testing conditions, teacher’s expectations, & the scoring of the test do not consistently favor any group (chance)

Observers recording what they expect participants will do rather than what the participants are actually doing

experimenters being biased when they administer the treatment

: participants changing their behavior to impress you or to help you

the participant acting in a way that makes the participant look good

a hint discovered by the participant that the participant will follow surely as if the researcher has demanded them to

a hint discovered by the participant that the participant will follow surely as if the researcher has demanded them to

stable, consistent scores that are not strongly influenced by random error

stable, consistent scores that are not strongly influenced by random error

degree to which answers to each question correlate with the overall test score; degree to which the test agrees with itself

percentage of times the raters agree; to obtain the interobserver reliability coefficient, researchers calculate a correlation coefficient between the different raters’ judgments of the same behaviors & then square that correlation

Relationship of reliability and validity, sample size, power, and effect size

- Reliability = prerequisite to validity; however reliability does not guarantee validity (i.e. a reliable measure may not be valid because it is reliably and consistently measuring the wrong thing)
- Easiest way to increase power of study is to add participants

degree to which the measure is measuring the construct that it claims to measure

1. Good content validity
                2. Good internal consistency
                3. Good convergent validity
                4. Good discriminant validity

Face: weakest kind, "does the test LOOK like it's right?"

Content: identify construct that is being measured, a good one samples the whole area