These slides discuss about the concept and definition of variables, variables in research, operationalisation, types and functions of variables and measurement scales.
2. Outline of today’s presentation
1. The concept and definition of
variable
2. Variables in research
3. Constructs versus variables
4. Operationalization
5. Types and functions of variables
6. Measurement Scales
3. Variables
It is very important
in research to see
variables, define
them, and control
or measure them.
4. The concept of variable
The concept of variable is basic but
very important in research. You will
not be able to do very much in
research unless you know how to
deal with variables.
5. The concept of variable
A variable is a measured characteristic that
can assume different values or level.
A measure that has only one value is called
a constant.
A variable can be defined as an attribute of a
person, a piece of text, or an object which
“varies” from person to person, text to text,
object to object, or from time to time.
5
6. Variables in the classroom
An EFL Student’s language skill may vary from
week to week.
The ability to speak a variety of languages. Some
people are monolingual, others are bilingual, and
others multilingual.
IQ Scores, reading speed, accuracy, fluency,
proficiency.
6
7. Some examples
Age can be considered a variable
because age can take different values
for different people or for the same
person at different times.
Similarly, country can be considered
a variable because a person's country
can be assigned a value.
8. Some examples
Grade level can be considered a
variable because Grade level can
take different values for different
people or for the same person at
different times.
Height, gender, e.t.c.
9. Variables in research
Variables are things that we measure,
control, or manipulate in research.
Variables can be very broad or very
narrow. For example, the discourse,
semantic, syntactic, phonological
elements of language are attributes of
language.
They are also something attributed to
people in varying degree of proficiency.
10. Variables in research
A variable such as phonological
system is broad, indeed, when
assigned to Students. The variable
rising tone is less so.
11. Remember
The broader the variable, the
more difficult it may be to define,
locate and measure accurately.
The more specific a variable is,
the easier it will be to locate and
measure.
12. Operationalization
Variables such as intelligence,
motivation, and academic achievement
are concepts, constructs, or traits that
cannot be observed directly.
They should be stated in precise
definitions that can be observed and
measured. This process is called
operationalization.
13. Operationalization
Intelligence
Trait or construct
Scores on the
Wechsler Adult
Intelligence
Scale
Operational definition of
intelligence
operationalization
15. Operational definition of a variable
With students’ intelligence scores or TOEFL
scores, we now have observable and
quantifiable definitions of what the
researcher means by the constructs of
“intelligence” and “proficiency”.
This is an operational definition of the
variable.
16. Important point!
Operational definitions must be based
upon a theory that is generally
recognized as valid.
For example, to operationalize the
construct of “proficiency” we should
construct a test based on an accepted
theory or model of language
proficiency.
17. Different types and functions
of variables
In addition to knowing how constructs
are operationalized as variables, it is
important to understand how such
variables are classified and
manipulated by researchers in their
quest to empirical knowledge.
To that end, we describe five different
functions of variables.
18. Functions of variables
To assess the relationship between
variables in research, we must be able to
identify each variable. Variables can be
classified as:
1.Independent
2.Dependent
3.Moderator
4.Control
5.Intervening
19. Independent vs. Dependent
Variables
An important distinction having to do with the
term 'variable' is the distinction between an
independent and dependent variable.
This distinction is particularly relevant when
you are investigating cause-effect
relationships (experiment). However, the
concept is also used in other research
designs.
20. Independent vs. dependent V.
In fact the independent variable
is what you (or nature)
manipulates -- a treatment or
program or cause. The dependent
variable is what is affected by
the independent variable -- your
effects or outcomes.
21. Independent Variables
The independent variable is the major
variable which you hope to investigate. It is
the variable which is selected, manipulated,
and measured (its effect) by the researcher.
Examples:
The effect of your instruction on reading
scores of your students.
The effect of social class on language use.
22. Dependent variable
The dependent variable is the variable
which you observe and measure to
determine the effect of the
independent variable.
In the previous examples, the reading
scores of your students and the use of
language would be the dependent
variable.
23. Two points
1. A variable that functions as a
dependent variable in one study may
be an independent variable in another
study.
2. Depending on the design of the study,
we may have more than one
independent and even more than one
dependent variable in the study.
24. Moderator variable
A moderator variable is a special type
of independent variable which you may
select for study in order to investigate
whether it modifies the relationship
between the dependent and
independent variables.
