Chapter 3 Sampling &

Scaling

[Link] Pithawala
Sampling

[Link] Pithawala
[Link] of Sampling

Before deciding the method of data collection (observation, interview,

mail survey, face to face) , the researcher has to decide whether
he/she wants to collect data from population or sample.

[Link] Pithawala
There are two laws on which the theory of sampling is based:

1. law of statistical regularity

 "the law of statistical regularity lays down that a moderately large number of
items chosen at random from a large group are almost sure on the average to
possess the characteristics of the large group”.
 For example, if we want to find out the average income of 10,000 people, we take
a sample of 100 people and find the average. Suppose another person takes
another sample of 100 people from the same population and finds the average,
the average income found out by both the persons will have the least difference.
On the other hand if the average income of the same 10,000 people is found out
by the census method, the result will be more or less the same.
2. law of inertia of large number
 It states that, other things being equal, larger the size of the sample, more
accurate the results are likely to be.

[Link] Pithawala
 Sampling is the process of selecting a segment from the population to
obtain information and on the basis of which a judgment is made for
population.
 Example1: before buying new book at bookstore, many people read a few
pages (sample) to see whether it is of interest to them.
 Example 2 : a prospective car buyer test drives an automobile(sample) to
judge the performance of the car.
 Example 3: housewife tests a few rice pieces(sample) during cooking to
find out whether the rice is fully cooked or not.

[Link] Pithawala
 1. save time: decision makers have time frame in which to make decision
based on whatever information can be obtain in that period.
 2. To save money: the cost of gathering information is very less in sampling.
 3. To have accurate measurement: the accuracy of information is greater in
sampling than complete population survey.
 4. To perform focused Activity: sampling remains the only solution when
population contains infinite numbers.
 5. To estimate the sampling error: sampling usually enables to estimates the
sampling error and thus assists in obtaining information concerning some
characteristics of the population
 6. To provide stable result : Sufficient accuracy in the sample will provide
stable result.

[Link] Pithawala
Census survey:
 A complete study of all the elements present in population is
known as survey.
 It is time consuming and costly method

Sample survey:
A study of selected elements , which represents population’s
characteristics and later on results can be generalized.
It saves time and cost.

[Link] Pithawala
 Conditions of the Go for Go for Description
study sample census

1. Budget available small Large If budget is high then go for census survey or
else can opt for sample survey

2. Time available Short Long If large amount of time available then go for
census else go for sample.

3. Population size Large Small If target population is small then go for census or
else go for sample.

4. Variance in the Small Large In case of small variance/errors in the

characteristics(mean/median/mode etc) is
characteristics present then go for sample or else go for census.

5. Cost of sampling Low High In case of wrong selection of samples (sampling

error) is low then go for sample otherwise go for
error census

6. Cost of non sampling High Low In case of questionnaire, researcher, interviewer,

respondent error is low then go for census or
errors. else go for sample.

7. Nature of Destructive Nondestructive In case of unit/elements under measurements

are discarded after research then go for sample
measurement testing testing or else go for census.

8. Attention to Yes No In case of individual attention to elements is

required then go for samples or else go for
individual cases [Link] Pithawala census.
Target population(N):
 The entire aggregation of items from which samples will be drawn is known as target
population.

Census survey:
 A complete study of all the elements presented in the target population is known as
Census survey.

Sample survey:
 A study of selected elements which represents populations characteristics.

Sampling element:
 A sampling elements (target respondent)is the object from which the information is
sought.

Sampling unit:
 A sampling unit is the single elements or group of elements of the population is known
as sample units

Sampling frame:
 A sampling frame is the list of representing elements of population from which the
sample may be taken.(geographical boundaries)

Precision:
 It is the measure of how close an estimate is expected
[Link] Pithawala , to the true value of parameters.
Example 1 : A survey by a kitchen appliances firm wants to find the demand of its
microwave ovens.
 Population: all women/men
Example 2 : A survey by a kitchen appliances firm wants to find the demand of
its microwave ovens among the women who are above the age of 20 years and
who cook.
 population: women
 Samples: women whose age >20 and who cook.
Example 3 : A survey by a kitchen appliances firm wants to find the demand of its
microwave ovens among the women who are above the age of 20 years and
who cook at Surat city.
 population: women
 Samples: women whose age >20 and who cook.
 Sample frame: Surat
Example 4 : A survey by a kitchen appliances firm wants to find the demand of its
microwave ovens among the1000 women who are above the age of 20 years
and who cook, at Surat city during a period of 2010 to 2019.
 population: women
 Samples: selection of 1000 women whose age >20 and who cook.
 Sample frame: Surat
 Sampling size: 1000(no of respondent)
 Sampling period : 10 years (2010 to 2019)
 Sampling unit: a household of apartment or society
 Sampling elements: all women above the age of 20 and who cook ,living in
that particular house or society.
[Link] Pithawala
4. Sampling process

