Types of extraneous and potentially confounding variables

Extraneous and potentially confounding variables are a type of experimental research. There are numerous types of extraneous and potentially confounding variables:

Individual participant differences

If more than one participant is used in a study, it is reasonable to suggest that individual characteristics of the participants may impact the findings. For example, one participant may naturally enjoy a greater attention span that a fellow participant. which could significantly affect the results irrespective of the manipulation of the independent variable. The experimental design of studies is usually carefully chosen to avoid such an issue.

Non-standardised instructions and procedures

Naturally, if participants are provided with different instructions, or are tested in different ways, results may not be the same as if they were treated in a uniform fashion. For example, if one participant is told to complete the arithmetic test as quickly as they can, whilst another is told to complete the test as accurately as they can, results are likely to be impacted.

Very minor differences in language, manner and order of presentation can influence the dependent variable.

Order effects

Order effects revolve around the sequence in which participants perform a task. There are two sides to order effects:

  1. The participant’s performance on a task may decline the second time due to boredom or fatigue
  2. The participant’s performance on a task may improve the second time due to prior knowledge and practice

Placebo effects

Placebo effects suggest that participants’ behaviour may be influenced by their prior knowledge of the study or how they think they ought to behave. For example, an individual with a broken leg may think they feel less pain if they truly believe that they have taken a pain killer. To counter this, a single-blind procedure (where the participants or, in some cases, the experimenters, are ignorant as to which group is the experimental group and which is the control group) may be utilised.

Experimenter effects

Further to non-standardised instructions, the experimenter themselves may impact the results. If the experimenter presents the instructions in a pleasant way to one group, and in an abrupt manner to another, the participants’ performance on a task may be affected accordingly. This is often a concern when the experimenter is aware of which group is the control group and which group is the experimental group. To counter this, a double-blind procedure (where the participants and any experimenters are ignorant as to which group is which) may be utilised.

See also:

Ways of minimising extraneous and potentially confounding variables.