Professional Documents
Culture Documents
Measurement and Scaling: by Prof (DR.) Harsh Vardhan 20 /11/2022
Measurement and Scaling: by Prof (DR.) Harsh Vardhan 20 /11/2022
By
Prof(Dr.) Harsh Vardhan
20 /11/2022
BRM/IIFT/Harsh/6 2
Objectives
Scales, Scales Properties
Attitude, Components of Attitude
Role of Attitude in Hypothesis Construct
Process of Attitude Measurement
Scaling Techniques
Comparative and Non - Comparative Scaling
True Score Model, Reliability & Validity
BRM/IIFT/Harsh/6 3
Measurement and Scaling
BRM/IIFT/Harsh/6 4
EXAMPLE
BRM/IIFT/Harsh/6 5
Scale
A Scale is any series of items that are progressively arranged
according to value or magnitude into which an item can be placed
according to its quantification.
A scale is a continuous spectrum or series of categories.
The purpose of scaling is to represent quantitatively, an item’s, a
person’s, or an event’s place in the spectrum.
BRM/IIFT/Harsh/6 6
Concept behind Scaling
Scales are developed based on three aspects:
Order : Numbers (categories) are ordered
Distance : Difference between numbers (categories) are ordered
and quantifiable. For our purposes and most of practical purposes
we assume this difference is same.
Origin : The series has a unique origin which is indicated by the
number zero.
BRM/IIFT/Harsh/6 7
SCALE
BRM/IIFT/Harsh/6 8
SCALE PROPERTIES
There are four Primary scales of measurements in an increasing order
of their properties:
Nominal: Uniquely Classifies
Ordinal: Preserves Order
Interval: Equal Intervals
Ratio: Natural Zero
BRM/IIFT/Harsh/6 9
Example Primary Scales of
Measurements
Scale
assigned to 7 11 3
Runners
Finish
Ordinal Rank Order of
Winner Third place Second place First place
BRM/IIFT/Harsh/6 11
Levels of Data Measurement
Nominal - In nominal measurement the values just "name" the
attribute uniquely.
No ordering of the cases is implied.
A person’s gender is nominal. It doesn’t matter whether you call them
boys vs. girls or males vs. females or XY vs. XX chromosomes.
Another example is geographic location – East , West, North & South.
Sex, Religion, Pan card number, AADHAR, Mobile number, Pin code.
Numbers are used to classify or categorize
Example: Employment Classification
1 for Educator
2 for Construction Worker
3 for Manufacturing Worker
BRM/IIFT/Harsh/6 12
Ordinal Scale
BRM/IIFT/Harsh/6 13
Ordinal Scale Properties
Uniquely classifies
Preserves order
Arrangement of objects is according to their magnitude.
Ordinal - A variable is ordinal measurable if ranking is possible for
values of the variable
Example
Excellent , Good, Fair (importance of object)
Excellent is better than Good but we do not know by how much.
A gold medal reflects superior performance to a silver or bronze medal in the
Olympics. But you can’t say a gold and a bronze medal average out to a silver
medal.
Preference scales are typically ordinal – how much do you like this cereal?
Like it a lot, somewhat like it, neutral, somewhat dislike it, dislike it a lot.
BRM/IIFT/Harsh/6 14
Ordinal Level Data
BRM/IIFT/Harsh/6 15
Ordinal Data
1 2 3 4 5
BRM/IIFT/Harsh/6 16
Interval Scale
A scale in which the numbers are used to rate objects such that
numerically equal distances on the scale represent equal distance in
the characteristic being measured.
An Interval Scale contains all information of an Ordinal Scale ,but it also
allows to compare the differences between objects.
Distances between consecutive integers are equal
Relative magnitude of numbers is meaningful
Differences between numbers are comparable
Location of origin, zero, is arbitrary
BRM/IIFT/Harsh/6 17
Interval Level Data
BRM/IIFT/Harsh/6 18
Interval Scale Properties
Uniquely classifies
Preserves order
Equal intervals
Consumer Price Index (base 100)
Fahrenheit temperature
F=(9/5)*C+32
Location of zero point is not fixed .Both zero point and unit of
measurement are arbitrary.