Seven steps for sampling process

Step 1 Defining the target population

Step 2 Specify the sampling frame

Step 3 Specifying the sampling unit

Step 4 Selection of the sampling method

Step 5 Determination of sampling size

Step 6 Specifying the sampling plan

Step 7 Execute the sampling process

[Link] Pithawala
5. Factors affecting Sample size decision

In non probability sampling procedures

• Allocation of budget
• Thumb rules and number of sub groups
• Importance of the decision
• Number of variables
• Nature of analysis
• Incidence rates
• Completion rates

In probability sampling procedures

• Formulas are used for determine sample size

[Link] Pithawala
 Essential condition to determine Sample size Decision

selection of Sample size

1. Several objectives Large sample size
2. Cost constrains Small sample size
3. Time constrains Small sample size
4. Nature of Data analysis For descriptive –small sample size
For depth analysis-large sample size
5. The extent of error Higher error- large sample size
6. The degree of confident (level of significant)desired For higher –large sample size
7. Estimation of standard deviation Based on calculation
8. Information required For more information-large sample size
9. Nature of research Exploratory-large sample size
conclusive-large sample size
10. Extent of variation Greater variation –large sample size
11. Population size Large population –large sample size
12. Importance of decision For high –large sample size
13. Completion rate For short –small sample size
14. Incident rate(no of respondent qualify) For high –small sample size
15. Sample size in earlier research on same topic Greater then previous sample size
[Link] Pithawala
 [Link] random sampling
2. Systematic sampling
Probability 3. Stratified sampling
sampling 4. Cluster sampling
5. Sequential sampling
6. Double sampling
Sampling
Techniques
1. Convenience sampling
Non 2. Judgmental sampling
Probability
3. Quota sampling
sampling
4. Snowball sampling

[Link] Pithawala
 Probability Sampling  Non Probability Sampling

 A sampling in which every member of  It involves the selection of units based

the population has a calculable and on factors other than random chance.
has equal probability of being
included in the sample  It is also known as deliberate
sampling and purposive sampling.
 each member has a fixed, known
opportunity to belong to the sample  There is no specific probability of an
individual to be a part of the sample.

[Link] Pithawala
A sampling process where each element in the target population has an equal
and known chance or probability of inclusion in the sample is called simple
random sampling.
In small populations random sampling is done without replacement.(sample
select once).

Mostly applied when:

target population size is small,
target population is homogeneous,
Sampling frame is clearly defined,
when not much information available about
[Link] population
Pithawala
Method of lottery
 Using the method of the lottery is one of the oldest methods and is a mechanical
example of random sampling. In this method, each member of the population has to
number systematically and in a consequent manner by writing each number on a
separate piece of paper. These pieces of paper are mixed and put into a box and then
numbers are drawn out of the box in a random manner.

Use of random numbers

 The use of random numbers is an alternative method that also involves numbering the
population. The use of a number table similar to the one below can help with this
sampling technique.

[Link] Pithawala
Example of simple random sampling of 10 households from a list of 40 households
We have a list of 40 heads of households. Each has a unique number, 1 through 40. We want to select 10 households
randomly from this list. Using a random number table, we select consecutive 2-digit numbers starting from the upper
left. If a random number matches a household's number, that household is added to the list of selected households. If a
random number does not match a household's number (for example, if it is greater than 40), then it does not select a
household. After each random number is used, it is crossed out so that it is never used again. We continue to select
households until we have 10.

[Link] Pithawala
 All elements in the target population are assigned numbers. Systematic sampling
involves the selection of every i th element from a sampling frame.
 Skip interval= population size/sample size
 For example if i=6 then the list starts with the number 1, then 6th , 12th 18th , etc.
 If the skip number is not rounded then make it nearest whole number.
 This method is more economical and less time consuming than simple random sampling.

