LIDE Learning Center, Inc.

Isabel, Leyte 6539

PRACTICAL RESEARCH 2
Second Semester, SY: 2023-2024

CH4. UNDERSTANDING DATA AND

WAYS TO SYSTEMATICALLY
COLLECT DATA
Melagrito, Ma. Lourdes F. and Diana J. Mendoza. Applied Research: An Introduction to
Quantitative Research Methods and Report Writing. Quezon City: Phoenix, 2016. Pages 101-
163.

Lesson 1: Measurement and Types of Data in Quantitative

Research (pp. 101-137)
 Guide Questions:
o What is measurement in quantitative research?
o What are levels of measurement or rating scale?
o What is the difference between measurement validity and measurement reliability?

WHY MEASURE?
o Quantitative research – examines precise concepts and issues that are the focus of their
studies
o Concepts – categories for the organization of ideas and observations (Bulmer 1984)
 Represent “a label that we give to elements of the social world that seem to
have common features and that strike us as significant.” (Bryman 2008, p. 143)
 E.g., Political Science – politics, power, state, government, nation, ideology, and
democracy
o Challenge of social science researchers – many of the phenomena that they study
involve abstract concepts that are not easily observable
 E.g., the political science terms above require you to specify clearly what you
mean when you use any of these concepts when doing quantitative research
o IF A CONCEPT IS TO BE USED IN QUANTITATIVE RESEARCH, THE CONCEPT WILL HAVE
TO BE MEASURED.
 2 tasks in the process of measurement - distinct but interrelated processes
 1. Conceptualization
 2. Operationalization

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 1 of 13
 o 1. Conceptualization – “the process through which we specify what we mean when we
use particular terms (concepts) in research” (Babbie 2010, p. 126)
 How? Don’t rely on Merriam-Webster dictionary!
 Examine the social theories and prior research relevant to the concept -> RRL
 Concepts - can be defined, approached, and explained in many ways; your task
is to specify how you are going to use it in your study
 Variable – a specific dimension or aspect of the concept
o 2. Operationalization – “the process through which we specify what we will observe”
that will indicate to use the presence or absence of the concept that we are studying
(Babbie 2010, p. 126)
 This moves our understanding of a concept from its abstract (theoretical) form
to its concrete (empirical/observable and measurable) form.
 It includes the creation or formulation of an indicator or a set of indicators that
will stand for and measure the concept or any of the dimensions of the concept,
the variable or variables
 Measure – “refer to things that can be relatively or unambiguously
counted”, i.e., income, age, number of children, number of years spent
in school (Bryman 2008, p. 145)
o Measures are quantities
 Indicator – “less directly quantifiable” and “something that is devised or
already exists and that is employed as though it were a measure of a
concept”
o Indicators are indirect measures of a concept
o Ways to devise indicators (Bryman):
 Through a question (or series of questions) that is part
of a structured interview schedule or self-completion
questionnaire (e.g., attitude -> job satisfaction, social
situation -> poverty, behavior - > leisure pursuits)
 Through the recording of individuals’ behavior using a
structured observation schedule (e.g., pupil behavior in
a classroom)
 Through official statistics, i.e., use of crime statistics to
measure criminal behavior
 Through an examination of mass media content through
content analysis, e.g., to determine changes in the
salience of an issue, i.e., AIDS
o SUMMARY: Why measure? 3 Roles:
 1. Allows us to delineate fine differences between people in terms of the
characteristics in question.
 2. Gives us a consistent device or yardstick for making distinctions.

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 2 of 13
  3. Provides the basis for more precise estimates of the degree of relationship
between concepts.

