Chapter 9 l Measurement Scaling Reliability and Validity
Chapter: # 9 l Measurement: Scaling, Reliability and Validity
Rating Scale Rating scale is scale with several response categories that evaluate an object on a scale. l l l Dichotomous scale. Category scale. Likert scale. Numerical scale. Semantic Differential scale. 2
Rating Scale (cont. ) l l l Itemized Rating scale. Fixed or Constant Sum Rating scale. Stapel scale. Graphic Rating scale. Consensus scale. 3
Ranking Scale Ranking scale is scale used to tap preferences between two or among more objects or items. Paired comparison. l Forced choice. l Comparative scale. l Item analysis. l 4
Goodness of Measures Accurately. l Actually. l 5
Reliability Stability of Measures ~ Test-Retest reliability. ~ Parallel-Form reliability. l Internal Consistency of Measures ~ Interitem consistency reliability. ~ Split-Half reliability. l 6
Test-Retest Reliability l The reliability coefficient obtained with a repetition of the same measure on a second occasion. 7
Parallel-Form Reliability l When responses on two comparable sets of measures tapping the same construct are highly correlated. 8
Internal Consistency of Measures l Indicative of the homogeneity of the items in measure that tap the construct. 9
Internal Consistency Reliability l A test of the consistency of respondents’ answer to all the items in a measure. 10
Spilt-Half Reliability l Reflects the correlations between two halves of an instrument. 11
Validity Content validity. l Criterion-Related validity. l Constract validity. l 12
Content Validity l A function of how well the dimensions and elements of a concept have been delineated. 13
Content Validity (cont. ) l Face validity is considered by some as a basic and a very minimum index of content validity. 14
Criterion-Related Validity l Established when the measure differentiates individuals on a criterion it is expected to predict. 15
Criterion-Related Validity (cont. ) l Concurrent validity or predictive validity is established when the scale discriminates individuals who are known to be different. 16
Construct Validity Convergent validity. l Discriminant validity. l 17
Convergent Validity l Established when the scores obtained by two different instruments measuring the same concept are highly correlated. 18
Discriminant Validity l Established when, based on theory, two variables are predicted to be uncorreleted, and the scores obtained by measuring them are indeed empirically found to be so. 19
Examples of Some Measures from management research. l Measures from marketing research. l 20
Measures from Management Research l l l Job involvement. Participation in decision making. Role conflict. Career salience. Least preferred coworker scale. 21
Measures from Marketing Research Index of consumer sentiment toward marketing. l Servqual-P battery. l Role ambiguity (salesperson). l 22
Servqual-P Battery l l To assess the quality of service rendered: Reliability. Responsiveness. Personalization. Tangibles. 23
- Slides: 23