Rating Scales 2003



A rating scale is a set of categories designed to elicit information about a quantitative or a qualitative attribute.



In psychometrics, rating scales are often referenced to a statement which expresses an attitude or perception toward something.



The basic feature of any rating scale is that it consists of a number of categories. These are usually assigned integers. An example of the use of a Likert scale:

 

Statement: I could not live without my computer. Response options: 1. Strongly Disagree 2. Disagree 3. Agree 4. Strongly Agree

 Thurstone method of equal appearing intervals  Guttman cumulative Scale(Scalogram)  Likert method of summated rating  Semantic differential  Graphic rating  Magnitude scaling

 Items are formed

 Panel of experts assigns values from 1 to 11 to each item  Mean or median scores are calculated for each item  Select statements evenly spread across the scale

Attitude towards Contraception How favourable

Value on 11-

Item

point scale Least

Neutral

1.3

Practising contraception should be punishable by law.

3.6

Contraception is morally wrong in spite of possible benefits.

5.4

Contraception has both advantages and disadvantages.

7.6

Contraception is a legitimate health measure.

9.6

Contraception is the only solution to many of our social problems.

Most

10.3

We should not only allow but enforce limitation on family size



Also known as Scalograms



Both the respondents and items are ranked



Cutting points are determined (GoodenoughEdwards technique)



Coefficient of Reproducibility (CReg) - a measure of goodness of fit between the observed and predicted ideal response patterns



Keep items with CReg of 0.90 or higher

Items on a Guttman Cummulative Scale GUTTMAN SCALE How acceptable Least

Statement housing Attitude towards mixed-ethnic

Generally speaking, people should be able to live anywhere they want. Real estate agencies should not discriminate against minority groups. The local council should actively support the idea of open housing. There should be a local review board that would rule on cases of extreme discrimination in housing.

Most

There should be laws to enforce mixed-ethnic housing.

Are you favour of having nuclear power plants in Britain?

1 STRONGLY APPROVE

2

3

4 NEUTRAL

5

6

7 STRONGLY DISAPPROVE

Semantic Differential Scale Osgood presented the concept directly and asked the individual to react to it through the use of a number of scales bounded by bipolar adjectives

SEMANTIC DIFFERENTIAL SCALE

Nuclear power GOOD

BAD

STRONG

WEAK

FAST

SLOW



Respondents mark their response on a continuum

Taste of my soft drink is:

Very important

Unimportant X

 

Invented in 1935 by Stephenson (physicist/psychologist) Aim to account for subjectivity e.g ◦ appraisal of health care ◦ political attitudes ◦ anything that is difficult to quantify and enumerate



Free software packages (e.g. Qmethod, Atkinson, 1992) are available via the web. http://www.rz.unibw-muenchen.de/~p41bsmk/qmethod/ http://www.qmethod.org/



Q-sort is the basis of Q Method i.e.

◦ obtain responses to a statement from strongly agree (+4) to strongly disagree (-4) ◦ rank order the responses.



Typically, in a Q methodological study people are presented with a sample of statements about some topic, called the Q-set.



Respondents, called the P-set, are asked to rankorder the statements from their individual point of view, according to some preference, judgement or feeling about them.



By Q sorting people give their subjective meaning to the statements, and by doing so reveal their subjective viewpoint (Smith 2001) or personal profile

● Create Nine Columns 44 Total Needs, Prioritized from 1 to 9 10 9 8 7 6 5 4 3 2 1 0

9

10

5 1

Number of Individual Needs Per Column

9 5

2

2

1

Highest

Thir d

Fif th

Seventh

Lowest

Pr ior ity

Highest

Highest

Highest

Pr ior ity

Pr ior ity

Pr ior ity

Pr ior ity

10 Needs 9 Needs

9 Needs

5 Needs

5 Needs

2 Needs

2 Needs

1 Need

Highest Priority

1 Need

Second Highest Priority

Third Highest Priority

Fourth Highest Priority

Fifth Highest Priority

Sixth Highest Priority

Seventh Highest Priority

Eighth Highest Priority

Lowest Priority

10 Needs

19

9 Needs

31

6

44

29

3

36

24

40

38

5 Needs

22

26

7

5 Needs

25

14

17

37

23

2

42

16

8

39

2 Needs

30

43

12

33

13

2 Needs

15

27

11

4

20

10

32

1

18

28

35

34

9

21

Fifth Highest Priority

Sixth Highest Priority

Seventh Highest Priority

Eighth Highest Priority

9 Needs

1 Need

41 Highest Priority

Second Highest Priority

Third Highest Priority

Fourth Highest Priority

1 Need

5 Lowest Priority



These individual rankings (or viewpoints) are then subject to factor analysis.



Correlation between personal profiles then indicates similar viewpoints, or segments of subjectivity which exist



By correlating people, Q factor analysis gives information about similarities and differences in viewpoint on a particular subject



The factors resulting from Q analysis thus represent clusters of subjectivity



Q can be very helpful in exploring tastes, preferences, sentiments, motives and goals, the part of personality that is of great influence on behaviour but that often remains largely unexplored