RM401E:
Research Method and
Statistics
Week 8: Measurement & Sampling
Three types of things that can be
measured
• A direct observable is physical phenomenon or feature
that can be observed directly, such as the number of
people present at a particular place or event.
• Somebody’s response to a questionnaire item about the
number of people working in an organization is an indirect
observable; that is, an indirect representation of a
characteristic or object.
• A construct is a creation based on observation, but it
cannot be observed either directly or indirectly. Examples:
customer satisfaction, job involvement, and price
consciousness.
Operationalization
• The process of translating abstract and subjective
constructs into concrete measures is called
operationalization.
• In this process, one needs to make several important
decisions on how to translate the abstract and
subjective construct into a measure.
Measurement
• Measurement: the assignment of numbers or other
symbols to characteristics (or attributes) of objects
according to a pre-specified set of rules.
– Objects include persons, strategic business units,
companies, countries, kitchen appliances, restaurants,
shampoo, yogurt and so on.
– Examples of characteristics of objects are arousal seeking
tendency, achievement motivation, organizational
effectiveness, shopping enjoyment, length, weight, ethnic
diversity, service quality, conditioning effects and taste.
Operationalizing Concepts
• Operationalizing concepts: reduction of abstract
concepts to render them measurable in a tangible
way.
• Operationalizing is done by looking at the behavioral
dimensions, facets, or properties denoted by the
concept.
Latent Variable
• A latent variable is a variable that is not directly observed but is i
nstead inferred from other variables that are observed.
• used in statistical models to explain the relationships between ob
served variables.
• useful for understanding the relationships between different facto
rs in a complex system.
• help to explain why some people are more successful than others
.
• Example of Latent Variable
• 1- In a study of the relationship between IQ and success, latent varia
bles could be used to represent factors such as motivation and oppo
rtunity.
• 2- in a study of the relationship between job satisfaction and job perf
ormance, latent variables could be used to represent factors such as
motivation and ability.
Steps
1. Provide conceptual definition of construct.
2. Develop pool of items related/important to the
construct.
3. Decide on response format (e.g., 5 point Likert-scales
with end-points ‘strongly disagree’ and ‘strongly agree’).
4. Collect data from representative sample from the
population.
5. Select items for your scale using ‘item-analysis’.
6. Test the reliability and validity of the instrument.
Scale
• Measurement means gathering data in the form of
numbers.
• To be able to assign numbers to attributes of objects
we need a scale: a tool or mechanism by which
individuals are distinguished as to how they differ
from one another on the variables of interest to our
study.
Four Types of Scales
• There are four basic types of scales: nominal, ordinal,
interval, and ratio.
• The degree of sophistication to which the scales are
fine-tuned increases progressively as we move from
the nominal to the ratio scale.
Nominal Scale
• A nominal scale is one that allows the researcher to
assign subjects to certain categories or groups.
• What is your department?
O Marketing O Maintenance O Finance
O Production O Servicing O Personnel
O Sales O Public Relations O Accounting
• What is your gender?
O Male
O Female
Ordinal Scale
• Ordinal scale: not only categorizes variables in such a
way as to denote differences among various
categories, but it also rank-orders categories in some
meaningful way.
• What is the highest level of education you have completed?
O Less than High School
O High School/GED Equivalent
O College Degree
O Masters Degree
O Doctoral Degree
Interval Scale
• In an interval scale, or equal interval scale,
numerically equal distances on the scale represent
equal values in the characteristics being measured.
• It allows us to compare differences between objects:
The difference between any two values on the scale
is identical to the difference between any two other
neighboring values of the scale.
– The clinical thermometer is an example; it has an arbitrary
origin and the magnitude of the difference between 36.5
degrees (the normal body temperature) and 37.5 degrees
is the same as the magnitude of the difference between 39
and 40 degrees.
Ratio Scale
• 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.
• What is your age?
Ordinal Scale or Interval Scale?
