Research Methods in Applied Linguistics

Chapter 6
Characteristics of a Variable
1. What is a variable?
A variable is defined as an attribute which changes from person to person, object to object, place to place, or time
to time.

2. Talk about kinds of variables briefly. Provide examples.

1. concrete variables can be measured objectively, such as height and size

2. abstract variables cannot be directly measured, such as knowledge, happiness, and motivation

3. discrete variables are of all-or-nothing nature, for instance, nationality and left-handedness

4. continuous variables can range from a minimum point up to a maximum point, such as height and population

3. What are the categorizations of different kinds of variables?

1. Variables can be discrete and at the same time concrete or abstract.

a. For instance, left-handedness is a variable which can be considered discrete and at the same time concrete.

b. While cognitive style is a discrete and at the same time an abstract variable.

2. Variables can be continuous and at the same time concrete or abstract

a. intelligence is a continuous and at the same time an abstract variable;

b. whereas height is a continuous and at the same time a concrete variable.

4. Complete the following:

The processes of narrowing down the topic and manageability rest upon the ………. and ………. of variables. That
is, by reducing the number of variables, the topic of research is ………., and by ………. of the variables, the topic
becomes manageable.
Researchers try to specify the variables by defining them ………..
1. number, 2. nature , 3. narrowed down, 4. specifying the characteristics, 6. as clearly and objectively as possible

5. Define the variables from two points of view:

A variable should be defined from two different perspectives: theoretical and operational.
1. theoretical: what the theory behind the variable is. A researcher should study the parameters of different

theories in order to come up with an acceptable theoretical definition for the variable.
2. An operational definition deals with the variable in terms of its measurable characteristics. That is, a variable is

defined operationally when the way it can be measured is clarified.

6. What are the measurement scales of variables? (4 measurements)

1. Nominal Scales
Some variables are discrete, one cannot measure how much of the variable exists. A researcher can identify

1|P a g e
subjects by assigning some arbitrary numbers to them. These numbers do not carry any mathematical values or
imply quantitative superiority. They are assigned only to name different levels of the variable.
2. Ordinal Scales
Certain variables cannot be easily measured. In such cases, the extent of the existence of a variable is rank-ordered
through ordinal scaling – people or objects are ranked on a particular variable along a scale from high to low or
vice-versa. Different cut-off points on the scale can be identified by a number. These numbers refer to the
superiority or inferiority of the class to which they are assigned. However, they do not specify the difference
between categories accurately or mathematically. In other words, the distances among the numbers are not
necessarily equal.
3. Interval Scales
Interval scaling determines how much of an attribute exists. The distances among the ranks are equal and have
mathematically specifiable values. In other words, the units of intervals are equal and constant from one class to
another. Due to this characteristic, interval scaling is the most objective scale of measurement in research.
4. Ratio Scales
It determines how much of an attribute exists and the intervals are equal too. The ratio scale has a true zero point.
This means that through a ratio scale, one can determine the negative values as well.

 Since the concept of true zero is exclusive to natural sciences, the ratio scale is often used in natural rather
than social sciences because the latter deal with human characteristics and traits. The minimum that can be
claimed in human attributes is the nonexistence of the attribute which is indicated by zero.

7. What are the functions of variables?

Independent variable is the variable that does not depend on any other variable and it is under the control of the

researcher. It is selected, manipulated, and measured by the researcher.

Dependent variable is the variable that the researcher observes and measures to determine the effect of the

independent variable.
Moderator variable is the variable selected in order to investigate whether the results of research are modified

because of this variable or not. It modifies the relationship between the independent and dependent variables.

 A moderator variable is similar to the independent variable because it influences the outcome of research,
but it is different from the independent variable because it cannot be manipulated by the researcher.

Control variable is a variable which is held constant in order to neutralize the potential effect it might have on the

outcome of research.
Intervening variable is the variable that stands between the independent and dependent variables and cannot be

measured or observed.
Provide an example which clarify the functions of variables.
A researcher is interested in investigating the relationship between the learning of vocabulary and fluency in
speaking among students of similar intelligence but different sexes. So, he makes the following directional
hypothesis:

2|P a g e
Among students with similar intelligence quotient (IQ), boys with instruction on vocabulary will perform better on
speaking tasks than boys without instruction while such a difference will not appear among girls.

In this hypothesis, the variable dealing with instruction, i.e., teaching vocabulary, is the independent variable.
The variable which is directly influenced by the independent variable, i.e., speaking tasks, is the dependent
variable. The variable which is hypothesized to change the effect of the independent variable on the dependent
variable, i.e., the sex, is the moderator variable. The variable which is held constant, i.e., the intelligence quotient,
is the control variable. And the variable which is not observed but is the underlying cause of performance on the
dependent variable, i.e., learning, is the intervening variable.

 It should be mentioned that like variable scales which are convertible to one another, variable functions are
research-dependent as well. That is, a variable does not, by nature, function as a dependent or an
independent. The function of a variable is determined by the researcher and the research method.

3|P a g e