[Link] Pithawala
 It is a two stage process:
Step 1: make a group of homogenous members of population.
stratified random is one where the population is divided in mutual exclusive and
collective exhaustive strata or sub group
 Here the population may be divided on the basis of age , gender , occupations,
education, region etc.

[Link] Pithawala
Step 2: In second stage, apply any of the random sampling method and select the
samples from each of the strata.

[Link] Pithawala
[Link] Pithawala
This method is comprises of two stages:
Stage1: the target population is first divided into mutually exclusive
and collective exhaustive subpopulations, or clusters.
Clustering involves grouping the population into various clusters
mostly based on geographical area.
Stage 2: after dividing population into some clusters, select entire
subgroup or cluster by using any of the random method.
It is heterogeneous within itself but homogeneous to other.
A common form of cluster sampling is area sampling.

[Link] Pithawala
[Link] Pithawala
 Once the cluster is formed the researcher can either go for one
stage, two stage or multi stage:
 Single stage: all the elements from each selected cluster are
studies.
 Two stage : 1) Researcher made clusters 2)The researcher use
random sample method to select few elements from cluster.
 Multi stage : 1) Researcher made clusters 2) made two or more
cluster units from main cluster 3) The researcher use random
sample method to select few elements from cluster.

[Link] Pithawala
 A sampling technique in which the population elements are
samples sequentially, data collection and analysis are done at
each stage and a decision is made as to whether additional
population elements should be samples or not.
 It is used to determine preferences for two competing
alternatives.
 Size of the sample is not fixed.
 Also known as lot wise sampling

[Link] Pithawala
 A sampling technique in which certain population elements
are sampled twice.
It is also known as two phase sampling.
 In first stage some basic information is collected and in
second stage few of them who are relevant for study or
necessary to be included are selected for the further data
collection.
 Example: Conduct unstructured interviews with a sample of
50. Repeat a structured interview with 30 out of 50 originally
sampled

[Link] Pithawala
 Convenience sampling attempts to obtain the sample of
convenient elements.
 The selection of sampling unit is left primarily to the interviewer’s
convenience.
 Often despondence are selected because they happen to be at
right place at right time as per the convenience of interviewer.
 Also known as grab sampling, accidental sampling, or opportunity
sampling

[Link] Pithawala
1. The most basic example of where the convenience sampling method is
used is when companies stop people at a mall or on a crowded street
to distribute their promotional pamphlets and ask questions.
2. Businesses use a convenience sampling method to gather information
about critical issues that are to be addressed almost immediately or
when a brand is collecting feedback about a particular feature or
newly launched product from the sample created using this method.
3. For instance, if you’re a marketing student who has been given a task
to get feedback on “scope of marketing in 2019”, you’d quickly create
an online survey and send a link to all the contacts on your phone,
share a link on social media and also talk to people you meet daily
face-to-face. [Link] Pithawala
 The researcher engaged in the study of university students might
visit the university canteen, library, some departments, play
ground, verandahs and interview certain number of students.

 Another example is of election study. During election times, media

personnel often present man – on – the – street interviews that are
presumed to reflect public opinion. In this sampling
representativeness is not significant.

[Link] Pithawala
[Link] Pithawala
Advantages Limitations
[Link] expensive and least time 1. selection bias is there
consuming. 2. it is not representative of any definable
[Link] sampling units are easy to population.
measure and cooperative by [Link] appropriate for marketing research
nature. project involving population inferences.
4 the findings of the sample survey cannot
be generalized to entire population , as
the samples are not representative.
5. As there is no set criteria for selecting
samples, there is a scope for research
being influenced by the bias of the
researcher.
7. It is not useful for descriptive or causal
but can used for exploratory for
generating ideas, insights or hypothesis.
[Link] Pithawala
 It is form of convenience sampling in which population elements
are selected based on the judgment of researcher.
 In this method the researcher acts as an expert, he/she exercise
judgment or expertise, select the elements to be included in the
sample because he/she believes that they are representative of the
population or are appropriate for study.