LEVELS OF MEASUREMENT
o “The values that an indicator can take but they say nothing about the indicator itself”
(Bhattacherjee 2012, p. 45)
o Variables – has categories of some sort, providing different types of information
 Discrete – each separate category represents a different status
 Continuous – a number represents a quantify that can be described in terms of
order, spread between the numbers, and/or relative amounts (Engel and Schutt
2012, p. 89)
o 4 Levels of Measurement
 1. Nominal – not quantitative
 2. Ordinal – quantitative
 3. Interval – quantitative
 4. Ratio – quantitative
o 2 important implications of knowing your variable’s level of measurement
 1. You can better understand the differences of your cases on a variable,
allowing you to understand more fully what you are actually measuring.
 2. You can identify the type of statistics or the type of statistical analysis that can
be applied to the variable.
o 1. Nominal Level of Measurement
 “Identifies variables whose values have no mathematical interpretation; they
vary in kind or quality, but not in amount.” (Engel and Schutt 2012, p. 89)
 E.g., gender – male or female; known as dichotomous variables,
variables with only two categories
 There is no order implied in the attributes of the nominal variable; no one
response is somehow higher or lower than another.
 The attributes that you use to categorize cases must be mutually exclusive and
exhaustive; when your variable meets these two requirements, every case
corresponds to one and only one attribute (Engel and Schutt 2012, pp. 90-91)
 Mutually exclusive – every case can have only one attribute
 Mutually exhaustive – every case can be classified into one of the
categories
 E.g., marital status – single, married, divorced, separated, widowed
 “other/s” – another category to put when you are unsure; captures the
attribute of every possible case
o 2. Ordinal Level of Measurement
 “The numbers assigned to cases specify only the order of the cases, permitting
greater than and less than distinctions.” (Engel and Schutt 2012, p. 89)
 Like nominal variables, these must be mutually exclusive and
exhaustive.

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 3 of 13
  E.g., level of education – high school, college, post-graduate
 Here, what matters is that you rank-order the categories; the distance between
categories does not matter (Babbie 2010, p. 138)
o 3. Interval Level of Measurement
 The variable values “represent fixed measurement units but have no absolute,
or fixed, zero point.” (Engel and Schutt 2012, p. 91)
 Also exclusive, exhaustive, and there is an order BUT the gaps between
the numbers are meaningful
 E.g., temperature and IQ tests
 Here, we can have 3 sets of data (Babbie 2010, p. 139):
 1. People are different from one another on tis variable
 2. One person is more than another person on this variable
 3. We can say “how much” more the person is than the other person
o 4. Ratio Level of Measurement
 Represents fixed measuring units and an absolute zero point (zero means
absolutely no amount of whatever the variable indicates).
 Ratio numbers can be added and subtracted; because the numbers
begin at an absolute zero point, they can be multiplied and divided (so
ratios can be formed between the numbers).
 E.g., ages – can be represented by values ranging from 0 years (or some
fraction of a year) to 120 or more.
 The numbers also are mutually exclusive, are exhaustive, have an order,
and have equal gaps. (Engel and Schutt 2012, p. 92)
 Ratio measures – have all the characteristics of interval measures BUT the
former have an absolute zero – a true point where zero really means the
absence of the attribute.
 E.g., age, income, length of residency, number of organizations, and
number of times you attend Church
o Types of Variables (Bryman 2008, p. 322 and Black 1999, p. 52)
 1. Ratio – the distances between the categories are identical across the range,
but there is an absolute zero and it has meaning – there is nothing there
 E.g., achievement test scores
 2. Interval – the distances between the categories are identical across the range
but there is no zero point where the trait does not exist
 E.g., IQ scores
 3. Ordinal – categories can be rank-ordered but the distances between the
categories are not equal across the range
 E.g., social class, opinions, job position
 4. Nominal – categories cannot be rank-ordered and have name value only;
must have at least 2 categories; binary – a subject either belongs or does not
belong

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 4 of 13
  E.g., gender, profession, school attended, country, etc.