Example, gender in the study of the
effect of instruction on students’
reading scores
25. Independent vs. moderator
variable
The essential difference between
independent and moderator variables lies
in how the researcher views each in the
study.
For independent variables, the concern is
with their direct relationship to the
dependent variable, whereas for
moderator variables, the concern is with
their effect on that relationship.
26. Suppose you were investigating the effect of
conversation practice on the speaking fluency
of foreign students. Conversation practice,
then , would be the independent variable that
you are interested in investigating. Fluency,
operationally defined, is the dependent variable.
However, you may have a hunch (feeling) that
conversation practice works better for your
Spanish students than for your Chinese
students. Or you may have a hunch that it
works better for men than for women or vice
versa. Thus, language and/or gender could be
moderator variable. 26
27. Control variables
It is virtually impossible to include all the
potential variables in each study. As a result,
the researcher must attempt to control, or
neutralize, all other extraneous variables
that are likely to have an effect on the
relationship between the independent,
dependent, and moderator variables.
28. Control variables
Control variables, then, are those that
the researcher has chosen to keep
constant, neutralize, or otherwise
eliminate so that they will not have an
effect on the study.
Example, the effect of outside practice
on reading in the previous example.
29. Intervening variables
Intervening variables are constructs
(other than the construct under study)
that may explain the relationship
between independent and dependent
variables but are not directly
observable themselves.
We are somehow aware of their
effects, but we are not able to account
for them.
30. Intervening variables
Usually the effect of the independent
variable on the dependent variable is
shown in terms of scores, counts, time
measurement, etc.
That is, the dependent variable is
measured is some way to determine
the effect of the independent variable.
31. Intervening variables
However, there is a process underlying
the behavior we are measuring which
is usually neither observable nor
measurable.
32. Intervening variables
For example, in the study of oral fluency,
oral fluency is measured. We have not,
however, said anything about the
process underlying the acquisition of
fluency. A number of variables have not
been measured which may or may not
be part of that process – learning ,
intelligence, frustration. These have not
been measured or manipulated. These
are called intervening variables.
33. The relationship among variables
Independent
Variable(s)
Dependent
Variable(s)
Intervening
Variable(s)
Moderator
Variable(s)
Control
Variable(s)
The Study
34. Two points
When designing a study, the
researcher determines which
variables fall into each category.
In real situations, all five types of
variables may not be included in all
studies.
35. Measurement
Measurement is defined as assigning
numbers to observations according to
certain rules.
Lyle F. Bachman (1990:19) explains
that measurement is the process of
quantifying the characteristics of an
object of interest according to explicit
rules and procedures.
36. Measurement Scales
To measure different variables, we
have four measurement scales:
1. Nominal Scale
2. Ordinal Scale
3. Interval Scale
4. Ratio Scale
37. Measurement Scales
For all four scales we use numbers, but the
numbers in each scale have different
properties and should be manipulated
differently.
It is the duty of the researcher to
determine the scale of the numbers used
to quantify the observations in order to
select the appropriate statistical test that
should be applied to analyzed the data.
38. Nominal Scale
Nominal scale classifies persons or
objects into two or more categories.
Members of a category have a
common set of characteristics, and
each member may only belong to one
category. Other names: categorical,
discontinuous, dichotomous (only two
categories).
39. Nominal Scale
In nominal scales, numbers are used
to label, classify, or categorize data.
For example, in coding data from a
survey to facilitate computer analysis,
boys may be coded as “1” and girls as
“2”. In this instance, it clearly does not
make sense to add or divide the
numbers.
40. True vs. artificial categories
True categories are those to which the
member naturally falls, such as gender
(male vs. female).
Artificial categories are those to which
the researcher places the members,
such as learning style (field
independent versus field dependent).
41. Ordinal Scale
Ordinal variables allow us to rank order the
items we measure in terms of which has less and
which has more of the quality represented by the
variable, but still they do not allow us to say "how
much more.“
Example: Ranking students
This scale has the concept of less than or more
than.
The three medals winners in the long jump at
the Olympic Games. The gold medalist
performed better than the silver medalist. The
silver medalist performed better than the bronze
medalist.
42. Ordinal Scale
Ordinal scales both classify
subjects and rank them in terms of
how they possess the characteristic
of interest. Members are placed in
terms of highest to lowest, or most
to least. Students may be ranked by
height, weight, or IQ scores. Ordinal
scales do not, however, state how
much difference there is between
the ranks.