• Circle the number that represents your feelings at this
particular moment best. There are no right or wrong answers.
Please answer every question.
1. I invest more in my work than I get out of it
I disagree completely 1 2 3 4 5 I agree completely
2. I exert myself too much considering what I get back in return
I disagree completely 1 2 3 4 5 I agree completely
3. For the efforts I put into the organization, I get much in return
I disagree completely 1 2 3 4 5 I agree completely
Properties of the Four Scales
Goodness of Measures
Validity and Reliability
• Validity refers the level of certainty that we are
indeed measuring the concept we set out to
measure and not something else
• Reliability indicates the extent to which it is without
bias and hence ensures consistent measurement
across time (stability) and across the various items in
the instrument (internal consistency).
Stability
• Stability: ability of a measure to remain the same
over time, despite uncontrollable testing conditions
or the state of the respondents themselves.
– Test–Retest Reliability: The reliability coefficient obtained
with a repetition of the same measure on a second
occasion.
– Parallel-Form Reliability: Responses on two comparable
sets of measures tapping the same construct are highly
correlated.
Internal Consistency
• Internal Consistency of Measures is indicative of the
homogeneity of the items in the measure that tap
the construct.
– Inter-item Consistency Reliability: This is a test of the
consistency of respondents’ answers to all the items in a
measure. The most popular test of inter-item consistency
reliability is the Cronbach’s coefficient alpha.
– Split-Half Reliability: Split-half reliability reflects the
correlations between two halves of an instrument.
Sampling
• Sampling: the process of selecting a sufficient number of
elements from the population, so that results from
analyzing the sample are generalizable to the population.
• The reasons for using a sample are self-evident. In research
involving hundreds or even thousands of elements, it
would be practically impossible to collect data from every
element. Even if it were possible, it would be prohibitive in
terms of time, cost, and other human resources.
Relevant Terms - 1
• 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. It comprises some
members selected from it.
• 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.
Relevant Terms - 2
• The characteristics of the population such as µ (the
population mean), σ (the population standard
deviation), and σ2 (the population variance) are
referred to as its parameters. The central tendencies,
the dispersions, and other statistics in the sample of
interest to the research are treated as
approximations of the central tendencies,
dispersions, and other parameters of the population.
Statistics versus Parameters
The Sampling Process
• Major steps in sampling:
– Define the population.
– Determine the sample frame
– Determine the sampling design
– Determine the appropriate sample size
– Execute the sampling process
Sampling Techniques
• Probability Sampling: elements in the population
have a known and non-zero chance of being chosen
– Simple Random Sampling
– Systematic Sampling
– Stratified Random Sampling
– Cluster Sampling
• Nonprobability Sampling
– Convenience Sampling
– Judgment Sampling
– Quota Sampling
Simple Random Sampling
• Procedure
– Each element has a known and equal chance of being
selected
• Characteristics
– Highly generalizable
– Easily understood
– Reliable population frame necessary
An Illustration of
Simple Random Sampling
A B C D E
1 6 11 16 21
2 7 12 17 22
3 8 13 18 23
4 9 14 19 24
5 10 15 20 25
Select five random
numbers from 1 to 25.
The resulting sample
consists of population
elements 3, 7, 9, 16,
and 24. Note, there is
no element from Group
C.
Systematic sampling
• Procedure
– Each nth element, starting with random choice of an
element between 1 and n
• Characteristics
– Easier than simple random sampling in implementation
– May increase the representation of the sample
– Systematic biases when elements are not randomly
listed
An Illustration of
Systematic Sampling
A B C D E
1 6 11 16 21
2 7 12 17 22
3 8 13 18 23
4 9 14 19 24
5 10 15 20 25
Select a random number
between 1 and 5, say 2.
The resulting sample
consists of population 2,
(2+5=) 7, (2+5x2=) 12,
(2+5x3=)17, and (2+5x4=) 22.