SCALES AND INDEXES

o Levels of Measurement
 Aka. Rating scales – used to capture a respondent’s reactions to a given item in
the scale:
 Nominal scale item – captures yes/no response
 Interval-scale item – between “strongly agree” to “strongly disagree”
o 1. Binary Scale
 A nominal scale consisting of binary items that assume one of two possible
values
 e.g., yes or no, true or false, male or female, full-time or part-time
o 2. Likert Scale
 Designed by Rensis Likert, it is a popular rating scale for measuring ordinal data
in research in the social sciences
 The Likert items in included in the scale are simply-worded statements to which
respondents can indicate their extent of agreement or disagreement on a five-
or seven-point scale ranging from “strongly agree” to “strongly disagree.”
 The overall scale score is a summation of the attribute values of each item
o 3. Semantic Differential Scale
 A composite or multiitem scale where respondents are asked to indicate their
opinions to a statement using different pairs of adjectives framed as polar
opposites
 Like Likert scale, the overall score is a summation of individual item scores
 Unlike Likert scale, the statement remains constant but the anchors (adjective
pairs) change across items
o 4. Guttman Scale
 Designed by Loius Guttman, this is another type of composite or multiitem scale
that uses a series of items arranged in increasing order of intensity of the
concept of interest, from least to most intense.
 Each item in the series has a weight which varies with the intensity of that item
 The weighted summation of each response is used as the aggregate measure of
an observation.

MEASUREMENT VALIDITY AND RELIABILITY

o Validity – the issue of whether an indicator (or set of indicators) that is devised to gauge
a concept really measures that concept
o Reliability – the consistency of a measure of a measure of a concept
o Types of Validity Tests or Techniques
 1. Face validity – the measure apparently reflects the content of the concept in
question

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 5 of 13
  Done by asking the experts of the field if the concept’s measure
captures what the concept is or what the concept means.
 2. Concurrent validity – requires using a criterion on which cases (e.g., people)
are known to differ and that is relevant to the concept in question.
 E.g., absenteeism – measures job satisfaction
 3. Predictive validity – a researcher uses a future criterion measure rather than
a contemporary one, as in concurrent validity
 4. Construct validity – a researcher is encouraged to deduce hypotheses from a
theory that is relevant to the concept
 E.g., examining the link between job satisfaction and job routine
o Satisfied – less likely to work on routine jobs
o Unsatisfied – more likely to work on routine jobs
 5. Convergent validity – done by comparing s new measure to measures of the
same concept developed through other methods
 E.g., measure how much time managers spend on various activities
 Potential Problem - it is not necessarily easy to establish which of the
two measures represents the more accurate picture or a definitive, valid
measure.
o 3 sources of unreliability
 1. The observer’s subjectivity
 2. Asking ambiguous questions
 3. Asking about issues that are respondents are not very familiar with or care
o How to create reliable measures
 1. Use questionnaires (objective) rather than observations (subjective)
 2. Ask questions on issues that the respondents know
 3. Avoid ambiguous measures
 Simplify the wordings in your indicators
o Types of Reliability Tests or Techniques
 1. Inter-rater reliability – aka. Inter-observer reliability; a measure of
consistency between two or more independent raters (observers) of the same
construct (concept).
 Raters check off which category each observation falls in, and the
percentage of agreement between the raters is an estimate of inter-
rater reliability.
 2. Test-retest reliability - a measure of consistency between two measurements
(tests) of the same construct administered to the same sample at two different
points in time.
 If the observations have not changed substantially between the two
sets, then the measure is reliable.
 3. Split-half reliability – a measure of consistency between two halves of a
construct measure.

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 6 of 13
  If you have a 10-item measure of a given construct, randomly split the
10 into 2 sets of 5 (unequal halves are not allowed)
 Administer the entire instrument to a sample of respondents
 Calculate the total score of each half for each respondent
 The correlation between the total scores for each half is a measure of
split-half reliability.
 4. Internal consistency reliability – a measure of consistency between different
items of the same construct
 If a multiple-item construct measure is administered to respondents,
the extent to which respondents rate those items in a similar manner is
a reflection of internal consistency.
o Random error – the error that can be attributed to a set of unknown and uncontrollable
external factors that randomly influence some observations but not others
 E.g., respondents with nice and bad moods
o Systematic error – an error that is introduced by factors that systematically affect all
observations of a construct across an entire sample.
 E.g., a financial crisis impacted the performance of financial firms
disproportionately more than other types of firms

Lesson 2: Sampling and Sampling Probability (pp.

