Which statistic is used to assess the internal consistency of a multi-item scale?
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.
ACronbach's alpha
BIntraclass correlation coefficient
CCohen's kappa
DContent validity index
Show answer and explanation
✓ Correct answerA. Cronbach's alpha
Explanation
Internal consistency is the degree to which the items of a scale measure the same underlying construct and agree with one another. Cronbach's alpha is the standard statistic for this, ranging from 0 to 1, with values of about 0.70 or above usually taken as acceptable for a research instrument. Other reliability statistics answer different questions: kappa is for agreement between raters on categorical ratings, and the intraclass correlation is mainly for agreement or consistency of continuous measurements across raters or occasions.
Why each option is right or wrong
✓A. Cronbach's alpha — Alpha summarises how closely a set of items on one scale hang together, which is exactly internal consistency.
✗B. Intraclass correlation coefficient — The intraclass correlation coefficient is used mainly for agreement of continuous measurements between raters or repeated occasions.
✗C. Cohen's kappa — Cohen's kappa measures agreement between two raters on categorical judgements, not the coherence of items within a scale.
✗D. Content validity index — The content validity index quantifies expert judgement about whether items cover the intended domain, which is validity rather than reliability.
Statistic
What it assesses
Typical use
Cronbach's alpha
Internal consistency of a multi-item scale
Questionnaires and Likert scales
Cohen's kappa
Agreement between raters beyond chance
Categorical ratings by two observers
Intraclass correlation
Agreement/consistency of continuous measures
Raters or repeated occasions
Content validity index
Expert judgement of item relevance
Scale development, validity
EXAM HINT Alpha = All items Agree (internal consistency); kappa = two raters.
اردو وضاحت
انٹرنل کنسسٹنسی کیا ہے؟
جب ہم کوئی اسکیل یا questionnaire بناتے ہیں جس میں کئی سوالات ہوتے ہیں، تو انٹرنل کنسسٹنسی کا مطلب یہ ہے کہ یہ سارے سوالات ایک ہی چیز کو ناپ رہے ہوں اور آپس میں ملتے جلتے جواب دیں۔ سیدھی بات یہ ہے کہ اگر ایک سوال کا جواب ہاں ہے تو دوسرے متعلقہ سوال کا جواب بھی ویسا ہی ہونا چاہیے۔
انٹرنل کنسسٹنسی ناپنے کے لیے کون سا statistic استعمال ہوتا ہے؟
Cronbach's alpha وہی statistic ہے جو بتاتا ہے کہ اسکیل کے سوالات آپس میں کتنا میل کھاتے ہیں۔ اس کی value 0 سے 1 تک ہوتی ہے اور عام طور پر 0.70 یا اس سے زیادہ کو قابلِ قبول سمجھا جاتا ہے۔ Kappa اور intraclass correlation الگ چیزوں کے لیے ہیں، جیسے دو raters کے درمیان اتفاق، اس لیے یہاں Cronbach's alpha ہی صحیح ہے۔