Any linear transformation y=a+bx will preserve properties of scale, x
is original scale value ,y is transformed scale value.
Two interval scales that rate objects A,B,C,D as 1,2,3,4 or 22,24,26 and 28
are equivalent. (use a=20,b=2).Zero has no value. Ratio between D to B
values changes from 2:1 to become 7:6
BRM/IIFT/Harsh/6 19
Ratio Scale
It is the highest scale as it possesses properties of Nominal, Ordinal
and Interval scale in addition ,an absolute zero point.
It allows researcher to identify or classify objects, rank order the
objects and compare intervals or difference.
It is also meaningful to compute ratios of scale values.
The difference between 2 and 5 is the same as the difference
between 14 and 17 but also 14 is 7 times large as 2 in the absolute
terms
Example of ratio scale: height, weight, age and money
In marketing sales, costs, market share, and number of customers
are variables on a ratio scale.
It allows transformation of y=bx, b>0
BRM/IIFT/Harsh/6 20
Ratio Level Data
BRM/IIFT/Harsh/6 21
Ratio Scale Properties
Uniquely Classifies
Preserves Order
Equal Intervals
Natural Zero
Weight and Distance
Temperature (Kelvin scale)
Used for Quantitative data
BRM/IIFT/Harsh/6 22
Mathematical and Statistical Analysis of Scales
Discrete Measures
Measures that can take on only one of a finite number of values.
Continuous Measures
Measures that reflect the intensity of a concept by assigning
values that can take on any value along some scale range.
BRM/IIFT/Harsh/6 23
Qualitative & Quantitative Data
Nominal & Ordinal data is also called Non Metric or Qualitative data.
Interval & Ratio data is called Metric or Quantitative data.
Two major types of inferential statistics are:
Parametric Statistics & Non - Parametric Statistics
Parametric Statistics uses Interval & Ratio data
Nonparametric Statistics uses Nominal & Ordinal data
BRM/IIFT/Harsh/6 24
Types of Variables
A variable is a characteristics of an individual or object that is capable of
taking different values and can be measured.
Examples: ROI, Stock price, EPS, Total Sales, Age of worker, Monthly salary, Time
spent on internet.
There are two types of variables-Qualitative & Quantitative.
Qualitative Variables are those variables which differ in kind rather than
degree. These could be measured on nominal or ordinal scale.
The Nominal scale indicates categorizing into groups or classes.
Example- gender, religion, race, color & occupation
The Ordinal scale indicates ordering of items.
Example agreement disagreement scales(Likert Scale)
BRM/IIFT/Harsh/6 25
Qualitative data could be dichotomous in which there are two categories
(gender) or multinomial in there are more than two variables
(geographic region).
Quantitative Variables are those variables which differ in degree
rather than kind. These can be measured on interval or ratio scale.
Interval scales indicates rank and distance from a arbitrary zero.
Example- temperature, examination scores.
Ratio scale indicates rank & distance from a natural zero
Example- height, monthly consumption, annual budget.
SPSS does not differentiate between interval & ratio scale and lists them
under label Scale.
BRM/IIFT/Harsh/6 26
Illustration of Primary Scales of Measurement
BRM/IIFT/Harsh/6 27
BRM/IIFT/Harsh/6 28
Concept Map for Primary Scale
BRM/IIFT/Harsh/6 29
Attitude
An enduring disposition to consistently respond in a given manner.
Components of attitudes:
Affective Component
The feelings or emotions toward an object
Cognitive Component
Knowledge and beliefs about an object
Behavioral Component
Predisposition to action
Intentions
Behavioral expectations
BRM/IIFT/Harsh/6 30
Attitudes as Hypothetical Constructs
BRM/IIFT/Harsh/6 31
What is Construct?
BRM/IIFT/Harsh/6 32
Techniques for Measuring Attitudes
Ranking
Requiring the respondent to rank order objects in overall performance on the basis of a
characteristic or stimulus.
Rating
Asking the respondent to estimate the magnitude of a characteristic, or quality, that an
object possesses by indicating on a scale where he or she would rate an object.
Sorting
Presenting the respondent with several concepts typed on cards and requiring the
respondent to arrange the cards into a number of piles or otherwise classify the concepts.
