Chapter 3
Designing Research Concepts,
Hypotheses, and Measurement
Research Design
 Must create a Research Design
 Questions are composed of concepts
Must start with a research question
Stages of Research
Developing Concepts
Selection of Research Method(s)
Sampling Strategy
Data Collection ‘Plan’
Results and Writing
 Also need to consider budget issues
 It is critical to survey research to
understand how to go from ideas to
concepts to variables – operationalization.
 Concept (p.35): an idea, a general mental formulation
summarizing specific occurrences
 A label we put on a phenomenon, a matter, a “thing” that
enables us to link separate observations, make
generalizations, communicate and inherit ideas.
 Concepts can be concrete, abstract, tangible or intangible.
Concrete: Height, Major
Abstract: Happiness, Love
Transferring Concepts into
something Measurable
 Variable:
 A representation of concept in its variation of
degree, varieties or occurrence.
A characteristic of a thing that can assume
varying degrees or values.
 Fixed meaning = constant
 Most variables are truly variable = multiple categories or
Example: Concept and Variable
 Concept:
Political participation
 Variables:
 Voted or not
 How many times a person has voted
 What party a person votes for
How to be measured?
 Conceptualization:
conceptualization includes coming to
agreement about the meaning of the concept
 In practice, you often move back and forth between
loose ideas of what you are trying to study and
searching for a word that best describes it.
 Sometimes you have to “make up” a name to
encompass your concept.
 As you flush out the pieces or aspects of a concept,
you begin to see the dimensions; the terms that
define subgroups of a concept.
 With each dimension, you must decide on
indicators – signs of the presence or absence of
that dimension.
Dimensions are usually concepts themselves.
Operationalizing Choices
 You must operationalize: process of converting
concepts into measurable terms
The process of creating a definition(s) for a concept
that can be observed and measured
 The development of specific research procedures that will
result in empirical observations
SES is defined as a combination of income and education
and I will measure each by…
The development of questions (or characteristics of data in
qualitative work) that will indicate a concept
Variable Attribute Choices
 Variable attributes need to be exhaustive and
 Represent full range of possible variation
 Degree of Precision
selection depends on your research interest
 Is it better to include too much or too little?
 The dependent variable is the variable that the
researcher measures; it is called a dependent variable
because it depends upon (is caused by) the
independent variable.
 The independent variable is the one that the
researcher manipulates.
 Example: If you are studying the effects of a new educational program
on student achievement, the program is the independent variable and
your measures of achievement are the dependent ones.
 Qualitative Variable: Composed of categories which are
not comparable in terms of magnitude
 Quantitative Variable: Can be ordered with respect to
magnitude on some dimension
 Continuous Variable: A quantitative variable, which can be
measured with an arbitrary degree of precision. Any two
points on a scale of a continuous variable have an infinite
number of values in between. It is generally measured.
 Discrete Variable: A quantitative variable where values can
differ only by well-defined steps with no intermediate values
possible. It is generally counted.
Level of Measurement
 Nominal
 Ordinal
 Interval
 Ratio
Nominal Measures
 Only offer a name or a label for a variable
 There is not ranking
 They are not numerically related
 Gender; Race
Ordinal Measures
 Variables with attributes that can be rank ordered
 Can say one response is more or less than another
 Distance between does not have meaning
lower class, middle and upper class
 Note: Scales and indexes are ordinal measures, but
conventions for analysis allow us to assume equidistance
between attributes (if it makes logical sense); treat them like
“interval” measures; and subject them to statistical tests
Interval Measures
 Distance separating attributes has meaning and is
standardized (equidistant)
 “0” value does not mean a variable is not present
 Score on an ACT test 50 vs. 100
 does not mean person is twice as smart
Ratio Measures
 Attributes of a variable have a “true zero point” that
means something
 Waist measures and Biceps measures
 Allows one to create ratios
 Hypotheses: (pg. 36) Untested statements
that specify a relationship between 2 or more
 Example: Milk Drinkers Make Better Lovers
Characteristics of a Hypothesis
 States a relationship between two or more variables
 Is stated affirmatively (not as a question)
 Can be tested with empirical evidence
 Most useful when it makes a comparison
 States how multiple variables are related
 Theory or underlying logic of the relationship makes sense
 Hypotheses should be clearly stated at the
beginning of a study.
Do not have to have a hypothesis to conduct
research, general research questions.
Positive and Negative (Inverse)
 Positive: as values of independent variable
increase, the values of the dependent variable
 Negative: as values of independent variable
increase, the values of the dependent variable
decrease (or vice versa)
Two-directional Hypotheses
 More general expression of a hypothesis
 Usually default in stat packages
 Suggests that groups are different or concepts
related, but without specifying the exact direction
of the difference
Example: Men and women trust UK security differently.
One-directional hypotheses
 More specific expression of a hypothesis
 Specifies the precise direction of the
relationship between the dependent and
independent variables.
Example: Women have greater trust in UK
security compared to men.
Determining Quality of Measurement
 Accuracy and Consistency in Measurement
 Validity is accuracy
 Reliability is consistency
 Definition -- The extent to which the same
research technique applied again to the same
object (subject) will give you the same result
 Reliability does not ensure accuracy:
a measure can be reliable but inaccurate
(invalid) because of bias in the measure or in
data collector/coder
 Definition -- The extent to which our
measure reflects what we think or want them
to be measuring
Face Validity
 Face validity: the measure seems to be related to
what we are interested in finding out even if it does
not fully encompass the concept
concept = intellectual capacity
 measure = grades (high face validity)
 measure = # of close friends (low face validity)
Criterion Validity
 Criterion validity (predictive validity): the measure
is predictive of some external criterion
Criterion = Success in College
Measure = ACT scores (high criterion validity?)
Construct Validity
 Construct Validity: the measure is logically related
to another variable as conceptualized it to be
construct = happiness
measure = financial stability
if not related to happiness, low construct validity
Content Validity
 Content Validity: how much a measure covers a
range of meanings; did you cover the full range of
dimensions related to a concept
Example: You think that you are measuring
prejudice, but you only ask questions about race
 what
about sex, religious etc.?
Methodological Approaches,
Reliability and Validity
 Qualitative research methods lend themselves to
high validity and lower reliability.
 Quantitative research methods lend themselves to
lower validity and higher reliability