Note, all the elements are
selected from a single row.
Stratified sampling
• Procedure
– Divide of population in strata
– Include all strata
– Random selection of elements from strata
• Proportionate
• Disproportionate
• Characteristics
– Inter-strata heterogeneity
– Intra-stratum homogeneity
– Includes all relevant subpopulations
An Illustration of
Stratified Sampling
A B C D E
1 6 11 16 21
2 7 12 17 22
3 8 13 18 23
4 9 14 19 24
5 10 15 20 25
Randomly select a number
from 1 to 5
for each stratum, A to E. The
resulting
sample consists of
population elements
4, 7, 13, 19 and 21. Note, one
element
is selected from each
column.
(Dis)proportionate stratified
sampling
• Number of subjects in total sample is allocated
among the strata (dis)proportional to the relative
number of elements in each stratum in the
population
• Disproportionate case:
– strata exhibiting more variability are sampled more than
proportional to their relative size
– Used when some strata are too small or too large, or most
variability within one or two strata
Example
Cluster sampling
• Procedure
– Divide of population in clusters
– Random selection of clusters
– Include all elements from selected clusters
• Characteristics
– Inter-cluster homogeneity
– Intra-cluster heterogeneity
– Easy and cost efficient
– Low correspondence with reality
An Illustration of
Cluster Sampling (2-Stage)
A B C D E
1 6 11 16 21
2 7 12 17 22
3 8 13 18 23
4 9 14 19 24
5 10 15 20 25
Randomly select 3 clusters,
B, D and E.
Within each cluster,
randomly select one
or two elements. The
resulting sample
consists of population
elements 7, 18, 20, 21, and
23. Note, no elements are
selected from clusters A and
C.
Strengths and Weaknesses of
Basic Sampling Techniques
Technique Strengths Weaknesses
Nonprobability Sampling
Convenience sampling
Least expensive, least
time-consuming, most
convenient
Selection bias, sample not
representative, not recommended for
descriptive or causal research
Judgmental sampling Low cost, convenient,
not time-consuming
Does not allow generalization,
subjective
Quota sampling Sample can be controlled
for certain characteristics
Selection bias, no assurance of
representativeness
Probability sampling
Simple random sampling
(SRS)
Easily understood,
results projectable
Difficult to construct sampling
frame, expensive, lower precision,
no assurance of representativeness
Systematic sampling Can increase
representativeness,
easier to implement than
SRS, sampling frame not
necessary
Can decrease representativeness
Stratified sampling Include all important
subpopulations,
precision
Difficult to select relevant
stratification variables, not feasible to
stratify on many variables, expensive
Cluster sampling Easy to implement, cost
effective
Imprecise, difficult to compute and
interpret results
Tradeoff between precision and
confidence
• We can increase both confidence and precision
by increasing the sample size
– Precision refers to how close our estimate is to the
true population characteristic.
– Confidence denotes how certain we are that our
estimates will really hold true for the population
• Given a certain sample size, the only way to
maintain the same level of precision is to
forsake the confidence with which we can
predict our estimates
Choice Points in Sampling Design
Sample size: guidelines
• In general: 30 < n < 500
• Categories: 30 per subcategory
• Multivariate: 10 x number of var’s
• Experiments: 15 to 20 per condition
Sampling in Qualitative Research
• Qualitative research generally uses nonprobability
sampling as it does not aim to draw statistical
inference.
• Purposive sampling is one technique that is often
used: subjects are selected on the basis of expertise
in the subject that is being investigated.
• Choose subjects in such a way that they reflect the
diversity of the population.
To do before next week’s lesson
• Install PSPP (a free statistics software) on your
own computer
– For the rest lessons of the course as well as your
project
– Please download the installation file from
https://sourceforge.net/projects/pspp4windows/
and install it on your computer before the lesson
next week
Mini-Quiz 2
Duration: 15 minutes
Ten Multiple Choice Questions on
Research Design and survey /experiment
method (Week 5-7)
The End

Chapter 8: Measurement and Sampling

  • 1.
