Operatinalization Measurement Sampling
Operatinalization Measurement Sampling
Operatinalization Measurement Sampling
MEASUREMENT SAMPLING
• RESEARCH SEMINAR I
• Dr. Lizan Perante - Calina
• RSH 630
» REPORTERS:
HARRY LECHICO
DIANNE JENNIFER MOLINA
WILHELMINA ROSARIO T. DAILO
Operationalization Research
•The process of creating a definition(s) for a concept
that can be observed and measured
Sample
The study
population
Groups in Sampling
The theoretical
population
The study
population
The study
population
The sampling
frame
Groups in Sampling
The theoretical
population
The study
population
The sampling
frame
The study
population
The sampling
frame
The sample
TYPES OF SAMPLING TECHNIQUES
PROBABILITY SAMPLING
PROBABILITY SAMPLING
• any method of sampling that utilizes some form of
random selection. (SocialResearchMethod.net)
• is a sampling technique wherein the samples are
gathered in a process that gives all the individuals
in the population equal chances of being selected.
(Explorable.com)
• each population member has a known, non-zero
chance of participating in the study. (Research-
Methodology.net)
• uses random sampling techniques to create a
sample. (SurveyGizmo.com)
Application of Probability
Sampling
FOUR STAGES OF
PROBABILITY SAMPLING
1. PROCESS
Identifying an appropriate sampling frame based on your
research question(s) and objectives.
2. Determining a suitable sample size.
3. Choosing the most appropriate sampling technique and
selecting the samples.
4. Checking if the sample is representative of the population.
Advantages of Probability
Sampling
1. LOTTERY METHOD
2. USE OF RANDOM
NUMBERS
STRATIFIED
SAMPLING
a probability
sampling method and a form
of random sampling in which
the population is divided
into two or more groups
(strata) according to one or
more common attributes.
intends to guarantee that
the sample represents
specific sub-groups or strata.
Application of Stratified Sampling
ABC Company
Total Population = 200
Actual Sample Size = 24
2. Sampling Fraction = 24/200 = 3/25 rounded to 1/8 (Can be rounded of to the nearest 10 or 100)
Accordingly, every 8th member of the sampling frame needs to be selected to participate in the study.
3. Choose the first sample randomly. Suppose you choose #47 as first sample.
4. Therefore, your sample group will comprise of ABC Company employees under the following
numbers: #47; #55; #63; #71; #79; #87; #95; #103; #111; #119; #127; #135; #143; #151; #159; #167;
#175; #183; #191; #199; #7; #15; #23; #31.
CLUSTER
SAMPLING
is a technique in which clusters
of participants that represent
the population are identified
and included in the sample.
involves identification of cluster
of participants representing the
population and their inclusion
in the sample group.
The main aim of cluster
sampling can be specified as
cost reduction and increasing
the levels of efficiency of
sampling.
CLUSTER
SAMPLING
conducted when the size of a
population is too large to
perform simple random
sampling.
A major difference between
cluster and stratified
sampling relates to the fact that
in cluster sampling a cluster is
perceived as a sampling unit,
whereas in stratified sampling
only specific elements of strata
are accepted as sampling unit.
Application of Cluster
Sampling
1. Select a cluster grouping as a sampling frame.
2. Mark each cluster with a unique number.
3. Choose a sample of clusters applying probability sampling.
MULTI-STAGE
SAMPLING
a more complex form of cluster
sampling which contains two or
more stages in sample
selection.
In simple terms, in multi-stage
sampling large clusters of
population are divided into
smaller clusters in several
stages in order to make primary
data collection more
manageable.
involves a combination of two
or more of the probability
sampling methods
Application of Multi-Stage
Sampling
Ethnographic 35 – 36
Grounded theory 20 – 35
Considering a homogeneous
population 4 – 12
Considering a heterogeneous
population 12 – 30
Advantages of Non-Probability
Sampling
We can logically arrange the four groups in terms of the relative amount of religiosity (the
shared attribute) each had. We arranged the four groups in a row ranging from the one
with lowest to the highest religiosity. This arrangement would provide a religious
representation of an ordinal measure.
Levels of Measurement
• Interval Measures. A level of measurement describing a variable
whose attributes are rank ordered and have equal distances
between adjacent attributes.
Illustration:
The interval separating IQ scores of 95 and100 may be
regarded as the same as the interval separating scores of 110
and 1115, by virtue of the distribution of observed scores
obtained by many thousands of people who have taken the
tests over the years. But it would be incorrect to infer that
someone with an IQ of 150 is 50 percent more intelligent than
some-one with an IQ of 100.
Attributes are ranked from lowest to highest IQ and their
variables are all 5.
Levels of Measurement
• Ratio measure A level of measurement describing a variable with attributes that
have all the qualities of nominal, ordinal and interval measures and in addition are
based on a "true zero" point.
Illustration:
All with $10,000 income together, the $20,000 income together, the $30,000
income, and so forth. The fact that members of a single group share the same income
and that each different group has a different shared income satisfies the minimum
requirement for a nominal measure. Arranging the several groups in a line from
lowest to highest income meets the additional requirements of an ordinal measure
and lets us determine if one person has a higher income than, or the same income as
another. If we space the groups equally far apart, we satisfy the additional
requirements of an interval measure and will be able to say who receives more
income than another. Finally, because one of the attributes included in income
represents a true zero (income extends from an absolute zero to $50,000). Comparing
two people in terms of a ratio variable then, allows us to determine ( I )that they are
different (or the same), (2) that one is more than the other, (3)how much they differ
and (4) the ratio of one to another.