[Link] Pithawala
 TV reporters stopping certain individuals on the street in order to ask
their opinions about certain political changes constitutes the most
popular example of this sampling method. However, it is important to
specify that the TV reporter has to apply certain judgment when deciding
who to stop on the street to ask questions; otherwise it would be the case
of random sampling technique.
 Departmental stores selected to test a new merchandising display system

[Link] Pithawala
[Link] Pithawala
Advantages Limitations
1. It is a low cost sampling. 1. It is not allow direct generalization to
2. It is convenient and quick. a specific population, usually
3 .it is useful if broad because population in not defined
population inferences are explicitly.
not required. 2. It is subjective and its value depends
entirely on researcher’s judgment,
expertise and creativity.

[Link] Pithawala
 Quota sampling may be viewed as two stages restricted
judgmental sampling.
 In Quota sampling, the entire population is segmented into
mutually exclusive groups, called quota.
Stage 1: develop the quotas or categories of population elements.
The population is divided into mutual exclusive and collective
exhaustive or sub groups.
Here the population may be divided on the basis of age , gender ,
occupations, education, region etc.

[Link] Pithawala
 Stage 2: with any of the non random sampling
methods(convenience or judgment sampling) samples are drawn
from each quota.
 Quota is homogeneous within itself and heterogeneous to other.
 No two quotas are similar to each other.

[Link] Pithawala
 The number of respondents that are to be drawn from each of
several categories (quota) is specified in advance and the final
selection of respondents is left to the interviewer who proceeds
until the quota for each category is filled.

Advantages Disadvantages
1. It is a low cost sampling. 1. There is no assurance that the

2. It is convenient and quick. sample is representative.

[Link] Pithawala
 In this method , first unit is selected as random or on the basis of
interviewer's judgment.
 The selection of other additional respondents are done based on
referral bases are know as snowball sampling.
 Here population from which the sample can be drawn Is not
available. therefore, the process of sampling depends on the chain
system of referrals.

Advantage Disadvantages
1. Low cost is a clear advantage. 1. Bias is one of its disadvantages.

[Link] Pithawala
 The 'snowball' effect occurs as referrals multiply at each step. For example
if you got two referrals from each person, then starting from two people get
four more, then eight, sixteen and so on.

 Snowball sampling uses a method beloved by sales people, where

customer referrals to new prospects have particular value as the
relationship of trust and obligation between the identified person and the
referrer makes it more likely that the new person will make a purchase.

 The way that the sample is chosen by target people makes it liable to
various forms of bias. People tend to associate not only with people with
the same study selection characteristic but also with other characteristics.
This increases the chance of correlations being found in the study that do
not apply to the generalized wider population.
[Link] Pithawala
 The need to get the person to give you a referral also means that the
researcher has to form a relationship with the person and be nice to them.
This can change the study results as affective biases in both the
researcher and the target person change how they think and behave.

 With care in selection (you do not have to use every referral) and
avoiding personal bias, snowball sampling can still be a useful method,
particularly if you have no other way of reaching the target population.

 Snowball sampling is also known as chain sampling, chain-referral

sampling or referral sampling.

[Link] Pithawala
[Link] Pithawala
For example; Meeting a homeless person, interviewing that person,
and then asking him/her to introduce you to other homeless people
you might interview.
A researcher is studying environmental engineers but can only find
five. She asks these engineers if they know any more. They give her
several further referrals, who in turn provide additional contacts. In
this way, she manages to contact sufficient engineers.

[Link] Pithawala
 Key Probability Sampling Non-Probability Sampling

Probability sampling is a
Nonprobability sampling is a
sampling technique, in which
method of sampling wherein, it
the subjects of the population
Meaning is not known that which
get an equal opportunity to be
individual from the population
selected as a representative
will be selected as a sample.
sample.
Alternately
Random sampling Non-random sampling
known as
Basis of
Randomly Arbitrarily
selection
Opportunity of
Fixed and known Not specified and unknown
selection
Research Conclusive Exploratory
Result Unbiased Biased
Method Objective Subjective
Inferences [Link] Pithawala Analytical
Scaling

[Link] Pithawala
1. Definition & general idea of scaling

Measurement is the process of observing and recording the observations that are
collected as part of a research effort.

scaling means assigning numbers or other symbols to characteristics of

object according to certain pre specified rules.

• One to one correspondent between the numbers and the

characteristics being measured.

• The rules for assigning numbers should be standardized and applied

uniformly.

• Rules must not change over object or time.