Choice
Asking a respondent to choose one alternative from among several alternatives; it is
assumed that the chosen alternative is preferred over the others.
BRM/IIFT/Harsh/6 33
The Attitude Measuring Process
The process comprises of four elements:
BRM/IIFT/Harsh/6 34
Ranking
Ranking task requires that the
respondent rank order a small number of
objects in overall performance on the
basis of some characteristic or stimulus.
BRM/IIFT/Harsh/6 35
Ranking/Rating Comparisons:
BRM/IIFT/Harsh/6 36
Rating
Rating asks the respondent to estimate the
magnitude of a characteristic, or quality, that an
object possesses. The respondent’s position on a
scale , where he or she would rate an object.
BRM/IIFT/Harsh/6 37
Types of attitude rating scales
Single-item scales
As the name itself suggests, single-item scales are those that have only one
item to measure a construct.
E.g., itemized-category scales, rank-order scales, constant-sum scales,
pictorial scales.
Multi-item scales
Continuous scales
BRM/IIFT/Harsh/6 38
Attitude Rating Scales
Simple Attitude Scale
Requires that an individual agree/disagree with a statement or respond to a
single question.
This type of self-rating scale classifies respondents into one of two categories
(e.g., yes or no).
Example: THE PRESIDENT SHOULD RUN FOR RE-ELECTION
_______ AGREE ______ DISAGREE
BRM/IIFT/Harsh/6 39
Single-item scales
Itemized-category scales
Example 1: There are four categories from which
respondents can choose to indicate their overall level of
satisfaction with their present health insurance plan:
___ Very satisfied
___ Quite satisfied
___ Somewhat satisfied
___ Not at all satisfied
BRM/IIFT/Harsh/6 40
Single-item scales
Example 2: “We are interested in your opinions about
the medical services offered by private doctors in this
area…”
Quality of medical care provided:
__ Excellent
__ Very good
__ Average
__ Below average
__ Don’t know
BRM/IIFT/Harsh/6 41
Category Scales
BRM/IIFT/Harsh/6 42
Attitude Rating Scales (cont’d)
Category Scale
A more sensitive measure than a simple scale in that it can have more than
two response categories.
Question construction is an extremely important factor in increasing the usefulness of
these scales.
Example: How important were the following in your decision to visit Laksh Dweep? (check
one for each item)
VERY SOMEWHAT NOT TOO
IMPORTANT IMPORTANT IMPORTANT
CLIMATE ___________ ___________ ___________
COST OF TRAVEL ___________ ___________ ___________
FAMILY ORIENTED ___________ ___________ ___________
EDUCATIONAL/HISTORICAL ASPECTS ___________ ___________ ___________
FAMILIARITY WITH AREA ___________ ___________ ___________
BRM/IIFT/Harsh/6 43
Scaling Techniques
BRM/IIFT/Harsh/6 44
TYPES OF SCALES
BRM/IIFT/Harsh/6 46
Non - Comparative Scale
BRM/IIFT/Harsh/6 47
Comparative Scale
Every response is to be scaled. To extract the necessary responses, one can
use any of the following scales:
Paired Comparison Scales: a scale showing a series of only two objects as
choices.
Rank Order Scale: a scale that ranks an object, concept or person in a
certain order.
Constant Sum Scale: a scale showing how a set number of points is
distributed among two more attributes.
Q-sorting Scale: a scale in which a set of objects are distributed into piles
according to specified rated categories.
BRM/IIFT/Harsh/6 48
Non - Comparative Scale
Graphic Rating Scales: a scale showing a graphic continuum that is typically
anchored by two extremes.
Itemized Rating Scales: a scale showing a limited number of ordered
categories.
Likert Scale: a scale that shows a series of attitudes toward an object, which
are given numerical values ranging from favorable to unfavorable.
Semantic Differential Scale: a scale that rates opposite pairs of words, or
phrases on a continuum, which are then plotted as a profile or image.
Stapel Scale: a scale that provides a single description in the center, which is
usually measured by plus o minus 5 points.
BRM/IIFT/Harsh/6 49
Comparative Scales: Paired Comparison -1
Respondent is presented with two objects at a time and asked to select
one object according to some criterion. The data obtained is of ordinal in
nature.
It is most popular comparative scaling technique.
