RM4 Rating scales

What are rating scales particularly useful for

measuring attitudes

6 different formats of rating scales

dichotomous/multichotomous, Likert scale, graphic rating scale, non-verbal scale, ranking scale, semantic differential scale

What is dichotomous scale

A scale with two response options, e.g., Yes/No.

What is multichotomies scale

one response option of multiple response options

What is a Likert scale?

A multi-point scale measuring degrees of agreement or opinion, typically with 5 or 7 points.

4 considerations for Likert scales

response acquiescence, double negatives, verbal response and neutral responses

When are non-verbal rating scales useful?

For children or cognitively impaired individuals.

What’s is occasionally necessary to include in non-verbal rating scales

labels to more clearly define the meanings of each category

What does a ranking scale measure?

The relative importance of several items.

How does a graphic rating scale work

mark along a continuous line which is anchored at each end

What does a semantic differential scale measure?

Attitudes indirectly, using bipolar opposite adjectives (e.g., "Valuable" vs. "Wasteful").

5 ways to maximise response rates

short, clear, pre-paid envelope, reminder and incentives

Why are multiple items often used for one variable?

to improve reliability and ensure different facets of the construct are measured.

How are variable scores calculated in questionnaires?

By totalling or averaging responses across items.

Why should the number of response options be carefully chosen?

Too few options reduce sensitivity; too many result reliability

Two criteria used to assess quality

temporal consistency and internal consistency

What is temporal consistency

the extent to which our tool would provide the same results under the same conditions

What is internal consistency

the extent to which our tool is measuring the construct we are interested in

What is construct validity?

Whether the construct being measured is valid, supported by cumulative research evidence.

How can construct validity be assessed in the short term?

Through convergent validity (correlates with related constructs) and discriminant validity (doesn't correlate with different constructs).

What are psychometric tests?

Standardized tools to measure traits or abilities

What key issue can arise with psychometric tests?

Reliability is often well established, but validity may be questionable.