Controlling Extraneous Variables
Filysia SetiawanSusan
The Validity of a study will be approached from four
perspectives
Environmental IssuesGrouping IssuesPeople IssuesMeasurement Issues
Environmental Issues
Naturally occurring variables Environmental situation naturally will affect the variables although it is indirectly.
Environmental Issues
Artificiality is surreptitiously alter the intentions of a study. Artificiality Experimental miceIn which the conditions are artificial, as they would in the real world
Grouping Issues Self – selectiongenerally refers to the practice of letting the subjects decide which group to join.
Mortality refers to students who self-select themselves out of group by dropping out of the study for any number of reasons.
Grouping Issues
Maturation Different individuals can have
different experiences that cause them to mature or change over time.
People Issues
Hawthorne effectit happens if the result of investigation are more closely related to this pleasure than to anything that actually occur in the research.
Halo effect is due to the tendency among human beings to respond positively to a person they like.
People Issues
Subject expectancyoccurs when the subjects think they have figured out what study is about.
People Issues
Researcher expectancyas Rosenthal showed can be a
problem in behavioral research.Is often controlled for with double
blind technique in medical research.
Measurements IssuesPractice effect essentially involves the potential
influence of the measures on each otherReactivity effectmay occur when this measures
themselves are causing a change in the subjects.
Measurement Issues
Instability of measures and resultsrefers to the degree to which the result on the measure are consistent.
ConclusionPerspectives Potential Problems Steps toward control
Environmental Natural Variables
Prearrangement of conditions
Artificiality Approximation of natural conditions
Grouping Self-selection Random, matched-pair, or stratified assignments
Mortality Short duration, track down missing subject
Maturation Short duration or built-in moderator or control variables
ConclusionPerspectives Potential Problems Steps toward control
People Hawthorne effect Double-blind techniqueHalo effect Built-in general attitude
as moderator or control variables
Subject expectancy Minimize obviousness of aims. Distraction from aims provided
Researcher expectancy Double –blind techniqueMeasurement Practice effect Counterbalancing
Reactivity Careful study of measuresInstability of measures and results
Statistical estimates of stability and probability