Exam 1

1. Metainformation

   Tag	Value
   file	Reliability_eur-reliability-103-en_eur-reliability-103-en
   name	eur-reliability-103-en
   section	Reliability/Analysis/Parallel forms, Reliability/Analysis/Test-retest
   type	string
   solution	""
   Type	Conceptual
   Program
   Language	English
   Level	Statistical Literacy

   Question

   Imagine the following two scenario’s:

   • You have two different, yet parallel tests that measure the same construct, with each test consisting of 8 items. You calculate an alternate forms estimate of the test’s reliability (correlation between alternate forms of a test).
   • You have one test consisting of 16 items. You calculate the split-half estimate of the reliability of the test.

   Explain why the (split‐half) estimate of the reliability of the 16-item scale is higher than the (alternate forms) estimate of the reliability of the 8 item scale.

   
   

   Solution

   The (split‐half) estimate of the reliability of the 16 item scale is higher than the (alternate forms) estimate of the reliability of the 8 item scale because of its longer test length. In general: the more measurements one takes of a single construct, the higher the reliability of the (average) measurement will be. S ‐ ceteris paribus – a scale score that is based on 16 items is more reliable than a scale score based on 8 items.