43. Interval Scale
Interval scales have the same properties as
ordinal scales, but they also have equal intervals
between the point of the scale.
Not only rank order the items that are measured,
but also to quantify and compare the sizes of
differences between them.
For example: students performance on a spelling
test A score of 16 will be higher than 14 and lower
than 18 and the difference between them is 2
points (equal intervals).
Interval scales normally have an arbitrary
minimum and maximum point. A score of zero in a
spelling test does not represent an absence of
spelling knowledge, nor does a score of 20
represent perfect spelling knowledge.
44. Table 1 Three Example Scales
Students Test Scores
(Interval)
Ranking
(Ordinal)
Frequencies
(Ordinal)
Remarks
Rosidi 97 1 1 “Top Group”
Milano 85 2 1
Liana 82 3 1
Dean 71 4 1
Heni 70 5.5 2 “Upper Middle
Billy 70 5.5 Group”
Komar 69 7 1
Randi 68 8 1
Monika 67 10 3 “Lower Middle
Wendi 67 10 Group”
Herman 67 10
Sena 66 12 1 “Lower Group”
Jeni 62 13 1
Elizabeth 59 14 1
Ardi 40 15 1
Linda 31 16 1 44
45. Ratio Scale
Very similar to interval scale; Ratio scale has all
the properties of interval variables, it has absolute
zero point. Height, weight, speed, and distance are
examples of ratio scales. Measurements made
with ratio scales can be added, subtracted,
multiplied, and divided. For example, we can say
that a person who runs a mile in 5 minutes is twice
as fast as a person who runs the mile in 10
minutes. Because ratio scales are often used in
physical measurements (where absolute zero
exists), they are not often employed in educational
research and testing.
46. Table 2 Four Scales of Measurement
Name
Categories
Shows
Ranking
Gives
Distances
Ratio Make
Sense
Nominal
Ordinal
Interval
Ratio
46
47. Remark:
The table shows that nominal scale name and
categorize only, while ordinal scales uses categories
but also give the ranking, or ordering of points within
categories.
Interval scales provide information about the
categories and ordering but also give additional
details about the distances, or intervals, between
points in that ranking.
Finally, ratio scales give the intervals, between
points in the ordering of certain categories, but with
even more information, because the ratio scales
have a zero, and points along the scale make sense
as multiples or ratios of other points on the scale.
47
48.
49. Table 3 Properties of Measurement Scales
from Agresti & Finlay , 1986:16)
49
Measurement
Scales
Properties Ways of comparing
measures
Typical examples
Ratio
continuous
Interval
continuous
Ordinal
continuous
Nominal
discrete
Absolute zero
Equal intervals
Ordering
Distinctiveness
Equal intervals
Ordering
Distinctiveness
Ordering
Distinctiveness
Distinctiveness
How many times larger ?
How much larger?
Which one is larger?
Are they different?
How much larger?
Which one is larger?
Are they different?
Which one is larger?
Are they different?
Are they different?
Age of length of
residence, cost
per student,
number of hour
spent in study
Test scores,
attitude scales
Ranking,
judgments or self-assessment,
using ratings
scales, grade or
level in school
Native language,
occupation,
classroom in
school
50. References
Main Sources
Coolidge, F. L.2000. Statistics: A gentle introduction. London: Sage.
Kranzler, G & Moursund, J .1999. Statistics for the terrified. (2nd ed.). Upper Saddle
River, NJ: Prentice Hall.
Butler Christopher.1985. Statistics in Linguistics. Oxford: Basil Blackwell.
Hatch Evelyn & Hossein Farhady.1982. Research design and Statistics for Applied
Linguistics. Massachusetts: Newbury House Publishers, Inc.
Ravid Ruth.2011. Practical Statistics for Educators, fourth Ed. New York: Rowman &
Littlefield Publisher, Inc.
Quirk Thomas. 2012. Excel 2010 for Educational and Psychological Statistics: A Guide
to Solving Practical Problem. New York: Springer.
Other relevant sources
Agresi A, & B. Finlay.1986. Statistical methods for the social sciences. San Francisco,
CA: Dellen Publishing Company.
Bachman, L.F. 2004. Statistical Analysis for Language Assessment. New York:
Cambridge University Press.
Field, A. (2005). Discovering statistics using SPSS (2nd ed.). London: Sage.
Moore, D. S. (2000). The basic practice of statistics (2nd ed.). New York: W. H.
Freeman and Company.
Thursday, October 30, 2014