RATING SCALES - www.muthupandi.co.in

advertisement
RATING SCALES
By
M. MUTHU SASI REKHA
SYNOPSIS











Introduction
Definitions
Rating
Principles of defining rating scale
Types of rating scales
Personality Profile
Sources of errors in rating scales
Merits of rating scales
Defects of rating scales
Improving the rating scales
Conclusion
INTRODUCTION
 Rating scale is an important technique of
evaluation. Rating is the assessments of a
person by another person. This is one of the
oldest methods of personality assessment.There
are certain general approaches to assess
personality like holistic or overall approach,
projective test approach and trait approach. In
this rating scales and inventories come under
the trait approach.
DEFINITIONS
 Barr & others define rating as: “Rating is a term
applied to expression of opinion or judgement
regarding some situation, object or character.
Opinions are usually expressed on a scale of
values. Rating techniques are devices by which
such judgements may be quantified”.

A rating scale is a device by which the
opinion concerning a trait can be systematized.
RATING
 Three point scale
Above average / Average / Below average
 Five point scale
Excellent / Very good / Good / Average / Poor
 Seven point scale
PRINCIPLES OF DEFINING A
RATING SCALE



The characteristics should be clearly
defined
The characteristics should be readily
observable
Degrees of the characteristics must
be defined
TYPES OF RATING
SCALES
 Numerical Rating scale
 Descriptive Rating Scale
 Graphical Rating Scale
 Score cards
 The rank Order Scale
 Method of Paired comparisons
 Man-To-Man Scale
Numerical Rating scale
In which numbers are assigned to each trait. If it is a seven
point scale, the number 7 represents the maximum amount
of that trait in the individual, and 4 represents the average.
The rater merely enters the appropriate number after each
name to indicate judgement of the person.
A
1
B
2
C
3
D
4
E
5
F
6
G
7
Descriptive Rating Scale
In which descriptive phrases or terms assigned
to each trait. The rater enters the appropriate
phrase after each name to indicate judgement
of the person.
A
B
C
D
E
Excellent Good
Average
Below average
poor
Graphical Rating Scale
A straight line, may be represented by
descriptive phrases at various points. To
rate the subject for a particular trait a check
mark is made at the particular point.
Low Moral Good Moral High Moral
0
25
50
75
100
Score cards
 It is a type of scale in which whatever is
being rated is analysed into its component
parts. An expert assigns each part of a
maximum score. The rater assign a value
to each item as he passes judgement, and
these values are totaled a final score is
pronounced.
The Rank Order Scale
 In this type the judge is simply required
to place the people being rated in a rank
order from high to low on the attitude or
opinion in question. A given individual’s
scale position is given in relation to other
people in the sample. The units of the
scale are unequal.
Method of Paired comparisons
 In which the rater compares each person
being rated with respect to the trait of
every other individual, being rated in the
general terms of equal better or worse.
Man-To-Man Scale
 In this case, an individual is asked to
rate the person to be rated (the rate) by
comparing him to a person already rated
and assigned a position on the scales.
The rate is assigned his position.
PERSONALITY PROFILE /
PSYCHOGRAPH
 When rating have been obtained on
several traits of the same individual they
may be combined in a diagram known as
the Personality profile.

A number of rating scales of the same
individual in a variety of traits may be
combined in the form of a psychograph
or personality profile.
PERSONALITY PROFILE (OF BANK CASHIER)
+3
+2
+1
Average
-1
Anxiety
Honesty
-3
Persistance
-2
SOURCES OF ERRORS IN
RATING SCALES
There are several common sources of error in
rating scales. All these sources affect the
validity of a rating. Errors may be due to :
1. Ambiguity
2. The personality of the rater - Halo effect
- Generosity error
- Severity error - Central tendecy
3. Logical error
4. Attitude of the rater.
5. Oppurtunity for adequate abservation.
SOURCES OF ERRORS IN
RATING SCALES – continued…
1.Ambiguity:
This refers to the wording and meaning of
the traits that are measured, e.g., To one rater,
aggressiveness may be a positive trait suggesting self
– assertion. To another it may connote hostility. The
term such as honest, effective and fruitful learning,
intelligent citizen, personality, ideal character etc.
must be clarified. Unless all pupils are rated on the
same attributes, the rating will be invalid and
unreliable.
SOURCES OF ERRORS IN
RATING SCALES – continued…
2. Personality of the Raters:
The Halo effect
Generosity error
Severity error
Central tendency error
SOURCES OF ERRORS IN
RATING SCALES – continued…
3. Logical error:
A logical error is closely related to the halo
effect, but is not due to personal bias. It occurs
when two traits, such as intelligence and socioeconomic status are closely related and the rater is
influenced in his rating of one by the presence or
absence of the other intelligent persons posses a
high socio-economic status.
SOURCES OF ERRORS IN
RATING SCALES – continued…
4. Attitude of Raters:
Accurate observation is a very time-consuming
process. Unless the raters truly believe that there is
some value to be derived from ratings, they may
consider them only as another administrative process
and not to do a conscientious job.
SOURCES OF ERRORS IN
RATING SCALES – continued…
5. Opportunity for Adequate Observation:
This is the most serious error. The error is
committed when the rater does not know well
enough the pupil he is rating. The only reasonable
thing to do is to refuse to rate the pupil on those
traits about which you have little or no knowledge.
MERITS OF RATING SCALES
Less subjective than casual observation,
An analytical judgement.
Helps comparison of individuals.
Stimulates person being rated.
DEFECTS OF THE RATING SCALES
Error of Lenience or severity
Error or central tendency
Halo effect
IMPROVING THE RATING SCALE
i) Identify educationally significant traits.
ii) Clearly define the traits to be rated and the scale
points to be used.
iii) Avoid technical jargon. If slang will help convey
the intent, use it by all means.
iv) Express the traits to be rated as question rather
then as declarative statements.
v) If the line showing the continuum is used, it
should follow immediately after the questions,
vi) Determine how discrimination you want the
ratings and divide the continuum accordingly.
(Three to seven
intervals).
CONCLUSION
In order to minimize the problem semantics and
to make Rating Scales more useful, it is necessary to
observe certain established principles. Firstly, the
trait must be clearly defined. Secondly, the degrees of
the trait must be clearly defined. Thirdly, the rater
may be asked to quote instances in support of his
judgement. Fourthly, the rater should be instructed
not to record anything on the scale. Fifthly, the rater
should be instructed to avoid Halo-Effect and finally,
it is better to obtain ratings of a particular person
rather than from more than one judge.
Download