138-145)
 Guide Questions:
o What is a sample?
o What is probability sampling in quantitative research?
o What are the different probability sampling techniques in quantitative research?

SAMPLING AND PROBABILITY SAMPLING

o SAMPLING - “the statistical process of selecting a subset (called a sample) of population
of interest for purposes of making observations and statistical inferences about that
population.”
o Population – “the group you want to generalize to”
 It is impossible to study entire populations due to feasibility and cost concerns.
o Sample – “the actual units selected for observation”
 It is important that the sample is truly a representative of the population.
o RANDOM SAMPLING – the only one used in quantitative research; defined as “that
method of drawing a portion (sample) of a population so that all possible samples of a
fixed size have the same probability of being selected.”
 Aka. PROBABILITY SAMPLING - “a technique in which every unit in the
population has a chance (nonzero probability) of being selected in the sample.”
o 3 Stages of the Sampling Process

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 7 of 13
  1. Defining the target population – using a unit of analysis, i.e., person, group,
organization, country, object, etc.
 2. Choosing a sampling frame – which is the list from which you can draw your
sample
 3. Choosing a sample from the sampling frame using a well-defined sampling
technique which are:
 A. Probability (random) sampling
 B. Nonprobability sampling
o Types of Probability Sampling Techniques
 1. Simple Random – from whole population; highly representative if all subjects
participate; not possible without complete list of population members
 2. Stratified Random – from identifiable groups (strata), subgroups, etc.; can
ensure that specific groups are represented, even proportionally, in the sample;
this requires greater effort and the strata must be clearly defined
 3. Cluster – random samples of successive clusters of subjects until small groups
are chosen as units; possible to select randomly when no single list of
population members exists but local lists do; clusters in a level must be
equivalent but some natural ones are not for essential characteristics
 4. Stage – combination of cluster (randomly selecting clusters) and random or
stratified random sampling; can make up probability sample by random at
stages and within groups; but is complex and combines limitations of cluster and
stratified random sampling

SIMPLE RANDOM SAMPLING

o This is the simplest and yet the most generalizable among all probability sampling
techniques because the sampling frame is not subdivided and the sample is not biased.
o Main difficulty – “obtaining a list of all the members of the population… from which to
select a sample”
o 2 ways to generate a random sample
 1. Random numbers table
 2. Computerized random number generator

Lesson 3: Types of Quantitative Data Collection (pp. 149-

157)
 Guide Questions:
o What are some of the data collection techniques in quantitative research?
o What are the strengths and weaknesses of these techniques?

SURVEY RESEARCH
o A research method involved the use of standardized questionnaires or interviews to
collect data about people and their preferences, thoughts, and behaviors in a systematic
manner.

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 8 of 13
 o Strengths:
 1. Excellent vehicle for measuring a wide variety of unobservable data
 2. Ideally suited for remotely collecting data about a population that is too large
to observe directly
 3. Questionnaire surveys are preferred by some respondents due to their
unobtrusive nature
 4. Interviews may be the only way of reaching certain population groups
 5. Large sample surveys may allow detection of small effects even while
analyzing multiple variables.
 6. Survey research saves researcher time, effort, and cost than other methods
o 2 ways of conducting surveys
 1. Questionnaire surveys (written) – may be mail-in, group-administered, or
online surveys
 2. Interview surveys (verbal) – may be personal, telephone, or focus group
interviews