Implications of Levels of
Measurement
• The variables to be examined in a research project
are limited to a particular level of measurement. At
the same time, you can treat some variables as
representing different levels of measurement. Ratio
measures are the highest level, descending through
interval and ordinal to nominal, the lowest level of
measurement.
• The level of measurement you'll seek then, is
determined by the analytical uses you've planned for
a given variable, as you keep in mind that some
variables are inherently limited to a certain level. If a
variable is to be used in a variety of ways, requiring
different levels of measurement, the study should be
designed to achieve the highest level required.
• For example, if the subjects in a study are asked their
exact ages and you have no need for ages at higher
than the ordinal level of measurement, they can later
be organized into ordinal or nominal groupings, you
may simply group their age range, such as 20 to 29,
30 to 39, and so forth.
Implications of Levels of
Measurement
• Always seek the highest level of measurement possible. Again,
although ratio measures can later be reduced to ordinal ones,
you cannot convert an ordinal measure to a ratio one. More
generally, you cannot convert a lower-level measure to a
higher-level one. That is a one-way street worth remembering.
• Typically a research project will tap variables at different levels
of measurement
For example, William and Denise Bielby (1999) set out to examine the world of film and television
in what they referred to as the "culture industry," the authors found that reputation (an ordinal
variable) is the best predictor of screenwriters' future productivity
They found that screen writers who were represented by "core" (or elite) agencies were far more
likely not only to find jobs (a nominal variable) but also to find jobs that paid more (a ratio
variable).
In other words, the researchers found that an agency's reputation (ordinal) was a key independent
variable for predicting a screen writer's success. The researchers also found that being older
(ratio), being female (nominal), belonging to an ethnic minority (nominal), and having more years
of experience (ratio) were disadvantageous for a screenwriter. On the other hand, higher earnings
from previous years (measured in ordinal categories) led to more success in the future.
Single or Multiple Indicators
• Many social research variables have fairly obvious, straightforward
measures. No matter how you cut it, gender usually turns out to be a
matter of male or female: a nominal-level variable that can be measured
by a single observation.
• Sometimes, however, there is no single indicator that will give you the
measure of a chosen variable.
• Many concepts are subject to varying interpretations each with several
possible indicators. In these cases, you'll want to make several
observations for a given variable. You can then combine the several pieces
of information you've collected to create a composite measurement of the
variable in question.
• Consider the concept "college performance.” All of us have noticed that
some students perform well in college courses and others do not.
• We might ask what characteristics and experiences are related to high
levels of performance. How should we measure over all performance?
Each grade in any single course is a potential indicator of college
performance, but it also may not typify the student's general performance.
The solution to this problem is so firmly established that it is, the grade
point average (GPA). Total the points earned by given student and divide
by the number of courses taken to obtain a composite measure.(If the
courses vary in number of credits, we adjust the point values accordingly.)
CRITERIA OF MEASUREMENT
QUALITY
Precision and Accuracy
• Precision and accuracy are important qualities in research
measurement.
• The description of a woman as "43 years old" is more precise
than "in her forties.”
• 43 years old because she was born on 1975, that’s accuracy.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
• Reliability is a matter of whether a particular technique,
applied repeatedly to the same object, yields the same result
each time
Illustration:
If you want to measure life health situation, you might want to
follow research from the department of health.
Several techniques developed for
cross-checking the reliability of the
measures.
• Reliability of Research Work. Measurement unreliability can
also be generated by research workers: interviewers and
coders, for example
Illustration:
To guard against interviewer unreliability, it is common
practice in surveys to have a supervisor call a
subsample of the respondents on the telephone and
verify selected pieces of information.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
Our IQ would seem a more valid measure of our intelligence than would the
number of hours you spend in the library. Though the ultimate validity of a
measure can never be proven, we may agree to its relative validity on the
basis of face validity, criterion-related validity, content validity, construct
validity, internal validation, and external validation. This must not be confused
with reliability.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
• Face validity is the quality of an indicator that makes it seem a
reasonable measure of some variable.
Illustration:
That the frequency of church attendance is some
indication of a person’s religiosity seems to make sense
without a lot of explanation. It has face validity.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
• Criterion-related validity is the degree to which a measure
relates to some external criterion, also called predictive
validity.
Illustration:
The validity of College Board exams is shown in their
ability to predict students’ success in college.
College success the criteria.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
• Construct validity is the degree to which a measure relates to
other variables as expected within a system of theoretical
relationships.
Illustration:
We might reasonably conclude that satisfied husbands
and wives will be less likely than dissatisfied ones to
cheat on their spouses. If our measure relates to
marital fidelity, that constitutes evidence of our
measure’s construct validity. Tests of construct validity,
then, can offer a weight of evidence that we measure
either does or does not tap the quality we want it to
measure.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
Illustration:
A test of mathematical ability cannot be limited to
addition alone but also needs to cover subtraction,
multiplication, division, and so forth.
CRITERIA OF MEASUREMENT
QUALITY
RELIABILITY AND VALIDITY
Reliability looks like a "tight
pattern," regardless of where
the shots hit, because
reliability is a function of
consistency. Validity on the
other hand, is a function of
shots being arranged
around the bull’s eye. The
failure of validity is
reliable valid valid systematically off mark.
but not and but not Notice that neither an
valid reliable reliable unreliable nor an invalid
measure is likely to be
useful
Who Decides What's Valid?
• Validity began with a reminder that we depend on agreements to determine
what's real. Sometimes Social researchers implicitly assumes that they are
somewhat superior to those they study
Example: The biologist feels superior to the frog on the lab table