Chapter 12 : Measurement: Scaling,

reliability and validity

• Measurement of the variables is an integral part of research and an important
aspect of research design.
• Unless the variables are measured in some way, we will not be able to find
answers to our research questions.
• Surveys and experimental designs, discussed in the previous chapters, often
use questionnaires to measure the variables of interest
• To test the hypothesis that workforce diversity affects organizational
effectiveness we have to measure workforce diversity and organizational
effectiveness.
• Measurement is the assignment of numbers or other symbols to characteristics
(or attributes) of objects according to a prespecified set of rules
• Now that we have learned how to operationally define
(operationalize) a concept (or variable), we need to assign numbers
(or other symbols) to it in some manner.
• Numbers allow us to perform statistical analysis on the resulting data
and (in deductive research) to test the hypotheses that we have
developed.
• Facilitate the communication of results.
• The types of scales that can be applied to assign numbers to
characteristics of objects
• Types of scale (nominal, ordinal, interval, ratio) and the statistical test
that can be performed on them.
• Attitudinal scales (rating scale, ranking scale)
 Four types of scales
• Measurement means gathering data in the form of numbers.
• To be able to assign numbers to attributes of objects we need a scale.
• A scale is a tool or mechanism by which individuals are distinguished
as to how they differ from one another on the variables of interest.
• Nominal Scale: A nominal scale is one that allows the researcher to
assign subjects to certain categories or groups , male/ female two
categories, can be coded as 1/2 simple category labels.
• No intrinsic value just assignment to non overlapping categories / mutually
exclusive / Exhaustive
• Ordinal Scale: An ordinal scale not only categorizes the variables in
such a way as to denote differences among the various categories, it
also rank-orders the categories in some meaningful way. i.e., best to
worst, first to last, 1234 etc.
• Provides much more information than the nominal scale
• See example on pg 209 of the book
• Does not give indication of the magnitude of the difference among the
ranks
• Interval Scale: In an interval scale, or equal interval scale,
numerically equal distances on the scale represent equal values in the
characteristics being measured.
• The difference between any two values on the scale is identical to the
difference between any other two neighboring values of the scale.
• The clinical thermometer is a good example of an interval-scaled instrument;
it has an arbitrary origin and the magnitude of the difference between 98.6
degrees and 99.6 degrees is the same as the magnitude of the difference
between 104 and 105 degrees.
• The interval scale, then, taps the differences, the order, and the equality of the magnitude of the
differences in the variable.
• Ratio Scale: The ratio scale overcomes the disadvantage of the
arbitrary origin point of the interval scale, in that it has an absolute (in
contrast to an arbitrary) zero point, which is a meaningful
measurement point.
• Thus, the ratio scale not only measures the magnitude of the
differences between points on the scale but also taps the proportions in
the differences.
• For instance, a person weighing 250 pounds is twice as heavy as
one who weighs 125 pounds.
• Likert scales: Students participation (1-5)(strongly disagree-strongly
agree – middle neither agree nor disagree)
• Allows researchers to measure averages and standard deviations and
other statistical techniques
• Read examples on pg 210-211
 Attitudinal scales (Ranking and Rating
scales)

• Rating scales each object is scaled independently of the other objects

under study.
• Ranking scales on the other hand, make comparisons between or
among objects and elicit the preferred choices and ranking among
them.
 Rating Scales

• Dichotomous scale - (nominal)

• Category scale - (nominal)
• Semantic differential scale – (ordinal/ interval) – (Bipolar objective /
non overlapping) - (Semantic space can be plotted) – (used to measure
the respondent attitude towards some brand , events, object etc.) Read
pg 214
• Numerical scale – (Similar to the semantic difference scale but with
numbers) – (Bipolar objectives / non overlapping)
• Itemized rating scale – (5-7 points with anchor with items numbered)
• Likert scale
• Fixed or constant sum rating scale – (ordinal)
• Stapel scale (gives both the intensity and the direction of attitude)
• Graphic rating scale
 Ranking scale
• Paired comparison
• Forced choice
• Comparative scale

• (Student reading pg 218-219)