[Link] Pithawala
 Measurement examples:

Example 1: To study people who attend a computer exhibition where all of the
computer’s new models are on display. You are interested in learning the
male-to-female ratio among visitors of the exhibition. You observe those enter
the exhibition area. • Record male as ‘m’ and female as ‘f’ or • Record
male as ‘1’ and female as ‘2’.

Example 2: To measure the opinion of people on several new computer models.

This can be achieved by interviewing a sample of visitors and assign their
opinions to scales ranging from Strongly Agree (1) … Neutral (3) … to
Strongly Disagree (5).
[Link] Pithawala
What is measured?
•A person’s physical properties may be stated in terms of weight, height,
posture.
• Psychological properties include attitudes and intelligence.
• Social properties include leadership, ability, class affiliation or status.

Primary scale of measurement:

[Link]
2. Ordinal
3. Interval
4. Ratio

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
 1. Nominal Scale
• In Nominal scale , numbers are given to identify persons, object or events.
The main purpose of code(number) is to identify a particular variable in
future.

• In business research ,nominal scale is used for identifying respondents,

brands, respondent’s attitudes, stores and other objects.

• Sometimes known as categorical scale, it assigns numbers to attributes for

easy identity. These numbers are however not qualitative in nature and
only act as labels.

It is a simple method among all other method.

It is least powerful level of measurement

It does not indicate order, distance or relationship and no arithmetic origin.

[Link] Pithawala
 Examples:
[Link](roll numbers) assigned to the students in a class or

2. Numbers(roll numbers) assigned to the patients in a hospital

3. Which political party are you affiliated with?

• Independent

• Republican

• Democrat

Labeling Independent as "1", Republican as "2" and Democrat as "3" does not in any way mean
any of the attributes are better than the other. They are just used as an identity for easy data
analysis.

4. Which of the following food items do you tend to buy at least once per month? (Please tick)

Okra Palm Oil Milled Rice

Peppers Prawns Pasteurized milk

[Link] Pithawala
5. Numbers(roll numbers) assigned to the players in cricket teams.

[Link] Pithawala
[Link] Pithawala
Data Analysis :
The numbers have no arithmetic properties and act only as labels.
[Link] only statistical analysis that can be performed on a nominal scale is the
percentage or frequency count. It can be analyzed graphically using a bar chart and
pie chart.

2. Hypothesis tests can be carried out on data collected in the nominal form.
H0: there is no significant difference among the variables.

[Link] Pithawala
3. The most likely would be the Chi-square test. However, it should be noted that the
Chi-square is a test to determine whether two or more variables are associated and the
strength of that relationship. It can tell nothing about the form of that relationship,
where it exists, i.e. it is not capable of establishing cause and effect.

[Link] Pithawala
4. Mode calculation

[Link] Pithawala
2. Ordinal Scale
Ordinal Scale involves the ranking or ordering of the attributes depending on the variable being
scaled.
• The attributes on an ordinal scale are usually arranged in ascending or descending order. It
measures the degree of occurrence of the variable.
• In this scale numbers are assigned to objects indicate relative extent of some characteristics
is possessed.
• An order scale represent order only. This scale does not represent the different the differences
of characteristic possessed by variables.
• It provide us the place or sequence only.
• In business research, ordinal scale are used to measured relative attitudes, opinions,
perceptions and preferences and customer satisfaction surveys. Measurement of this
judgment includes “greater than ” or “less than” judgment form of respondents.
• It uses qualifiers like very, highly, more, less, etc. to depict a degree.

[Link] Pithawala
Examples:
1. A software company may need to ask its users: How would you rate our app?

2. determine farmers' preferences among 5 brands of mobile.

Order of preference Brand

1 samsung
4 Vivo
2 nokia
3 panasonic
5 oppo

From such a table the researcher knows the order of preference but nothing about how much more one brand is
preferred to another, that is there is no information about the interval between any two brands. All of the
information a nominal scale would have given is available from an ordinal scale. In addition, positional statistics

such as the median, quartile and percentile can be determined.