EXPERIMENT
o One of the most rigorous of all research designs
o 3 key features:
 1. One or more independent variables are manipulated by the researcher (as
treatments)
 2. Subjects are randomly assigned to different treatment levels (random
assignment)
 3. The results of the treatments on outcomes (dependent variables)
o This is best suited for explanatory research (rather than for descriptive or exploratory
research) – the goal of the study is to examine the cause-and-effect relationships and
independent variables that can be manipulated or controlled
o 2 Possible Settings
 1. Laboratory experiments – tend to be high in internal validity but low in
external validity (generalizability)
 The artificial (laboratory) setting in which the study is conducted may
not reflect the real world
 2. Field experiments – high in both internal and external validity; e.g., done in a
real organization
 But these are relatively rare because of the difficulties in manipulating
treatments and controlling extraneous effects in a field setting
o 2 Broad Categories of Experimental Research – both require treatment manipulation
 1. True experimental designs – require random assignment
 2. Quasi-experimental designs – doesn’t require random assignment
o Basic concepts in experimental research
 Treatment and control groups

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 9 of 13
  Treatment group – a treatment, an experimental stimulus, is
administered to them (many treatments can be given)
 Control group – they are not given the stimulus
 MEASURE OF SUCCESS – if the treatment group rate more favorably on
outcome variables than the control group
 Treatment manipulation
 This helps control for the “cause” in the cause-effect relationships.
 MEASURE OF SUCCESS – the validity of experimental research depends
on how well the treatment was manipulated
 Checked in 2 steps:
o 1. Pretest – measurements done before the treatment
o 2. Posttest – measurements done after the treatment
 Random Selection and assignment
 Random selection – process of randomly drawing a sample from a
population or a sampling frame
o More closely related to external validity (generalizability) of
findings
 Random assignment – a process of randomly assigning subjects to
experimental or control groups
o Related to internal validity since it is related to design
 Experimental research – can use either or both
 Quasi-experimental research – doesn’t use neither
o Strength
 Its internal validity (causality) that is due to its ability to link cause and effect
through treatment manipulation while controlling for the spurious effect of
extraneous variable.
o Weaknesses
 1. Without theories, the hypotheses being tested tend to be ad hoc, possibly
illogical, and meaningless
 2. Many measurement instruments are not tested for reliability and validity
 3. Many experimental research use inappropriate research designs; these lack
internal validity and are highly suspect
 4. The treatments (tasks) used in experimental research may be diverse,
incomparable, and inconsistent across studies and sometimes inappropriate for
the subject population

OFFICIAL STATISTICS
o Advantages
 1. The data have already been collected
 2. The problem of reactivity will be much less pronounced than when data are
collected by interview or questionnaire

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 10 of 13
  3. Precisely because the data are compiled over many years, it is possible to
chart trends over time and perhaps to relate these to wider social changes
 4. There is the prospect as well of cross-cultural analysis since the official
statistics from different nations can be compared for a specific area of activity
 5. Unobstrusive measure or method – “any method of observation that directly
removes the observer from the set of interactions… being studied.”

CONTENT ANALYSIS
o “An approach to the analysis of documents and texts that seeks to quantify content in
terms of predetermined categories and in a systematic and replicable manner”
o Quantitative Model (Deductive)
 1. Research question and hypothesis
 2. Conceptualization – what valuables are used and how they will be defined
 3. Operational measures – aimed at gaining internal validity and face validity
 3a. unit of analysis
 3b. measurement
 Categories can be exhaustive and mutually exclusive or a priori
 4. Coding
 5. Sampling – random sampling a subset of content
 6. Reliability
 Can use: two codes for intercoder reliability or computer program for
validation
 7. If reliability was determined by hand (step 6) then apply a statistical check
 8. Tabulation and representation
o Its Contributions by Social Scientists
 1. Quantitative researchers have shown the value of the method in gaining
“hard data.”
 2. Examining patterns and themes within the objects produced in a given
culture, analyzing preexisting data to expose and unravel macro processes, and
presenting their findings on easily readable charts and tables has been
significant when getting public attention.
 3. Became the standard method for analyzing the role of mass-produced texts in
the socialization process.
 4. Helps shape social policy by calling attention to systematic inequalities in
need of change

Lesson 4: Use of Statistical Analysis in Quantitative

Research (pp. 158-163)
 Guide Questions:
o What is the use of statistics in quantitative research?
o What is the difference between descriptive and inferential statistics?

