Understanding Control of Extraneous Variables in Research

In research, especially in experimental designs, we aim to understand the effect of an independent variable on a dependent variable. However, other variables, known as extraneous variables, can also influence the dependent variable. If these extraneous variables are not controlled, they can introduce alternative explanations for the observed results, making it difficult to conclude that the independent variable caused the change in the dependent variable. Therefore, controlling extraneous variables is crucial for establishing a clear cause-and-effect relationship and increasing the internal validity of a study.

Let's examine the given techniques and evaluate their effectiveness in controlling extraneous variables:

• (A) Change of instrument: Changing the measurement instrument might be necessary if the original instrument is unreliable or invalid, but it is not a primary technique for *controlling* the influence of extraneous variables themselves. An instrument measures variables; it doesn't typically control external factors that might affect the participants or the setting.
• (B) Randomisation: Randomisation is a powerful technique used primarily in experimental designs. It involves randomly assigning participants to different groups (e.g., treatment group and control group). The principle behind randomisation is that it distributes the effects of unknown or uncontrollable extraneous variables roughly equally across all groups. This helps to ensure that any differences observed between groups are likely due to the independent variable rather than systematic differences caused by extraneous factors.
• (C) Matching: Matching is another technique used to control extraneous variables, especially in smaller samples or quasi-experimental designs. It involves pairing participants based on one or more important extraneous variables (e.g., age, gender, IQ) that are thought to influence the dependent variable. One participant from each pair is then assigned to one group (e.g., treatment) and the other to another group (e.g., control). This ensures that the groups are equivalent on the matched variables, thus controlling their influence.
• (D) Removing variables: If possible, simply removing or eliminating an extraneous variable from the research setting or participant pool is an effective control method. For example, if noise level is an extraneous variable in a study on concentration, conducting the study in a soundproof room (removing the variable) controls its effect. Similarly, if age is an extraneous variable, restricting the sample to participants within a narrow age range effectively removes the variability of age.
• (E) Changing the research method: Changing the overall research method (e.g., from experimental to survey) might be done for various reasons related to the research question or feasibility, but it is not a specific technique *within* a method to control extraneous variables. Different methods have different strengths and weaknesses regarding control, but "changing the method" itself is not a technique like randomisation or matching.

Based on standard research methodology practices, Randomisation, Matching, and Removing variables are recognized techniques for controlling extraneous variables. Changing the instrument might address measurement issues, but not necessarily control extraneous variables directly. Changing the research method is too broad and not a specific control technique.

Therefore, the techniques used to control extraneous variables from the given options are Randomisation, Matching, and Removing variables.

The options that list these techniques are (B), (C), and (D).

Revision Table: Key Concepts in Controlling Variables

Additional Information on Control Techniques for Extraneous Variables

Controlling extraneous variables is a cornerstone of good research design, particularly in quantitative studies aiming for high internal validity. Beyond randomisation, matching, and removing variables, other techniques include:

• Holding variables constant: Ensuring that all participants experience the same level of an extraneous variable (e.g., conducting all experiments at the same time of day, in the same room).
• Statistical Control: Measuring extraneous variables and using statistical techniques (like ANCOVA or partial correlation) to remove their influence during data analysis. This is often used when physical control is difficult or impossible.
• Counterbalancing: Used in within-subjects designs where participants are exposed to multiple conditions. It involves presenting the conditions in different orders to different participants to control for order effects (like practice or fatigue).
• Blinding: Preventing participants (single blind) or both participants and researchers (double blind) from knowing which treatment condition participants are in. This controls for participant expectations and researcher bias.

The choice of control technique depends on the nature of the extraneous variable, the research design, ethical considerations, and practical feasibility.