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
Data Analysis :
The items in this scale are classified according to the degree of occurrence of the
variable in question.
1. We can perform statistical analysis like median and mode using the ordinal scale,
but not mean.
2. Rank order correlation
3. Non parametric analysis tests(like mann whitney, kruskal wallies, wilcoxon,
Friedman ANOVA etc)
4. Modeling techniques(like chart, flow chart, data flow diagram, activity
diagram)

[Link] Pithawala
3. Interval Scale
• The interval scale of data measurement is a scale in which the levels are ordered
and each numerically equal distances on the scale have equal interval difference

• An interval scale contains all the properties of an ordinal scale and it also allows
comparing the differences between objects.

• For example, the difference between “1” and “2” is equal to the difference between
“3” and “4”. Further, the difference between “2” and “4” is twice the difference
between “1” and “2”.

• However in an interval scale, zero point position is not fixed. Here zero point and
even the units units used for measurement are arbitrary . Zero scaling does not
mean absence of particular characteristics but that particular quantum of
characteristic has been assigned zero point.

[Link] Pithawala
With an interval scale, you not only know that a given attribute A is bigger than another
attribute B, but also the extent at which A is larger than B. Also, unlike ordinal and nominal
scale, arithmetic operations can be performed on an interval scale.

A 5 Minutes Interval Time Scale

[Link] Pithawala
 Examples of interval scales in numeric and semantic formats

1. Please indicate your views on Balkan Olives by scoring them on a scale of 5 down
to 1 (i.e. 5 = Excellent; 1 = Poor) on each of the criteria listed
Balkan Olives are:
Circle the appropriate score on each line
Succulence 5 4 3 2 1
Fresh tasting 5 4 3 2 1
Free of skin blemish 5 4 3 2 1
Good value 5 4 3 2 1
Attractively packaged 5 4 3 2 1
(a)

2. Please indicate your views on Balkan Olives by ticking the appropriate responses below:
Excellent Very Good Good Fair Poor
Succulent
Freshness
Freedom from skin blemish
Value for money
Attractiveness of packaging
(b)
[Link] Pithawala
[Link] Pithawala
Data Analysis :
Interval scale would use parametric statistical technique:
1. Mean and standard deviation
2. Corelation
3. Regression
4. ANOVA
[Link] analysis
6. Modeling techniques
Non parametric can also be used but it is less powerful than parametric techniques.

[Link] Pithawala
 4. Ratio Scale
• Ratio scale is the highest scale among the all.
• It possesses all the properties of the nominal, ordinal and interval scale.
• Ratio scale has a natural or absolute zero point, which means absence of particular
characteristics.
• Ratio scale can be used to identify or classify objects, rank the objects and compare
intervals or differences.
• It is also meaningful to compute ratios of scale values.
Eg. Ratio scales permit the researcher to compare both differences in scores and the relative
magnitude of scores. For instance the difference between 5 and 10 minutes is the same
as that between 10 and 15 minutes, and 10 minutes is twice as long as 5 minutes.
With respect to market research, the common ratio scale examples are price, number of
customers, employees, competitors, etc. It is extensively used in marketing, advertising,
and business sales.

[Link] Pithawala
Examples:
1. Which of the following Weight category do you fall in?
more than 100 kgs
81 - 100 kgs
61 - 80 kgs
40 - 60 kgs
Less than 40 kgs
2. Examples of variables which are ratio scaled include weights, lengths and times.
[Link] Pithawala
Data Analysis :
1. The ratio scale of data measurement is compatible with all statistical analysis
methods like the measures of central tendency (mean, median, mode, etc.) and
measures of dispersion (range, standard deviation, etc.).
2. Correlation
3. Regression
4. ANOVA
[Link] analysis
6. Modeling techniques
Non parametric can also be used but it is less powerful than parametric techniques.

[Link] Pithawala
[Link] Pithawala
3. Discussion of reliability, validity & sensitivity of scales
Reliability of the scale :
-Reliability refers to the extent to which a scale produces consistent result if repeated
measurement are made.

-Reliability is assessed by determining the proportion of systematic variation in a scale.

-It is done by determining the associations between scores obtained from different
administrations of the scale. If association is high, the scale yield consistent result and
therefore it is reliable.

There are three approaches to assessing reliability:

1. Test –retest reliability

Step 1: Fill up a questionnaire/take a test of a group of individual

Step 2: Then again re-test , the same test to same group at some later time.

Step 3: Correlate the 1st set scores with 2nd . If association is high then scale is reliable.

