Scaling Techniquea- Likert

7/30/2019 Scaling Techniquea- Likert

1/26

SCALING TECHNIQUES

Sem - II


2/26

LIKERT SCALE


3/26

Dr. Rensis Likert

(1903 - 1981)

The original idea for the likert scale is found in Rensis Likerts 1932 article inArchive of psychology titled A technique for the measurement ofAttitudes. This idea was expanded by Likerts 1934 Journal of social

psychology article titled A simple and Reliable method of scoring theThurstone Attitude Scales.


4/26

WHATIS LIKERTSCALE?

It is a psychometric scale commonly involved in research thatemploys questionnaires.

It is the most widely used approach to scaling responses in surveyresearch.

Likert scales are a non-comparative scaling technique and are one-dimensional in nature.

When responding to a Likert questionnaire item respondents specifytheir level of agreement or disagreement on a symmetric agree-disagree scale for a series of statements.

Thus, the range captures the intensity of their feelings for a givenitem, while the results of analysis of multiple items reveals a patternthat has scaled properties of the kind Likert identities


5/26

FIVEPOINT LIKERTITEM

Likert Scale

Difference

Likert item


6/26

Likert item is considered symmetric or balanced because there are equalamounts of positive and negative positions.

Often five ordered response levels are used, although many

psychometricians advocate using seven or nine level, a recent empiricalstudy found that a 5 or 7 point scale.

1. STRONGLYDISAGREE2. DISAGREE3. NEITHERAGREENORDISAGREE4. AGREE5. STRONGLYAGREE

The format of a typical five-level Likertitem


7/26

EXAMPLE:

Codes: 1-strongly disagree, 2- disagree, 3- neutral, 4-agree, 5- strongly agree

S/N Affirmative perception statements Codes

(i) OPAC / Web OPAC was Easier to use than I expected 1 2 3 4 5

(ii) It was Fun to use 1 2 3 4 5

(iii) It was Easy to use 1 2 3 4 5

(iv) It helped me in finding the documents faster 1 2 3 4 5

(v) It is Very difficult to use 1 2 3 4 5

(vi) It is Very confusing to use 1 2 3 4 5

(vii) I found more items than expected 1 2 3 4 5

(viii) I am comfortable with simple search 1 2 3 4 5

(ix) I am comfortable complex/Advance search 1 2 3 4 5

(x) I am comfortable quick search 1 2 3 4 5

(xi) I am comfortable when using OPAC/Web OPAC 1 2 3 4 5

Q.18.Please measure the following affirmative perceptions about your library OPAC

and Web OPAC use .


8/26

Codes: 1-strongly disagree, 2- disagree, 3- neutral, 4-agree, 5- strongly agree

S/N Parameters Codes

(i) I access OPAC/Web OPAC stand alone system 1 2 3 4 5

(ii) I access OPAC/Web OPAC library premises 1 2 3 4 5

(iii) It is easy to be familiar with this OPAC/Web OPAC 1 2 3 4 5

(iv) The OPAC/Web OPAC should have more flexible

interfaces

1 2 3 4 5

(v) Library searching will be easier and faster with the Web

OPAC

1 2 3 4 5

(vi) It is easy to read information provided in the Web OPAC 1 2 3 4 5

(vii) A OPAC/Web OPAC search by author is easy 1 2 3 4 5

(viii) A OPAC/ Web OPAC search by call number is easy 1 2 3 4 5

(ix) OPAC/ Web OPAC scanning through a long display

(forward or backward) is easy

1 2 3 4 5

(x) OPAC/ Web OPAC reducing the result when too much is

retrieved is easy

1 2 3 4 5

Q.19. Please mark your appreciation towards the use of your library OPAC/Web

OPAC.


9/26


10/26

The format of a typical Seven-level Likert item


11/26


12/26


13/26

ANALYSIS METHODS

Depending on how the Likert scale questions are treated a number ofdifferent analysis methods can be applied

1. Analysis methods used for individual questions (ordinal data) Bar charts and dot plots

Not histograms (data is not continuous) Central tendency summarised by median and mode

Not mean Variability summarised by range and interquartile range

Not standard deviation Analysed using non-parametric tests

(difference between the medians of comparable groups) Mann- whitney U test Wilcoxon signedrank test Kruskalwallis test


14/26

2. When multiple Likert question responses are summed together (intervaldata)

All questions must use the same Likert scale

Must be a defendable approximation to an interval scale (i.e. codingindicates magnitude of difference between items but there is noabsolute zero point) All items measure are single latent variable (i.e. a variable that is notdirectly observed, but rather inferred from other variables that areobserved and directly measured)

Analyzed using parametric tests Analysis of variance (ANOVA)

3. Analysis methods used when reduced to nominal level of agree vs.disagree

Chisquare test Cochran Q testMcNemar test


15/26

ADVANTAGES

Item analysis increases the degree of homogeneity or internal

consistency in the set of statements. Subjects generally find it easy to respond because they have a wide

range of answers(usually five) to choose from instead of only twoalternative responses, i.e., agree or disagree.