  • 2.
    Three types ofthings that can be measured • A direct observable is physical phenomenon or feature that can be observed directly, such as the number of people present at a particular place or event. • Somebody’s response to a questionnaire item about the number of people working in an organization is an indirect observable; that is, an indirect representation of a characteristic or object. • A construct is a creation based on observation, but it cannot be observed either directly or indirectly. Examples: customer satisfaction, job involvement, and price consciousness.
  • 3.
    Operationalization • The processof translating abstract and subjective constructs into concrete measures is called operationalization. • In this process, one needs to make several important decisions on how to translate the abstract and subjective construct into a measure.
  • 4.
    Measurement • Measurement: theassignment of numbers or other symbols to characteristics (or attributes) of objects according to a pre-specified set of rules. – Objects include persons, strategic business units, companies, countries, kitchen appliances, restaurants, shampoo, yogurt and so on. – Examples of characteristics of objects are arousal seeking tendency, achievement motivation, organizational effectiveness, shopping enjoyment, length, weight, ethnic diversity, service quality, conditioning effects and taste.
  • 5.
    Operationalizing Concepts • Operationalizingconcepts: reduction of abstract concepts to render them measurable in a tangible way. • Operationalizing is done by looking at the behavioral dimensions, facets, or properties denoted by the concept.
  • 6.
    Latent Variable • Alatent variable is a variable that is not directly observed but is i nstead inferred from other variables that are observed. • used in statistical models to explain the relationships between ob served variables. • useful for understanding the relationships between different facto rs in a complex system. • help to explain why some people are more successful than others . • Example of Latent Variable • 1- In a study of the relationship between IQ and success, latent varia bles could be used to represent factors such as motivation and oppo rtunity. • 2- in a study of the relationship between job satisfaction and job perf ormance, latent variables could be used to represent factors such as motivation and ability.
  • 7.
    Steps 1. Provide conceptualdefinition of construct. 2. Develop pool of items related/important to the construct. 3. Decide on response format (e.g., 5 point Likert-scales with end-points ‘strongly disagree’ and ‘strongly agree’). 4. Collect data from representative sample from the population. 5. Select items for your scale using ‘item-analysis’. 6. Test the reliability and validity of the instrument.
  • 8.
    Scale • Measurement meansgathering data in the form of numbers. • To be able to assign numbers to attributes of objects we need a scale: a tool or mechanism by which individuals are distinguished as to how they differ from one another on the variables of interest to our study.
  • 9.
    Four Types ofScales • There are four basic types of scales: nominal, ordinal, interval, and ratio. • The degree of sophistication to which the scales are fine-tuned increases progressively as we move from the nominal to the ratio scale.
  • 10.
    Nominal Scale • Anominal scale is one that allows the researcher to assign subjects to certain categories or groups. • What is your department? O Marketing O Maintenance O Finance O Production O Servicing O Personnel O Sales O Public Relations O Accounting • What is your gender? O Male O Female
  • 11.
    Ordinal Scale • Ordinalscale: not only categorizes variables in such a way as to denote differences among various categories, but it also rank-orders categories in some meaningful way. • What is the highest level of education you have completed? O Less than High School O High School/GED Equivalent O College Degree O Masters Degree O Doctoral Degree
  • 12.
    Interval Scale • Inan interval scale, or equal interval scale, numerically equal distances on the scale represent equal values in the characteristics being measured. • It allows us to compare differences between objects: The difference between any two values on the scale is identical to the difference between any two other neighboring values of the scale. – The clinical thermometer is an example; it has an arbitrary origin and the magnitude of the difference between 36.5 degrees (the normal body temperature) and 37.5 degrees is the same as the magnitude of the difference between 39 and 40 degrees.
  • 13.
    Ratio Scale • Ratioscale: 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. • What is your age?
