instrument دو مختلف موقعوں پر دے کر دونوں scores کو correlate کیا جاتا ہے۔ اگر correlation زیادہ ہو تو سمجھیں instrument مستحکم ہے۔ content validity یا دو raters کا آپس میں موازنہ کچھ اور چیزیں ناپتے ہیں، وقت کے ساتھ استحکام نہیں۔
Key terms
correlating — A statistical way of checking how closely two sets of scores move together. (ایک statistical طریقہ جو بتاتا ہے کہ دو scores کتنی قریب قریب چلتے ہیں۔)