GROUP 6

SAMIYA
DANIYA
WALI
KSHITIJ
SIFAT
RICHITA
CHAPTER 13

MEASUREMENT SCALES
Page: 330-336
 THE NATURE OF ATTITUDES
 An attitude is a learned, stable predisposition to respond to oneself.
 As attitude is predisposition, the more likely that the product or service will be
purchased.
 Examples of Myra
1. She is convinced that mind writer has has great talent, terrific product and
superior opportunities for growth.
2. She loves working at mind writer.
3. She expects to stay with the firm and work hard to achive rapid promotion for
greater visibility and influence.
 SELECTING A MEASUREMENT SCALE
 Research objectives
 Response types
 Data properties
 Number of dimensions
 Balanced or unbalanced
 Forced or unforced choices
 Number of scale points
 Rater errors
 RESEARCH OBJECTIVES

 These are too numerous to list

 Researchers however face two general types of scaling objectives:

 To measure characteristics of the participants who participate in the study.

 To use participants as judges of the objects or indicants presented to them.

 RESPONSE TYPES
 Here measurement scales fall into one of four general types:
 Rating- participants score an object or indicant without making a direct
comparison to another object or attitude.
 Ranking- constrain the study participant to making cpmparisons and
determining order among two or more properties, or their indicants or objects.
 Categorization- asks participants to put themselves or property indicants in
groups or categories.
 Sorting- requires that participants sort cards(concepts or constructs) into piles
using criteria established by the researcher.
 DATA PROPERTIES
 Decision about measurement of scale is often made with regard to the data
properties generated by each scale
 Scales are classified in increasing order of:
 Power
 Nominal scales- classify data into categories without indicating order, distance or
unique origin.
 Ordinal scales- show relationship of more and less than but have no distance or
unique origin.
 Interval scales- have both order and distance but no unique origin.
 NUMBER OF DIMENSIONS

• There are two types unidimensional and multidimensional

 BALANCED OR UNBALANCED
• Balanced rating scale - has equal number of categories above and below the
midpoint .
• Balanced scale may take the form of
 “very good – good -average – poor – very poor”
• Unbalanced rating scale- has unequal number of favorable and unfavorable
response choices
• Unbalanced scale may take the form of
 “Poor – fair – average – good – very good- excellent ”
FORCED OR UNFORCED CHOICES
 Unforced choice rating scale – provides participants with to express no
opinion when they are unable to make a choice among the alternative
offered.

 Forced rating scale – require the participant select one of the offered
alternative .

 Researched can exclude the response choice “no opinion” , “undecided” ,

“don’t know”
 NUMBER OF SCALE POINTS
 A scale should be appropriate for its purpose
 For a scale to be useful
a. It should match the stimulus presented and extracted information
proportionate to the complexity of the attitude object , concept or construct.
 A 3-Piont scale (Better than average– Average ---- Worde than average ) is
probably sufficient for a deodorant, a fast food burger, gift-wrapping or a snack.
There is a little support for choosing a scale with 5 or more than points in this
instance.
 When the product is complex, plays an important role in the consumer’s life and
is costly a scale with 5 to 11 points should be considered.
 RATER ERRORS
 The value of rating scale depends on the assumption that a person
can and will make good judgement.
 These errors most often occur when the rater does not know the
object or property being rated.
1. Halo Effect: A mistake or bias that can occur in evaluating an
individual's performance where they are consistently rated based on the
evaluator's overall impression, rather than on their actual performance.
2. Central Tendency Error: Central tendency error is the raters’ tendency
to avoid making “extreme” judgments of employee
performance resulting in rating all employees in the middle part of a
scale.
3. Leniency Error: Leniency error is when a raters’ tendency is to rate all
employees at the positive end of the scale (positive leniency) or at the
negative end of the scale (negative leniency). when a manager over-
emphasizes either positive or negative behaviors.
RATING SCALES

 It is a set of categories designed to elicit information about a

quantitative or a qualitative attribute.