Facebook Twitter RSS Reset

Variables in Research – Dependent, independent, confounding and other types

VARIABLES

  • In quantitative studies, concepts are usually called variables.
  • A variable, as the name implies, is something that varies. e.g. weight, anxiety and blood pressure are variables, each varies from one person to another.
  • Most aspects of the humans are variables.
  • If every one weighted 150 pounds, weight would not be a variable, it would be a constant.
  • Quantitative researchers seek to understand how and why things vary and to learn if differences in one variable are related to differences in another.
  • A variable, then, is any quantity of a person, group, or situation that varies or takes on different values.
  • Variables are building blocks of quantitative studies.

HETEROGENEITY AND HOMOGENEITY

  • When an attribute is extremely varied in the group under study, the group is heterogeneous with respect to that variable.
  • If the amount of variability is limited, the group is homogeneous.
  • Degree of variability or heterogenicity of a group of people has implications for study design.

VARIABLES – INHERENT OR CREATED?

  • Variables are inherent characteristics of people, such as their age, blood type, or weight.
  • But sometime, researchers create a variable for conducting the studies. E.g. In testing the effectiveness of pain management methods, different methods used are variables.

TYPES OF VARIABLES

Continues variables

They have a wide range of values.The values are not restricted to the whole numbers. They have values along a continuum.They can assume an infinite number of values between two points.Weight is a continues variable: between 1 and 2 pounds, the number of values is limitless: 1.05, 1.8, 1.333 and so on.

Discrete variable

It has a finite number of values between any two points, representing discrete quantities.For example, if people were asked how many children they have, they might answer 0,1,2,3 or more. The value of number of children is discrete, because a number such as 1.5 is not meaningful. And also between 1 and 3 the only possible value is 2. It means that in this variable, the values are restricted to the whole numbers.

Catagorical variables

These variables take on a small range of values that do not represent a quantity. They take a handful of discrete nonquantitative values. Blood type, for example, has four values- A, B, AB and O.

Dichotomous variable

When catagorical variables take on only two values, they are called as dichotomous variables. Gender, for example, is a dichotomous: male and female.

Dependent and independent variables

Many studies seek to understand the causes of phenomena as: Does smoking cause lung cancer? The presumed cause is the independent variable and the presumed effect is the dependent variable.

Outcome variable

the variable capturing the outcome of interest.

Dependent  and Independent Variables

  • Variability in the dependent variable is presumed to depend upon the variability in the independent variable.
  • For example, researcher study to which extent lung cancer (dependent variable) depends on smoking(independent variable).
  • Most dependent variables have multiple causes or antecedents. e.g. if we are studying factors that influence people’s weight, we might consider their height, physical activity and diet as independent variables.
  • Two or more dependent variables also may be of interest. For example, a researcher may compare the effects of smoking on oral cancer and lung cancer.
  • Several dependent variables could be used to assess treatment effectiveness, such as length of hospital stay, number of recurrent respiratory infections and so on.
  • It is common to design studies with multiple independent and dependent variables.
  • Variables are not inherently dependent or independent. A dependent variable in one study could be an independent variable in another. e.g Effect of exercise on osteoporosis(dependent variable). On the other hand, effect of osteoporosis(independent variable) on bone fracture.
  • So we can say that whether a variable is dependent or independent is a function of role it plays in a particular study.

Confounding(extraneous) variables

  • A variable that is extraneous to the research question and that confounds the relationship between the independent and dependent variables.
  • Confounding variables need to be controlled either in research design through statistical procedures.
  • For example, if we want to study the effect of age on bearing the low birth weight babies in teenager (15 to 19 years) mothers and older (25 to 29 years) mothers, then beside age, other factors like nutritional status and prenatal care also play a role.
  • So in this study, mother’s nutritional status and prenatal care are the confounding variables. These variables are extraneous to the research question.
  • Our task is to control both these variables in order to shed light on the variable under study; i.e. age.

How To Recognize The Variables To Be Controlled?

  • First we should isolate the independent and dependent variables in which we are interested.
  • Then identify the confounding variables need to be controlled.
  • Confounding variables need to be controlled only if they simultaneously are related to both the dependent and independent variable.

MODERATOR VARIABLE

  • It is a variable that affects the strength or direction of a relationship between the independent and dependent variables.
  • For instance, we want to study the effect of nurses’ use of humor(Independent Variable i.e IV) on patient’s stress(Dependent Variable i.e DV) hospitalized for cancer.
  • But we might also be interested in whether the relationship between IV and DV is influenced by a third variable. i.e. gender?
  • Does the nurses use of humor have a different effect on stress on male versus female patients?

DUMMY VARIABLES

  • Dichotomous variables are created for use in many multivariate statistical analyses, typically using codes of 0 and 1(e.g. yes=1, no=0)
  • This variable is a cumulative count based on other variables in the datasheet.
  • These transformations are done to render data appropriate for statistical tests.

CORE VARIABLE

  • Grounded theory researchers seek to understand actions by focusing on the main concern or problem that the individual behaviour is designed to resolve.
  • The manner in which people resolve this main concern is called as core variable.
  • In a grounded theory study, it is the central phenomenon that is used to integrate all categories of the data.
  • In this, the researcher’s seek to understand actions by focusing on the main concern.