Operationalising Variables in Research

Operationalising Variables

Operationalisation of Variables

• Definition: Operationalisation of variables is an initial and important part of the research process that involves establishing the elements or characteristics that constitute measurement of the variable(s) in the research.

• Requirements for Researchers:

  • State explicitly the variable(s) involved in the research.

  • State explicitly how the variable(s) will be measured.

Example Operationalisation of School Bullying

• School bullying involves various elements and characteristics. Measuring these components helps in assessing the extent of bullying. The elements include:

  • Being called ‘mean’ names; made fun of, or teased in a hurtful way.

  • Being excluded from activities or completely ignored.

  • Being hit, kicked, punched, or shoved around.

  • Others spreading rumours about the individual.

  • Being forced to do things that the individual did not want to do.

• The frequency of these above behaviours serves as indicators of the extent of bullying.

Some Variable Types

Independent (IV) and Dependent (DV) Variables

• Independent Variable (IV): The factor manipulated by the experimenter.

• Dependent Variable (DV): The outcome measured under each condition of the independent variable.

• Note: Conditions of the independent variable refer to the levels of the manipulation.

Example One: Statistics Course

• Hypothesis: Success on a statistics course is a direct function of the amount of time spent studying by the student.

• Experimental Design: The researcher randomly assigns students to one of four conditions:

  • 0 hours study

  • 5 hours study

  • 10 hours study

  • 20 hours study (per day!)

• Dependent Variable: Errors made in the statistics examination.

• Independent Variable: Hours of study (0, 5, 10, 20 hours).

• Conditions: 0, 5, 10, 20 hours of study.

Example Two: Behavior Modification vs. Psychoanalysis

• Hypothesis: The behavior modification approach is a better therapeutic treatment for depression than the psychoanalytic approach.

• Experimental Design: 50 patients with depression are randomly assigned to one of two practitioners:

  • One using behavior modification.

  • One using the psychoanalytic approach.

• Measurement Tool: Beck Depression Inventory (BDI), with scores indicating levels of depression:

  • 0–9: minimal depression

  • 10–18: mild depression

  • 19–29: moderate depression

  • 30–63: severe depression

• Dependent Variable: Difference between pre- and post-treatment BDI scores.

• Independent Variable: Type of Therapy (Behavior Modification vs. Psychoanalytic).

Example Three: New Drug Testing

• Hypothesis: A new drug will prevent colds and flus.

• Experimental Design: Volunteers of the same age are randomly assigned to one of three conditions:

  • New drug

  • Placebo

  • Nothing at all

• Dependent Variable: Number of colds and flus (and associated symptoms) recorded over the period of one year.

• Independent Variable: Type of treatment administered (New drug, placebo, or nothing).

Predictor and Criterion Variables

• Definition: These terms are equivalent to independent (predictor) and dependent (criterion) variables, primarily used in observational research contexts like correlational studies. In experimental contexts, causation can be inferred more readily by manipulating the IV, while in correlational studies, the terms help to avoid assumptions of causation.

Extraneous Variables

• Extraneous Variable: Refers to any variable that affects the dependent variable apart from the independent variable. The experimenter aims to control or eliminate these variables, usually by holding them constant across all conditions.

Confounding Variables

• Confounding Variable: A type of extraneous variable that is not controlled, which obscures the effect of the independent variable on the dependent variable. Confounding variables can mislead results and interpretations.

Populations and Samples

• Population: Refers to all elements contained within a specified group. Examples include:

  • All people on earth.

  • All Irish Mothers.

  • All students at Magee College.

• Sample: A subset of the elements from the population, such as:

  • A randomly drawn sample from all people on earth.

  • A randomly drawn sample from all Irish Mothers.

  • A randomly drawn sample from all students at Magee College.

Parameters and Statistics

• Parameter: A numerical description of some aspect of the population. Generally, population parameters are unknown due to the difficulty in gathering information on the entire population.

• Statistic: A numerical estimate of some aspect of the population based on information gathered from a sample.

Summary of Parameters and Statistics