[Link] Pithawala
2. Alternative forms of Reliability
In alternative-forms reliability, two equivalent forms of the scale are constructed and the same
respondents are measured at two different times, with a different form being used each time.

3. Internal consistency Reliability

Internal consistency reliability determines the extent to which different parts of a summated scale
are consistent in what they indicate about the characteristic being measured.
i) split half reliability
In split-half reliability, the items on the scale are divided into two halves and the resulting half
scores are correlated.
ii) coefficient alpha or Cronbach ‘s alpha
The coefficient alpha, or Cronbach's alpha, is the average of all possible split-half coefficients
resulting from different ways of splitting the scale items. This coefficient varies from 0 to 1, and
a value of 0.6 or less generally indicates unsatisfactory internal consistency reliability.

[Link] Pithawala
Validity of the scale:
-The validity of the scale is defined as the extent to which differences in observe scale score
reflect the true differences among the objects on the characteristics bring measured rather
than systematic or random errors.
In particular scale, my data is appropriate in particular scale or not.
Types of validity that researcher may assess:
1) Content validity (it is subjective but systematic evaluation of content is needed)
2) Criteria validity(to check measurement scale perform as per expectation in relation to
other variables)
3) Construct validity/concurrent validity
i) convergent validity(it measures the extent to which the scale correlates with other
construct of the same construct)
ii) Discriminate validity(it measures the extent to which the scale correlates with other
construct of the different construct)
iii) Nomo logical validity(it helps to find correlation between scale and theory)
[Link] Pithawala
Sensitivity in scale:
-Sensitivity is the extent to which the scale is sensitive enough to pick up the changes.

-It must be able to detect small but critically significant changes in the phenomenon
over time.

-It may be done by comparing longitudinal measures of the same outcome.

Some other measures of the Scale:

-Practicality of the scale refers to whether the scale is suitable for use in ordinary
practice.

-Utility of the scale refers to whether the scale is able to produce useful results.

-Specificity of the scale is a measure of how accurate a test is against incorrect

[Link] cam be considered as the percentage of times a test will correctly identify
a negative result.

[Link] Pithawala
4. Types of Measurement Scale
There are two main types of measurement scales, namely; comparative scales and non-
comparative scales.
1) Comparative Scales
In comparative scaling, respondents are asked to make a comparison between one object
and the other. When used in market research, customers are asked to evaluate one
product in direct comparison to the others.
Comparative scales can be further divided into the,
pair comparison,
rank order,
constant sum ,
q-sort scales
magnitude Estimation scaling.

[Link] Pithawala
 Paired Comparison Scale
- Paired Comparison scale is a scaling technique that presents the respondents with two
objects at a time and asks them to choose one according to a predefined criterion.
- The resulting data from this scale is ordinal in nature.
- The resulting data can be analyzed in several ways out of which most common one is
calculating percentage of despondences who prefer one over another.
- With help of “transitivity to preference”, it is possible to convert paired in to rank
order data.
For example, there are 3 new features in the last release of a software product. But the
company is planning to remove 1 of these features in the new release. Therefore, the
product researchers are performing a comparative analysis of the most and least preferred
feature.
Which feature is most preferred to you between the following pairs?
Filter - Voice recorder Filter - Video recorder Voice recorder - Video recorder

[Link] Pithawala
1. Which brand of soft drink do you prefer out of the following?
a. Pepsi b. Coke
2 Which brand of soft drink do you prefer out of the following?

[Link] Pithawala
[Link] Pithawala
 Rank Order Scale:
- In rank order scaling technique, respondents are simultaneously provided with multiple
options and asked to rank them in order of priority based on a predefined criterion.
- It is mostly used in marketing to measure preference for a brand, product, or feature.
-When used in competitive analysis, the respondent may be asked to rank a group of
brands in terms of personal preference, product quality, customer service, etc.
-The results of this data collection are usually obtained in the conjoint analysis, as it
forces customers to discriminate among options.
- The rank order scale is a type of ordinal scale because it orders the attributes from the
most preferred to the least preferred but does not have a specific distance between the
attributes.

