Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

Validity and Reliability Analysis 
Validity is concerned with whether a variable measures what it is suppose to measure. Construct validity is 
based on a philosophical view that hypothetical constructs are unobservable and cannot be measured directly but 
that measures can be developed which at least partially represent the constructs (Peter, 1979). Construct validity 
assesses whether a measure relates to other observed variables in a way that is consistent with theoretically derived 
predictions (Bollen, 1989). Reliability is the consistency of measurement. Reliability is the part of measure that is 
free of purely random error and thus reliability coefficients estimate the amount of systematic variance in a measure 
(Peter, 1979). In general, four types of procedures are used for assessing reliability: test-retest, alternative forms, 
split-halves, and Cronbach’s alphas. This Cronbach’s alpha reliability estimates are a result of interplay between the 
number of items comprising the instrument and the interrelations among the items. Cronbach’s alpha (Nunnally, 
1988) is the most popular reliability coefficient in social science research and Cronbach’s alphas coefficients for the 
retained items ranged from .791 to .897 an indication of the reliability of the measurement (see Table 1).