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 11 of 13
STATISTICS
o Defined in the following ways:
 1. The overall domain concerned with the mathematical treatment of variability.
 2. A subset of theoretical or mathematical statistics - concerned with the use of
mathematical principles and probability theory in the development and testing
of methodology for the treatment of variability.
 3. A subset of applied statistics – use of already developed and accepted
statistical methodology as an aid in the research effort
 4. The specific procedures developed by statisticians for the mathematical
treatment of variability.
o The textbook’s definition – statistics is concerned with (1) the application of accepted
statistical procedures as an aid in the research effort; (2) as specific statistical
procedures used in modern research.
o 2 Levels of Statistics
 1. Descriptive
 2. Inferential

1. DESCRIPTIVE STATISTICS
o Definition – “to statistically describing, aggregating, and presenting the constructs of
interest or associations between these constructs.”
 The Researcher – “stops after looking at a finite data set to numerically
delineate some variable or to portray, explain, or predict variability.”
 Goal – treatment of variability in the case of samples
o 2 Types of Analysis
 A. Univariate – analysis of a single variable; a set of statistical techniques that
can describe the general properties of one variable; it includes:
 i. Frequency distribution – summary of the percentages of individual
values or ranges of values for that variable
 ii. Central tendency – an estimate of the center of a distribution of
values
o mean – average
o median – middle
o mode – most frequent
 iii. Dispersion – the way values are spread around the central tendency
 B. Bivariate – examines how 2 variables are related to each other
 i. Bivatiate correlation (aka. “correlation”) – most common bivariate
statistic; “the statistical procedure associated with the correlation
model when 2 variables are involved”
 ii. Cross-tabulation (aka. “cross-tab” or “contingency table”) – describes
the frequency of all combinations of 2 or more nominal or categorical
variables.

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 12 of 13
2. INFERENTIAL STATISTICS
o Definition – “the statistical testing of hypotheses (theory testing)” or “the statistical
procedures that are used to reach conclusions about associations between variables”
 The Researcher – attempts to see how far she can go in generalizing statistical
findings to groups other than those examined
 Goal – test the validity of generalizations from samples to populations
o Hypothesis testing – you test generalizations from samples to populations
o Hypothesis – “states something about whether what is true of the sample is also true of
the population”
 2 Forms of hypothesis
 i. null form
 ii. alternative form
o A. Two-group comparison – one of the simplest inferential analyses; “done by
comparing posttest outcomes of treatment and control group subjects in a randomized
posttest only control design”
o B. Factor analysis – a data reduction technique that is used to statistically aggregate a
larger number of observed measures (items) into a smaller set of unobserved (latent)
variables, called factors, based on their underlying bivariate correlation patterns.
o C. Discriminant analysis – a classificatory technique that aims to place a given
observation in one of several nominal categories based on a linear combination of
predictor variables.
o D. Logistic regression or logit model – a general linear model (GLM) in which the
outcome variable is binary (0 or 1) and is presumed to follow a logistic distribution.
o E. Probit regression/model – is a GLM in which the outcome variable can vary between
0 and 1 (or can assume discrete values 0 and 1) and is presumed to follow a standard
normal distribution.
o F. Path analysis – a multivariate GLM technique for analyzing directional relationships
among a set of variables.
o G. Time series analysis – a technique for analyzing time series data or variables that
continually changes with time

Lesson 1: Measurement and Types of Data in Quantitative Research (pp. 101-137) Page 13 of 13