Measurement and Scaling Scales ⢠Scale definition
Lecture 5 – Measurement and Scaling Scales • Scale definition: the combined set of points that anchor the measurement tool • The process of assigning a set of descriptors (label, rank, number, score, etc.) to represent the range of possible responses to a question about a particular construct. • Scale properties: o Assignment – unique descriptors identify each object or level o Order – establishes ‘relative magnitudes’ between the descriptors, creating hierarchical rank-order relationships among objects o Distance – absolute differences between objects or levels (Likert scale) o Origin – scale includes a ‘true natural zero’ • Assignment – unique descriptions identify each object in a set o E.g. numbers (10, 38, 44, 18 etc.) o Colors (red, blue, green, pink) o Yes and no for questions that place objects into mutually exclusive groups • Order – establishes ‘relative magnitudes’ between the descriptors, creating hierarchical rankorder relationships among objects. o E.g. first place is better than a fourth place finish – this person is lighter than this other person • Distance – express absolute differences between objects o E.g. 6 children is 2 more than 4 children • Origin – includes a ‘true natural zero’ or ‘true state of nothing’ o Weight o Age o Times one shops at a supermarket
Attitude measurement • Attitudinal measurement is difficult because it deals with: people’s thoughts, feelings, intended behaviors and characteristics etc. • An attitude is a learned predisposition to react in some consistent manner • To measure attitudes, researchers may use the trilogy, attitude-towards-object, or the affect global approach Trilogy (tri-part) three part • Cognition o Thoughts & beliefs (measure what you know) o A person’s information about an object e.g. recall of laptop brand names • Affect o Feelings (measure how you feel) o Summarises overall feelings towards an object, e.g. like or dislike for a laptop brand. • Connation o Actions o Expectations of future behaviour toward an object e.g. likelihood to purchase a laptop brand. Likert Scale • Ordinal scale that asks respondents to indicate the extent to which they agree or disagree with a series of mental or behavioral beliefs about a given object. • Initially, five scale descriptors were used:
Strongly agree Agree Neither agree or disagree Disagree Strongly disagree § A modified Likert scales expands this set to six or seven categories Characteristics include: only summated rating scale that uses a set of agree/disagreement scale descriptors Measures cognitive components; does not measure affective or conative components Best utilised when self-administered surveys or personal interviews are used to collect data. o o o o o
• • •
Semantic Differential Scale o Semantic Differential (SD) is a type of a rating scale designed to measure the connotative meaning of objects, events, and concepts. The connotations are used to derive the attitude towards the given object, event or concept. Criteria for good measurement: RVS • Reliability – the degrees to which measures are free from random error and therefore yield consistent results (test-retest method) • Validity – the ability of a scale to measure what was intended to be measured • Sensitivity – the ability to accurately measure variability in stimuli or responses. Establishing Validity • Face or content validity – professional agreement that a scale’s content logically appears to accurately reflect what was intended to be measured • Criterion validity – the ability of a measure to correlate with other standard measures of the same construct or established criterion • Construct validity – the ability of a measure to provide empirical evidence consistent with a theorybased concept • Two dimensions: repeatability and internal consistency
Attitude measurement • Attitudinal measurement is difficult because it deals with: people’s thoughts, feelings, intended behaviors and characteristics etc. • An attitude is a learned predisposition to react in some consistent manner • To measure attitudes, researchers may use the trilogy, attitude-towards-object, or the affect global approach Trilogy (tri-part) three part • Cognition o Thoughts & beliefs (measure what you know) o A person’s information about an object e.g. recall of laptop brand names • Affect o Feelings (measure how you feel) o Summarises overall feelings towards an object, e.g. like or dislike for a laptop brand. • Connation o Actions o Expectations of future behaviour toward an object e.g. likelihood to purchase a laptop brand. Likert Scale • Ordinal scale that asks respondents to indicate the extent to which they agree or disagree with a series of mental or behavioral beliefs about a given object. • Initially, five scale descriptors were used:
Strongly agree Agree Neither agree or disagree Disagree Strongly disagree § A modified Likert scales expands this set to six or seven categories Characteristics include: only summated rating scale that uses a set of agree/disagreement scale descriptors Measures cognitive components; does not measure affective or conative components Best utilised when self-administered surveys or personal interviews are used to collect data. o o o o o
• • •
Semantic Differential Scale o Semantic Differential (SD) is a type of a rating scale designed to measure the connotative meaning of objects, events, and concepts. The connotations are used to derive the attitude towards the given object, event or concept. Criteria for good measurement: RVS • Reliability – the degrees to which measures are free from random error and therefore yield consistent results (test-retest method) • Validity – the ability of a scale to measure what was intended to be measured • Sensitivity – the ability to accurately measure variability in stimuli or responses. Establishing Validity • Face or content validity – professional agreement that a scale’s content logically appears to accurately reflect what was intended to be measured • Criterion validity – the ability of a measure to correlate with other standard measures of the same construct or established criterion • Construct validity – the ability of a measure to provide empirical evidence consistent with a theorybased concept • Two dimensions: repeatability and internal consistency
