Chapter 11- More on Experiments: Confounding and Obscuring Variables

Threats to Internal Validity/Types of Counfounds

• Changes in participants

• Changes in measurements

Changes in Participants

• History

• Maturation

• Testing Effects

History

Refers to external events that occur during the course of an experiment which may affect participants' responses, potentially leading to confounding results.

Maturation

Refers to the psychological or physical changes that participants undergo over time, independent of the experimental treatment, which can influence their responses and confound results.

Testing Effects

Refers to the effects resulting from participants being tested multiple times, which can influence their subsequent performance due to practice or fatigue.

Changes in Measurement

• Instrumentation

• Regression to the mean

• Attrition

Instrumentation

Refers to changes in the measurement tools or procedures used in a study that may affect the consistency and validity of the results.

Regression to the Mean

A statistical phenomenon where extreme scores tend to move closer to the average upon repeated measurement, potentially impacting the interpretation of experimental results.

Attrition

Refers to the loss of participants from a study over time, which can lead to biased results if the attrition is related to the treatment or outcome.

Issues for all Designs

• Selection

• Observer bias (even with a comparison group)

• Demand Characteristics (even with a comparison group, do a double blind/masked study to avoid)

• Placebo Effect

Selection Effects

Systematic differences between groups that may influence outcomes in an experiment, often arising from non-random assignment.

Observer Bias

A type of bias that occurs when a researcher's expectations or beliefs influence their observations or interpretations of data, which can skew results. This is problematic even when a comparison group is present.

Demand Characteristics

• Participants' behavior influenced by their awareness of the study's purpose or expectations, which can lead to biased outcomes.

• Do a double blind/masked study to avoid

Placebo Effect

A phenomenon where participants experience changes in their condition simply because they believe they are receiving treatment, even if it is inactive or a placebo.