  • 14.
    Ordinal Scale orInterval Scale? • Circle the number that represents your feelings at this particular moment best. There are no right or wrong answers. Please answer every question. 1. I invest more in my work than I get out of it I disagree completely 1 2 3 4 5 I agree completely 2. I exert myself too much considering what I get back in return I disagree completely 1 2 3 4 5 I agree completely 3. For the efforts I put into the organization, I get much in return I disagree completely 1 2 3 4 5 I agree completely
  • 15.
    Properties of theFour Scales
  • 16.
  • 17.
    Validity and Reliability •Validity refers the level of certainty that we are indeed measuring the concept we set out to measure and not something else • Reliability indicates the extent to which it is without bias and hence ensures consistent measurement across time (stability) and across the various items in the instrument (internal consistency).
  • 18.
    Stability • Stability: abilityof a measure to remain the same over time, despite uncontrollable testing conditions or the state of the respondents themselves. – Test–Retest Reliability: The reliability coefficient obtained with a repetition of the same measure on a second occasion. – Parallel-Form Reliability: Responses on two comparable sets of measures tapping the same construct are highly correlated.
  • 19.
    Internal Consistency • InternalConsistency of Measures is indicative of the homogeneity of the items in the measure that tap the construct. – Inter-item Consistency Reliability: This is a test of the consistency of respondents’ answers to all the items in a measure. The most popular test of inter-item consistency reliability is the Cronbach’s coefficient alpha. – Split-Half Reliability: Split-half reliability reflects the correlations between two halves of an instrument.
  • 20.
    Sampling • Sampling: theprocess of selecting a sufficient number of elements from the population, so that results from analyzing the sample are generalizable to the population. • The reasons for using a sample are self-evident. In research involving hundreds or even thousands of elements, it would be practically impossible to collect data from every element. Even if it were possible, it would be prohibitive in terms of time, cost, and other human resources.
  • 21.
    Relevant Terms -1 • 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. It comprises some members selected from it. • 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.
  • 22.
    Relevant Terms -2 • The characteristics of the population such as µ (the population mean), σ (the population standard deviation), and σ2 (the population variance) are referred to as its parameters. The central tendencies, the dispersions, and other statistics in the sample of interest to the research are treated as approximations of the central tendencies, dispersions, and other parameters of the population.
  • 23.
  • 24.
    The Sampling Process •Major steps in sampling: – Define the population. – Determine the sample frame – Determine the sampling design – Determine the appropriate sample size – Execute the sampling process
  • 25.
    Sampling Techniques • ProbabilitySampling: elements in the population have a known and non-zero chance of being chosen – Simple Random Sampling – Systematic Sampling – Stratified Random Sampling – Cluster Sampling • Nonprobability Sampling – Convenience Sampling – Judgment Sampling – Quota Sampling
  • 26.
    Simple Random Sampling •Procedure – Each element has a known and equal chance of being selected • Characteristics – Highly generalizable – Easily understood – Reliable population frame necessary
  • 27.
    An Illustration of SimpleRandom Sampling A B C D E 1 6 11 16 21 2 7 12 17 22 3 8 13 18 23 4 9 14 19 24 5 10 15 20 25 Select five random numbers from 1 to 25. The resulting sample consists of population elements 3, 7, 9, 16, and 24. Note, there is no element from Group C.
  • 28.
    Systematic sampling • Procedure –Each nth element, starting with random choice of an element between 1 and n • Characteristics – Easier than simple random sampling in implementation – May increase the representation of the sample – Systematic biases when elements are not randomly listed
  • 29.
    An Illustration of SystematicSampling A B C D E 1 6 11 16 21 2 7 12 17 22 3 8 13 18 23 4 9 14 19 24 5 10 15 20 25 Select a random number between 1 and 5, say 2. The resulting sample consists of population 2, (2+5=) 7, (2+5x2=) 12, (2+5x3=)17, and (2+5x4=) 22. Note, all the elements are selected from a single row.