 International Dimensions of Scaling

• Now that we have seen how to operationally define variables and

apply different scaling techniques, it is important to make sure that the
instrument (questionnaire) that we develop to measure a particular
concept is indeed accurately measuring the variable.
• Hence, in some way, we need to assess the “goodness” of the measures
developed.
• We need to be reasonably sure that the instruments we use in our
research do indeed measure the variables they are supposed to, and
that they measure them accurately.
• Reliability is a test of how consistently a measuring instrument
measures whatever concept it is measuring.
• Validity is a test of how well an instrument that is developed measures
the particular concept it is intended to measure.
• Validity is a test of how well an instrument that is developed measures
the particular concept it is intended to measure.
• Content validity ensures that the measure includes an adequate and
representative set of items that tap the concept.
• Criterion-related validity is established when the measure differentiates
individuals on a criterion it is expected to predict. Read examples 222
• Construct validity testifies to how well the results obtained from the use of
the measure fit the theories around which the test is designed
• Reliability is a test of how consistently a measuring instrument
measures whatever concept it is measuring.
Stability of Measure (Ability of the measure to remain the same over time)
• The reliability coefficient obtained by repetition of the same measure on a
second occasion is called the test–retest reliability.
• When responses on two comparable sets of measures tapping the same
construct are highly correlated, we have parallel-form reliability.
Internal consistency of measure (The homogeneity of the items in the measure that taps the construct)
• test of the consistency of respondents’ answers to all the items in a measure,
interitem consistency reliability
• Split-half reliability reflects the correlations between two halves of an
instrument
 Students reading assignment

• Reflective versus formative measurement scales

 Chapter 13 Sampling

• The population refers to the entire group of people, events, or things

of interest that the researcher wishes to investigate.
• An element is a single member of the population.
• A sample is a subset of the population.
• The sampling unit is the element or set of elements that is available
for selection in some stage of the sampling process.
• A subject is a single member of the sample, just as an element is a
single member of the population
Normality of distribution
 Sampling Process

• Define the population

• Determine the sample frame
• Determine the sampling design (probability / Non probability)
• Determine the appropriate sample size
• Execute the sampling process
 Probability Sampling

• Simple random sampling (unrestricted)

• Every element has the equal chance of selection
• The distribution pattern of characteristics of interest are distributed normally
• Complex probability sampling (restricted)
 Complex Probability Sampling
• Efficiency is improved in that more information can be obtained for a given
sample size using some of the complex probability sampling procedures than the
simple random sampling design.
1. Systematic sampling, (Choosing nth element) Example 243
2. Stratified random sampling, (Proportionate / disproportionate), (Homogeneity
within group and heterogeneity among groups) Example 245
3. Cluster sampling (Area sampling special case), (Can be single stage and
multistage sampling)
4. Double sampling (Subset of the original sample for further inquiry and
interviews)
 Non probability sampling

1. Convenience sampling
2. Purposive sampling (Judgement sampling / Quota sampling),
(Similar to the stratified proportional sampling but based on
convenience)

Read table 13.2 on page 249

Read figure 13.4 on page 251
Practical examples from page 252 - 256
 Issues of precision and confidence in
determining sample size
• A reliable and valid sample should enable us to generalize the findings
from the sample to the population under investigation
• the sample statistics should be reliable estimates and reflect the population
parameters as closely as possible within a narrow margin of error.
• No sample statistic (X) is going to be exactly the same as the population
parameter (µ), no matter how sophisticated the probability sampling design
is.
• Though the point estimate X may not accurately reflect the population
mean, µ, an interval estimate can be made within which µ will lie, with
probabilities attached – that is, at particular confidence levels.
 Precision
• Precision refers to how close our estimate is to the true population
characteristic.
• Usually, we estimate the population parameter to fall within a range, based
on the sample estimate
• The narrower this interval, the greater the precision
••  
Precision is a function of the range of variability in the sampling distribution
of the sample mean

Which one is more precise ???

• Even if we take only one sample of 30 subjects from the population,
we will still be able to estimate the variability of the sampling
distribution of the sample mean. This variability is called the standard
error, denoted by . The standard error is calculated by the following
formula:
• 
 Confidence
• Whereas precision denotes how close we estimate the population
parameter based on the sample statistic.
• Confidence denotes how certain we are that our estimates will really
hold true for the population. 90 percent confident, 95 percent
confident etc
• A 95% confidence is the conventionally accepted level for most
business research, most commonly expressed by denoting the
significance level as p ≤ 0.05.
• In other words, we say that at least 95 times out of 100 our estimate
will reflect the true population characteristic.
 Google formula of sample
standard deviation.

The use of t or z distribution is dependent on the sample size

usually z distribution is used if sample size is greater than 30 other
wise for sample size below or equal to 30 t distribution
 Sample data and hypo testing

• So far we have discussed sample data as a means of estimating the

population parameters, but sample data can also be used to test
hypotheses about population values rather than simply to estimate
population values.
Customers expend the same average amount in purchases in Department Store
A as in Department Store B.

• You require confidence in your finding and that confidence should be presented in probability e.g. out
of 100 time of similar situation 95 time this outcome was observed (There are no certainties in life
only probabilities)
• As the sample is below 30 for each store so t distribution is required
Sample size