Validity & Reliability

Face Validity

the content of the measure appears to reflect the construct being measured

A new measure of extroversion includes questions like “I like to interact with other people” and “when I meet a new person I am not shy.”

Predictive Validity

Scores on the measure predict behavior on a criterion measured at a future time

A questionnaire adminitered at a speed-dating event its used to predict who will call whom for a date

Concurrent Validity

Scores on a measure are related to a criterion measured at the same time

A questionnaire measures whether participants prefer Coke or Pepsi while researchers also measure how much each beverage is consumed

Convergent Validity

scores on the measure are related to other measures of the same construct

A new measure of neuroticism is related to scores on the Big 5 sub scale of neuroticism

Discriminant Validity

scores on the measure are not related to other measures that are theoretically different

A measure of mood is unrelated to a measure of state anxiety

Construct Validity

A operational definition is adequate to reflect the true theoretical meaning of a variable

Using the Beck Depression Inventory to measure depression is a good way yo measure the thoughts and behaviors associated with depression

Internal Validity

The certainty with which the results of an experiment can be attributed to the manipulation of the IV rather than some other variable

Researchers control extraneous variables and manipulate the number of sexist words in a paragraph. They conclude that the more sexist words in a paragraph causes participants to subsequently use more sexist language themselves.

External Validity

The degree to which the results of an experiment can be generalized beyond the sample or situation

A researcher finds that men who are exposed to violent media subsequently act more aggressively towards those romantic partners. Another study is run testing whether women show the same tendencies before applying the results to women as well

Reliability

Consistency of a measure

A bathroom scale yields similar weights each time you step on it over the course of a five minute period

Test-Retest Reliability

Administering the same measure to the same participants on two separate occasions and correlating those results

An IQ test yields similar scores when students take it two time over the course of two months

Split-Half Reliability

A measure is randomly split into two parts, and each ½ is correlated with the other

One half of the Resistance to Mate Guarding Scale is correlated with the other half of the same scale

Inter-Rater Reliability

Different observers agree in their ratings of a target behavior

Two teachers agree that a student is behaving aggressively on the playground

Internal Consistency Reliability

Correlations among items on the same measure

Each item on the Perceptions of Infidelity scale is correlated with each other item