[Link] Pithawala
For example:
Rank the following brands from the most preferred to the least preferred.
Coca-Cola
Pepsi Cola
Dr pepper
Mountain Dew

[Link] Pithawala
[Link] Pithawala
 Constant Sum Scale
- Constant Sum scale is a type of measurement scale where the respondents are asked to
allocate a constant sum of units such as points, credits, money etc among the stimulus
objects according to some specified criterion.
- The constant sum scale assigns a fixed number of units to each attribute, reflecting the
importance a respondent attaches to it.
- This type of measurement scale can be used to determine what influences a customer's
decision when choosing which product to buy. For example, you may wish to determine
how important price, size, fragrance, and packaging is to a customer when choosing
which brand of perfume to buy.
- This scale is considered as ordinal scale as it is comparative in nature.

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
 Q-Sort Scale
Q-Sort scale is a type of measurement scale that uses a rank order scaling
technique to sort similar objects with respect to some criterion.
The respondents sort the number of statements or attitudes into piles, usually of
11.
The Q-Sort Scaling helps in assigning ranks to different objects within the same
group, and the differences among the groups (piles) are clearly visible. It is a fast
way of facilitating discrimination among a relatively large set of attributes.
For example, a new restaurant that is just preparing its menu may want to
collect some information about what potential customers like:
The document provided contains a list of 50 meals. Please choose 10 meals you
like, 30 meals you are neutral about (neither like nor dislike) and 10 meals you
dislike.

[Link] Pithawala
[Link] Pithawala
 Magnitude Estimation scaling.
- It is a comparative scale in which magnitude of agreement or disagreement is
estimates.
- In this method numbers are assigned to object such those ratios between the
assigned numbers reflect ratios on the specified criteria.
For example respondents may be asked to indicate whether they agree or disagree
with each of a series of statement measuring attitude and then assign a number
between 0 to 100 to each statement to indicate the intensity of their agreement or
disagreement.

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
2) Non-Comparative Scales
In non-comparative scaling, customers are asked to only evaluate a single object. This
evaluation is totally independent of the other objects under investigation. Sometimes
called monadic or metric scale.
Non-Comparative scale can be further divided into,
[Link]
2. Itemized rating scales
a) Likert scale
b) Semantic differential scale
c) staple scale
d) Thurston Scale

[Link] Pithawala
 Continuous Rating Scale
In continuous rating scale, respondents are asked to rate the objects by placing a
mark appropriately on a line running from one extreme of the criterion to the other
variable criterion.
Also called the graphic rating scale, it gives the respondent the freedom to place the
mark anywhere based on personal preference.
Once the ratings are obtained, the researcher splits up the line into several
categories and then assign the scores depending on the category in which the ratings
fall.
This rating can be visualized in both horizontal and vertical form.
Although easy to construct, the continuous rating scale has some major setbacks,
giving it limited usage in market research.

[Link] Pithawala
[Link] Pithawala
 Itemized Rating Scale
The itemized rating scale is a type of ordinal scale that assigns numbers each
attribute. Respondents are usually asked to select an attribute that best
describes their feelings regarding a predefined criterion.
Itemized rating scale is further divided, namely;

Likert scale,
semantic
Staple scale, and
Thurston scale.

[Link] Pithawala
1) Likert Scale:
A Likert scale is an ordinal scale with five response categories, which is used to order a
list of attributes from the best to the least.
This scale uses adverbs of degree like very strongly, highly, etc. to indicate the different
levels.

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
2) Stapel Scale:
This a scale with 10 categories, usually ranging from -5 to 5 with no zero point. It is a
vertical scale with 3 columns, where the attributes are placed in the middle and the least
(-5) and highest (5) is in the 1st and 3rd columns respectively.

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
Semantic Differential Scale:
This is a seven-point rating scale with endpoints associated with bipolar labels (e.g. good
or bad, happy, etc.). It can be used for marketing, advertising and in different stages of
product development.

[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
[Link] Pithawala
Thurston Scale:
This is also known as equal interval scale which was developed by Thurston & chave to
measure the attitude towards a given concept or construct.
While using this scale large number of statements are collected /made which are related
to the study.
The respondents rates this statements on 11 category scale in which each category
expresses a different degree of favorableness toward the concept. Which as follow:
A B C D E F G H I J K

Unfavorable neutral favorable

Thurston & chave have used cards A to K letters to assess the attitude of respondents.
For each of the statements respondents show cards and accordingly score was obtained.
Letter A to K are converted in to 1 to 11 digits then items are ranked according to the
mean or median ratings.
[Link] Pithawala