Methodology
Research Design
Variables
VARIABLES
“an observable characteristic of an object , event, or
people that can be described according to some well-
defined classification or measurement scheme” (Williams, 1986, p. 4).
a characteristic or attribute that has two or more mutually
exclusive values or properties (Ardales, 1992);
“building blocks of hypotheses”
Mutually exclusive – the attributes are distinct, one cannot be represented in the other, no overlap .
Classification of Variables
Quantitative : measured numerically
Qualitative: variables which can’t be measured
quantitatively
ex. Sex, civil status, stress, head injury, etc..
Characteristics of variables that are
necessary considerations in most research
Definition
Function
Type of measurement (scale)
Variables and their Definitions
An operational definition “assigns meaning to a construct or a
variable by specifying the activities or “operations” necessary to
measure it...It is a specification of the activities of the researcher
in measuring the variable or manipulating it” (Kerlinger, p. 28).
Types of operational definitions are:
(a) measured, “which describes how a variable will be
measured” and includes the source of the data (e.g., a specific
standardized instrument or author developed questionnaire)
(b) experimental, which “spells out the details of the
investigator's manipulation of the variable” (Kerlinger, 1986.
p. 29) (e.g., the specific details and procedures of the
intervention or treatment).
Variables and Their Functions
Functions of variables are most
frequently related to
presumed causality
the purposes of the inquiry
Presumed Causality
A. Variable functions related to presumed causality
include: Independent and Dependent Variables
Independent variable (X): “ the factor that is
manipulated or controlled by the researcher” (Marczyk et al. 2005, p.
42)
the one varied during the experiment;
presumed cause of the dependent variable
It is what causes or influences the outcome” (Marczyk et al., p. 46).
it predicts the relationships
Also referred to as Explanatory Variables
manipulated, antecedent, stimulus
Variables: Function Cont.
Dependent variable: (Y)– the one measured or
observed;
the presumed effect of the independent variable.
(Tuckman, 1988 ).
The term dependent implies “it is influenced by the
independent variable (Marczyk, et al, p. 46).
Also called Criterion, response, consequence, or
outcome variables.
outcome, consequence, response
Note that the dependent and independent classifications are not as
readily applicable to ex post facto studies in which relationships
rather than causality are studied. They are similarly not applicable to
descriptive studies.
Variables: Function Cont.
B. Variable functions related to the purposes of
inquiry include Moderator and Control.
Control variables are introduced to remove their influence
from the relationship of the other variables,
whereas, moderator variables are introduced to further
elucidate the nature of the relationships among the
variables.
Classification of Variables (Con’t)
Control Variable: any variable that is held constant in a
research study by observing only one of its instances or
levels.
Moderating Variable : Suspected or known to
impact / influence the DV
Ex:Does level of education determine how much
money customers spend on books?
MV: income levels of customers
Extraneous Variables
Extraneous variable / Exogenous variable; factors not
considered in the study but which affect/ influences in anyway
the study variables;
TEST FACTOR which when introduced, substantially changes or
causes the loss of the original bivariate (X and Y) relationship;
the original relationship between X and Y is called spurious
causal relationship.
The researcher should have control over it (through randomization
and matching); if beyond control, it should be taken into account
in the analysis and interpretation
Extraneous Variables
Number of policemen
Size of city
Extent of criminality
Number of schools
Rural/ Urban
Level of education
Antecedent variable- a variable previous to the
independent variable.
Ex:
Education SES of Family Class Performance
Parents’
Occupation
Intervening Variables
Intervening variable – also called in-the-head variable, internal
and unobservable psychological processes that account for
behavior (Kerlinger, 1986); when introduced, the original
relationship between the X and Y is lost or weakened.
Intervening Variable: a hypothetical variable that is not observed
directly in the research study, but is inferred from the relationship
between the independent and dependent variable.
Intervening variable – logical consequence of X and logical
cause of Y; when introduced, the original relationship between
the X and Y is lost or weakened.
Ind V Int.V DV
Business Attitude toward Loan
Income loan repayment repayment
How a Variable Becomes an Intervening Variable:
There’s a relationship between X and Y.
There’s a relationship between X and int. variable.
There’s a relationship between int. variable and Y
Suppressor or Distorter Variables
Suppressor variable – its presence could suppress the
relationship between the independent and
dependent variables.
Distorter variable – there is originally no relationship
between X and Y, but when the distorter variable is
introduced, there now exists a relationship between
X and Y.
Variables: Measurement Scales
There are two different scales for measurement of
variables.
1. Variables can be: continuous or categorical (Kerlinger, 1986)
AND
2. Variables can be nominal, ordinal, interval, or ratio
(Williams, 1986)
Continuous vs. Discrete vs. Categorical Variables
Continuous variable – can take on an ordered set of numerical
values within a certain range, e.g., age, distance, volume,
temperature, height, weight, calorie intake, etc.
Discrete variable – can take specific values only; can only have
specific value which has semblance of absoluteness; e.g., class
size, family size (can’t have 2.5 children).
Categorical variable – belongs to nominal measurement (classifies
observations by name only, does not connote any value, does not
give the idea that one is greater than the other); e.g., religion.
Quantification of Variables
Measurement: the application of rules in assigning
numbers to cases so as to represent the presence or
absence of quantity of an attribute possessed by
each case.
Four (4) scales of measurement
Scales of Measurement
Nominal Scale Measurement (Lowest)
Ordinal Scale Measurement
Interval Scale Measurement
Ratio Scale Measurement (Highest)
Variables measured at higher levels can be scaled
down to lower levels
Nominal and Ordinal Scale
Nominal is just a descriptive name - no value, no
ranking - like colors (red, blue, yellow) or gender
(male, female) and are mutually exclusive (for the
purposes of the study).
Ordinal - ranked highest to lowest. Cities ranked by
size of population. Baseball players ranked by
batting average.
Interval Scale
“Continuous variable” in which a number is assigned
to the amount of an attribute and IN WHICH ZERO
DOES NOT MEAN THE ABSENCE OF THE ATTRIBUTE
E.g. like temperature - 0 does not mean there is not
temperature. Test scores - zero doesn’t mean the
absence of intelligence/knowledge.
Ratio Scale
“Continuous variable” in which a number is assigned to
the amount of an attribute and IN WHICH ZERO HAS A
MEANING - like in the amount of money in your pocket -
you can’t go below zero. Number of children, amount of
food you have, height, weight, age
“Variables measured at higher levels can be scaled
down to lower levels” : you can rank or name ratio or
interval scales, making them ordinal or nominal, but not
vice versa. You can do more statistical processes with the
“higher” level variables.