  • 30.
    Stratified sampling • Procedure –Divide of population in strata – Include all strata – Random selection of elements from strata • Proportionate • Disproportionate • Characteristics – Inter-strata heterogeneity – Intra-stratum homogeneity – Includes all relevant subpopulations
  • 31.
    An Illustration of StratifiedSampling A B C D E 1 6 11 16 21 2 7 12 17 22 3 8 13 18 23 4 9 14 19 24 5 10 15 20 25 Randomly select a number from 1 to 5 for each stratum, A to E. The resulting sample consists of population elements 4, 7, 13, 19 and 21. Note, one element is selected from each column.
  • 32.
    (Dis)proportionate stratified sampling • Numberof subjects in total sample is allocated among the strata (dis)proportional to the relative number of elements in each stratum in the population • Disproportionate case: – strata exhibiting more variability are sampled more than proportional to their relative size – Used when some strata are too small or too large, or most variability within one or two strata
  • 33.
  • 34.
    Cluster sampling • Procedure –Divide of population in clusters – Random selection of clusters – Include all elements from selected clusters • Characteristics – Inter-cluster homogeneity – Intra-cluster heterogeneity – Easy and cost efficient – Low correspondence with reality
  • 35.
    An Illustration of ClusterSampling (2-Stage) A B C D E 1 6 11 16 21 2 7 12 17 22 3 8 13 18 23 4 9 14 19 24 5 10 15 20 25 Randomly select 3 clusters, B, D and E. Within each cluster, randomly select one or two elements. The resulting sample consists of population elements 7, 18, 20, 21, and 23. Note, no elements are selected from clusters A and C.
  • 36.
    Strengths and Weaknessesof Basic Sampling Techniques Technique Strengths Weaknesses Nonprobability Sampling Convenience sampling Least expensive, least time-consuming, most convenient Selection bias, sample not representative, not recommended for descriptive or causal research Judgmental sampling Low cost, convenient, not time-consuming Does not allow generalization, subjective Quota sampling Sample can be controlled for certain characteristics Selection bias, no assurance of representativeness Probability sampling Simple random sampling (SRS) Easily understood, results projectable Difficult to construct sampling frame, expensive, lower precision, no assurance of representativeness Systematic sampling Can increase representativeness, easier to implement than SRS, sampling frame not necessary Can decrease representativeness Stratified sampling Include all important subpopulations, precision Difficult to select relevant stratification variables, not feasible to stratify on many variables, expensive Cluster sampling Easy to implement, cost effective Imprecise, difficult to compute and interpret results
  • 37.
    Tradeoff between precisionand confidence • We can increase both confidence and precision by increasing the sample size – Precision refers to how close our estimate is to the true population characteristic. – Confidence denotes how certain we are that our estimates will really hold true for the population • Given a certain sample size, the only way to maintain the same level of precision is to forsake the confidence with which we can predict our estimates
  • 38.
    Choice Points inSampling Design
  • 39.
    Sample size: guidelines •In general: 30 < n < 500 • Categories: 30 per subcategory • Multivariate: 10 x number of var’s • Experiments: 15 to 20 per condition
  • 40.
    Sampling in QualitativeResearch • Qualitative research generally uses nonprobability sampling as it does not aim to draw statistical inference. • Purposive sampling is one technique that is often used: subjects are selected on the basis of expertise in the subject that is being investigated. • Choose subjects in such a way that they reflect the diversity of the population.
  • 41.
    To do beforenext week’s lesson • Install PSPP (a free statistics software) on your own computer – For the rest lessons of the course as well as your project – Please download the installation file from https://sourceforge.net/projects/pspp4windows/ and install it on your computer before the lesson next week
  • 42.
    Mini-Quiz 2 Duration: 15minutes Ten Multiple Choice Questions on Research Design and survey /experiment method (Week 5-7)
  • 43.