No outside group of judges is involved in selecting statements andgiving values to them.

LIMITATIONS

Ties in ranks occur quite frequently.

The response pattern of an individual is not revealed.

A respondent is required to answer all questions on the scale.

A problem of interpretation arises with this type of scale.

In this scale all statements of a universe are deemed to be of

equal attitude value.


16/26

CONCLUSION

A summated rating scale is a set of attitude statements all of whichsubjects respond with degrees of agreement or disagreement carryingdifferent scores. These scores are summed or summed and averaged toyield an individuals attitude score. The objective is to avoid the use ofonly a single statement to represent a concept and instead to useseveral statement as indicators, all representing different facets of theconcept to obtain a more well rounded perspective.


17/26

GUTTMAN SCALING


18/26

GUTTMAN SCALING

Also known as

Scalogram analysis

Cumulative scaling

Purpose:

Establish a one-dimensional continuum

Perfectly predict item responses from total score Seldom perfect in practice


19/26

GUTTMAN SCALING

Work quicklyand

effectivelyunder

pressure49

Organize thework when

directions arenot specific.

39

Decide how tomanage

multiple tasks.20 Manage resources effectively.

4

Start with a large set of items that you think all reflect thesame construct.

Work quicklyand

effectivelyunder

pressure49



39

Decide how to

managemultiple tasks.

20 Manage resources effectively.4

Work quicklyand

effectivelyunder

pressure49



39

Decide how tomanage


4Work quickly

and

effectivelyunder

pressure49Organize the

work whendirections arenot specific.

39

Decide how tomanage

multiple tasks.20

Manageresourceseffectively.

4

Work quicklyand effectivelyunder pressure

49



39

Decide how to

managemultiple tasks.

20 Manage resources effectively.4

Work quicklyand

effectivelyunder

pressure49Organize the work

when directions arenot specific.

39

Decide how tomanage


4


49



39

Decide how tomanage

multiple tasks.20


4


49



39

Decide how tomanage

multiple tasks.20


4

Work quicklyand

effectivelyunder

pressure49



39

Decide how tomanage


4

Work quicklyand

effectivelyunder



39

Decide how tomanage


4

Work quicklyand

effectivelyunder



39

Decide how tomanage

multiple tasks.20

Manage resourceseffectively.

4


20/26

GUTTMAN SCALING

Have many judges rate each statement with respect to whetherit is favorable with respect to the construct -- a yes/norating.


21/26

GUTTMAN SCALING

Lay out a table:


22/26

GUTTMAN SCALING

Lay out a table:

Item Item Item Item Item Item

Respondent 2 7 5 3 8 ...

7 Y Y Y Y Y Y15 Y Y Y -- Y --

3 Y Y Y Y -- --

29 Y Y Y Y -- --

19 Y Y Y -- -- --

32 Y Y -- Y -- --

41 Y Y -- -- -- --

6 Y Y -- -- -- --

14 Y -- -- Y -- --

33 -- -- -- -- -- --


23/26

GUTTMAN SCALING

Lay out a table:



7 Y Y Y Y Y Y15 Y Y Y -- Y --

3 Y Y Y Y -- --

29 Y Y Y Y -- --

19 Y Y Y -- -- --

32 Y Y -- Y -- --

41 Y Y -- -- -- --

6 Y Y -- -- -- --

14 Y -- -- Y -- --

33 -- -- -- -- -- --

When sorted by row and column it

will show whether there is

a cumulative scale.


24/26

GUTTMAN SCALING

Lay out a table:



7 Y Y Y Y Y Y15 Y Y Y -- Y --

3 Y Y Y Y -- --

29 Y Y Y Y -- --

19 Y Y Y -- -- --

32 Y Y -- Y -- --

41 Y Y -- -- -- --

6 Y Y -- -- -- --

14 Y -- -- Y -- --

33 -- -- -- -- -- --

When sorted by row and column it

will show whether there is

a cumulative scale.

Note the exceptions.


25/26

GUTTMAN SCALING

There are several statistical techniques for examining the tableto find a cumulative scale.

Because there is seldom a perfectly cumulative scale, weusually have to test how good it is.

These statistics also estimate a scale score value for each item.

HOW DOES RESPONDENT USE?

For selected items (the final scale), they simply answer whetherthey agree or not (yes/no).

Since each item has a scale value, we can add these values.

Total score is the sum of the scale items respondent agreed to.


26/26

THANKYOU

Documents

Scaling Techniquea- Likert