Example- A respondent may be asked whether they shop more from Spenser or
Reliance Fresh.
Paired comparison of two shampoos brand out of five brands. 5*(5-1)/2=10 pairs .It
is possible to convert paired comparison data to a rank order, the researcher can
find number of times each brand is preferred over the other by simple total.
Paired comparison is feasible and practical when the numbers of brands are
limited.
I would like to know your overall opinion of two brands of adhesive bandages. They
are MedBand and Super-Aid. Overall, which of these two brands—MedBand or
Super-Aid—do you think is the better one? Or are both the same?
MedBand is better _____
Super-Aid is better _____
They are the same _____ BRM/IIFT/Harsh/6 50
Obtaining Shampoo Preferences Using Paired Comparisons
Instructions: We are going to present you with ten pairs of shampoo brands. For
each pair, please indicate which one of the two brands of shampoo you would prefer
for personal use.
Recording Form:
Jhirmack Finesse Vidal Head & Pert
Sassoon Shoulders
Jhirmack 0 0 1 0
5*(5-1)/2=10
Finesse 1a 0 1 0
Vidal Sassoon 1 1 1 1
Head & Shoulders 0 0 0 0
Pert 1 1 0 1
Number of Times 3 2 0 4 1
Preferredb
1 in a particular box means that the brand in that column was preferred over the brand in the
corresponding row. A 0 means that the row brand was preferred over the column brand.
The number of times a brand was preferred is obtained by summing the 1s in each column.
BRM/IIFT/Harsh/6 51
Rank Order Scaling-2
A Comparative Scaling Technique in which respondents are presented
several objects simultaneously and are asked to order or rank them
according to some criterion.
An ordinal scale may be developed by asking respondents to rank order
(from most preferred to least preferred) a set of objects or attributes.
Example
Respondent may be asked to rank brands of toothpaste according to overall
preference. Respondent may assign rank 1 to most preferred brand,2 to second
most preferred and so on…..
Rank order also lead to ordinal data.
It takes less time.
Respondent understands easily instructions.
Rank order data can be converted to equivalent paired comparative data.
BRM/IIFT/Harsh/6 52
Preference for Toothpaste Brands
Using Rank Order Scaling
Instructions
Rank the various brands of toothpaste in order of preference. Begin by picking
out the one brand that you like most and assign it a number 1.
Then find the second most preferred brand and assign it a number 2.
Continue this procedure until you have ranked all the brands of toothpaste in
order of preference. The least preferred brand should be assigned a rank of 10.
No two brands should receive the same rank number.
The criterion of preference is entirely up to you. There is no right or wrong
answer.
BRM/IIFT/Harsh/6 53
Preference for Toothpaste Brands Using Rank Order Scaling
Brand Rank Order
1. Crest _________
2. Colgate _________
3. Aim _________
4. Gleem _________
5. Sensodyne _________
6. Ultra Brite _________
7. Close Up _________
8. Pepsodent _________
9. Plus White _________
10. Stripe _________
BRM/IIFT/Harsh/6 54
Constant-Sum Scale-3
Constant-Sum Scale require respondents to allocate a fixed number of
rating points (usually 100) among several objects, to reflect the relative
preference for each object.
O1 O2 O3 O4 are four objects allocated fixed number of rating points out of total
100 – O1 given 40, O2 - 30, O3 - 10, O4 - 20 .
If an attribute is unimportant ,the respondent assigns it 0 point. If the attribute is
twice as important to some attribute, it receives twice as many points.
The sum of points is 100 that is why it is known as Constant Sum Scale.
BRM/IIFT/Harsh/6 55
Constant-Sum Scale
Brand A _________
Brand B _________
Brand C _________
BRM/IIFT/Harsh/6 56
Constant-Sum Scales
Example: Please divide 100 points among the following
characteristics so the division reflects the relative importance of
each characteristic to you in the selection of a health plan.
BRM/IIFT/Harsh/6 57
Importance of Bathing Soap Attributes
Using a Constant Sum Scale
Instructions
There are eight attributes of bathing soaps. Please allocate
100 points among the attributes so that your allocation
reflects the relative importance you attach to each attribute.
The more points an attribute receives, the more important the
attribute is. If an attribute is not at all important, assign it zero
points. If an attribute is twice as important as some other
attribute, it should receive twice as many points.
BRM/IIFT/Harsh/6 58
Importance of Bathing Soap Attributes
Using a Constant Sum Scale
Form
Average Responses of Three Segments
Attribute Segment I Segment II Segment III
1. Mildness 8 2 4
2. Lather 2 4 17
3. Shrinkage 3 9 7
4. Price 53 17 9
5. Fragrance 9 0 19
6. Packaging 7 5 9
7. Moisturizing 5 3 20
8. Cleaning Power 13 60 15
Sum 100 100 100
BRM/IIFT/Harsh/6 59
Q-Sort Procedures-4
It is Comparative Scaling technique that uses a rank order procedure to
sort objects based on similarity with respect to some criterion.
It discriminates a large number of objects quickly.
Objects are sorted in to piles based on similarities with respect to some
criterion.
Example - Respondents are given 100 attitude statements on individual cards &
ask them to place in 7 piles .Ranging from most highly agreed with least highly
agreed with.
BRM/IIFT/Harsh/6 60
Non - Comparative Scaling
BRM/IIFT/Harsh/6 61
Scale Basic Characteristics Example Advantages Disadvantages
BRM/IIFT/Harsh/6 62
Continuous Rating Scale
BRM/IIFT/Harsh/6 63
Continuous Rating Scale
How would you rate Reliance as a department store?
VERSION 1
Probably the worst ……..|………………………..Probably the best
VERSION 2
Probably the worst ……..|……………………………………………Probably the best
0 10 20 30 40 50 60 70 80 90 100
VERSION 3
Very bad Neither good Very good
nor bad
Probably the worst ……..|……………………………………………Probably the best
0 10 20 30 40 50 60 70 80 90 100
BRM/IIFT/Harsh/6 64
Itemized Rating Scales
BRM/IIFT/Harsh/6 65
The Likert Scale
An extremely popular means for measuring attitudes. Named after Rensis
Likert.
Respondents indicate their attitudes by checking how strongly they agree or
disagree with statements.
Five Response alternatives: “strongly agree”, “agree”, “uncertain”, “disagree”,
and “strongly disagree”.
It is easy to construct and administer.
Disadvantage- it takes longer to complete than other itemized scale because
respondent has to read each statement.
It is more fun to play a tough, competitive tennis match than to play an easy
one.
___Strongly Agree ___Agree ___Not Sure ___Disagree ___Strongly Disagree
BRM/IIFT/Harsh/6 66
Selected Category Scales
BRM/IIFT/Harsh/6 67
Likert Scale
Each statement is assigned a numerical score ranging either from -2 to +2
or 1 to 5.
The analysis can be conducted on an item-by-item basis (Profile Analysis)
or a total (Summated) score can be calculated for each respondent by
summing across items.
Compare two stores: Profile Analysis will involve comparing two stores in
terms of average respondent ratings of each item such as quality of
merchandise, in-store service and variety of products.
In Summated approach determine the total score for each respondent on
each store. High & Low score will reflect favorable or unfavorable
response.
Summated score is quite popular and is preferred that is why Likert scale
is also known as Summated Scale.
BRM/IIFT/Harsh/6 68
Computing a Composite Scale : Trust
worthiness of a web site- page 297
BRM/IIFT/Harsh/6 70
Likert Scale For Measuring Attitudes
Toward Tennis
BRM/IIFT/Harsh/6 71
Likert Scale for Measuring Attitudes
Toward Tennis
It is more fun to play a tough, competitive Tennis match
than to play an easy one.
___Strongly Agree
___Agree
___Not Sure
___Disagree
___Strongly Disagree
BRM/IIFT/Harsh/6 72
Likert Scale for Measuring Attitudes
Toward Tennis
BRM/IIFT/Harsh/6 73
Semantic Differential
BRM/IIFT/Harsh/6 74
Semantic Differential Scales For Measuring
Attitudes Toward Tennis
Exciting ___ : ___ : ___ : ___ : ___ : ___ : ___ : Calm
BRM/IIFT/Harsh/6 75
A Semantic Differential Scale for Measuring Self- Concepts,
Person Concepts, and Product Concepts
1) Rugged :---:---:---:---:---:---:---: Delicate
2) Excitable :---:---:---:---:---:---:---: Calm
3) Uncomfortable :---:---:---:---:---:---:---: Comfortable
4) Dominating :---:---:---:---:---:---:---: Submissive
5) Thrifty :---:---:---:---:---:---:---: Indulgent
6) Pleasant :---:---:---:---:---:---:---: Unpleasant
7) Contemporary :---:---:---:---:---:---:---: Obsolete
8) Organized :---:---:---:---:---:---:---: Unorganized
9) Rational :---:---:---:---:---:---:---: Emotional
10) Youthful :---:---:---:---:---:---:---: Mature
11) Formal :---:---:---:---:---:---:---: Informal
12) Orthodox :---:---:---:---:---:---:---: Liberal
13) Complex :---:---:---:---:---:---:---: Simple
14) Colorless :---:---:---:---:---:---:---: Colorful
15) Modest :---:---:---:---:---:---:---:
BRM/IIFT/Harsh/6
Vain 76
Numerical Scales
Similar to Semantic Differential, but Numerical Scales have numbers as
response options rather than “semantic space’ or verbal descriptions,
to identify categories (response positions).
You have your automobile for about one year, please tell us how
satisfied you are with your vehicle
BRM/IIFT/Harsh/6 77
Other Scale Types
Image Profile
A graphic representation of semantic differential data for
competing brands, products, or stores to highlight comparisons.
Because the data are assumed to be interval, either the
arithmetic mean or the median will be used to compare the
profile of one product, brand, or store with that of a competing
product, brand, or store.
BRM/IIFT/Harsh/6 78
Image Profiles of Commuter Airlines versus Major Airlines
BRM/IIFT/Harsh/6 79
Stapel Scales
BRM/IIFT/Harsh/6 80
Select a plus number for words that you think describe the store accurately.
The more accurately you think the work describes the store, the larger the
plus number you should choose.
Select a minus number for words you think do not describe the store
accurately. The less accurately you think the word describes the store, the
larger the minus number you should choose;
Therefore, you can select any number from +3 for words that you think are
very accurate all the way to -3 for words that you think are very inaccurate.
BRM/IIFT/Harsh/6 81
A Stapel Scale for Measuring a Store’s Image
Department
Store Name
+3
+2
+1
Wide selection, High quality, Poor service[any one]
-1
-2
-3
BRM/IIFT/Harsh/6 82
Graphic Rating Scales
BRM/IIFT/Harsh/6 83
Summary of Advantages & Disadvantages of Rating Scales
320
BRM/IIFT/Harsh/6 84
Behavioral Differential
BRM/IIFT/Harsh/6 85
Selecting a Measurement Scale
Is a ranking, rating or choice techniques best. Ranking
provide ordinal data- limits statistical analysis.
Should a monadic or comparative scale be used?
What type of category labels, if any should be used?
How many scale positions are needed?
Should a balanced or unbalanced rating scale be chosen?
Single measure V/s Index measure
BRM/IIFT/Harsh/6 86
Monadic Rating Scale
Now that you’ve had your automobile for about 1 year, please
tell us how satisfied you are with its pickup.
Completely Very Fairly Well Somewhat Very
Satisfied Satisfied Satisfied Dissatisfied Dissatisfied
BRM/IIFT/Harsh/6 87
A Comparative Rating Scale
BRM/IIFT/Harsh/6 88
An Unbalanced Scale
How satisfied are you with the bookstore in the ABC mall?
Neither Satisfied Quite Very
Satisfied Nor Dissatisfied Satisfied Dissatisfied
BRM/IIFT/Harsh/6 89
Three Criteria for Good Measurement
Reliability Validity
Good
Measurement
Sensitivity
BRM/IIFT/Harsh/6 90
Two general traits characterize the
Scientific Method
Two important characteristics of any measurement procedure are:
Reliability
Validity
BRM/IIFT/Harsh/6 92
Reliability
Reliability is a Statistical measure of how reproducible the survey
instruments’ data are.
It is the ability of a measure to produce the same or highly similar
results on repeated administrations.
Reliability of a questionnaire relates to the consistency of responses
across retesting with the same or equivalent instrument.
In short, reliability means stability in results.
A survey is said to be a reliable one if it provides a consistent
measure of important characteristics despite background
fluctuations.
BRM/IIFT/Harsh/6 93
Reliability
BRM/IIFT/Harsh/6 94
Reliability
RELIABILITY
BRM/IIFT/Harsh/6 95
Reliability
Reliability
The degree to which measures are free from random error and
therefore yield consistent results.
An indicator of a measure’s internal consistency.(Exam page 329)
Internal Consistency
Represents a measure’s homogeneity or the extent to which each
indicator of a concept converges on some common meaning.
Measured by correlating scores on subsets of items making up a scale.
Internal Consistency Reliability is an approach for assessing internal
consistency of set of items when several items are summated in order
to form a total score for scale.
BRM/IIFT/Harsh/6 96
Internal Consistency
Split-Half Method
Assessing internal consistency by checking the results of one-half of a
set of scaled items against the results from the other half.
Split Half Reliability is a form of internal consistency reliability in which
the items constituting the scale are divided into two halves and
resulting half scores are correlated.
1. Administer the test to a large group students (ideally, over about 30).
2. Randomly divide the test questions into two parts. For example,
separate even questions from odd questions.
3. Score each half of the test for each student.
4. Find the correlation coefficient for the two halves.
BRM/IIFT/Harsh/6 97
Internal Consistency
Coefficient alpha (α) is a measure of internal consistency reliability that is
average of all possible split half coefficients resulting from different splitting of
the scale items.
It lies between 0 to 1 and value less than 0.6 indicates unsatisfactory internal
consistency reliability.
The most commonly applied estimate of a multiple item scale’s reliability.
Represents internal consistency by computing the average of all possible split-half
reliabilities for a multiple item scale.
The coefficient demonstrates whether or not the different items converge.
Ranges in value from 0 (no consistency) to 1 (complete consistency).
Generally, scales with a coefficient α:
.80 - .95: very good reliability
.70 - .80: good reliability
.60 - .70: fair reliability
below .60: poor reliability
BRM/IIFT/Harsh/6 98
Test-Retest Reliability
Test-Retest Method
Administering the same scale or measure to the same respondents at two
separate points in time to test for stability.
Represents a measure’s repeatability.
When a measuring instrument produces unpredictable results from one testing
to the next, the results are said to be unreliable because of error in
measurement.
Problems
The pre-measure, or first measure, may sensitize the respondents and
subsequently influence the results of the second measure.
Time effects that produce changes in attitude or other maturation of the
subjects.
BRM/IIFT/Harsh/6 99
Validity
BRM/IIFT/Harsh/6 101
Validity
Validity
CONCURRENT PREDICTIVE
BRM/IIFT/Harsh/6 102
Establishing Validity
The four basic approaches to establishing validity are face validity, content validity,
criterion validity, and construct validity
Face Validity
A scale’s content logically appears to reflect what was intended to be measured.
Content Validity
The degree that a measure covers the breadth of the domain of interest.
Criterion Validity
The ability of a measure to correlate with other standard measures of similar
constructs or established criteria.
Construct Validity
Exists when a measure reliably measures and truthfully represents a unique
concept.
BRM/IIFT/Harsh/6 103
Validity
Convergent Validity
Another way of expressing internal consistency; highly reliable scales contain
convergent validity.
Discriminant Validity
Represents how unique or distinct is a measure; a scale should not correlate
too highly with a measure of a different construct.
BRM/IIFT/Harsh/6 104
Reliability and Validity on Target
The differences between the two are illustrated with rifle targets
A: The shots from the older gun are scattered low reliability.
B: The shots from the newer gun are closely clustered and on target
high reliability and validity
C: The shots from a newer gun are closely clustered but off target
high reliability but low validity.
BRM/IIFT/Harsh/6 105
Sensitivity
BRM/IIFT/Harsh/6 106
Measurement Accuracy
BRM/IIFT/Harsh/6 107
True Score Model
True Score Model provides a frame work for
understanding the accuracy of measurement.
X0 = XT + XS + XR ,
BRM/IIFT/Harsh/6 108
True Score Model
BRM/IIFT/Harsh/6 109
Thanks