test 2 notes

Question 1

Reliability

Answer 1

Property or attribute of consistency in measurement for a test

Question 2

Reliability Coefficient

Answer 2

Statistic that ranges from 0 to 1.0

Question 3

Classical Test Theory

Answer 3

Variance of any score is due to “true” measurement plus error

Question 4

True Score

Answer 4

Reflects a person’s true ability/attribute/trait that test is trying to measure

Question 5

Measurement Error

Answer 5

Influence of any other variable that could change a “true” score, e.g., low test reliability

Question 6

Error

Answer 6

Actual test score minus the true score.

Question 7

Measurement Error Sources

Answer 7

Unsystematic errors and systematic errors

Question 8

Unsystematic Errors

Answer 8

Not as serious, from random scores, item selection, test administration, and test scoring

Question 9

Systematic Errors

Answer 9

More serious, poor content domain sampling, question on a social intro version test that actually measures general anxiety

Question 10

Coefficient Alpha

Answer 10

Ranges from 0 to 1

Question 11

Cronbach’s α

Answer 11

Mean of all possible split-half reliability coefficients for a given test. Shows correlation of scores with each other

Question 12

Kuder-Richardson-20

Answer 12

Similar to α, but used when test responses are dichotomous (True/False)

Question 13

Decisions About People

Answer 13

.90 to .95 Reliability Coefficient

Question 14

Research Tests

Answer 14

.80 reliability coefficient

Question 15

Tests that seem promising but need more development

Answer 15

.70 reliability coefficient

Question 16

Relationship of Reliability to SEM

Answer 16

Lower SEM=Higher Reliability

Question 17

Validity

Answer 17

A unitary concept that reflects the extent to which a test measures what it aims to measure

Question 18

Utility

Answer 18

Inferences made from the test are appropriate, meaningful and useful

Question 19

Construct

Answer 19

A complex psychological concept that cannot be directly measured. Happiness, depression, love.

Question 20

Face Validity

Answer 20

Questions are clear and understood by
examinee to reflect what’s being tested
or measured. Example: BDI-II. However, face valid tests are susceptible
to response bias.

Question 21

Content Validity

Answer 21

Degree to which the questions, tasks, or
items on a test are representative of the
universe of behavior the test was designed
to sample. Example: Beck Depression Inventory

Question 22

Criterion Related Validity

Answer 22

when test is effective in estimating an examinee’s performance on some outcome measure

Question 23

Criterion

Answer 23

a concrete real-world outcome (e.g., college acceptance, employment status, work produced, grade)

Question 24

What are the two types of Criterion Validity?

Answer 24

Concurrent and Predictive

Question 25

Concurrent Validity

Answer 25

criterion is assessed at the SAME TIME as the measure

Question 26

Predictive Validity

Answer 26

Criterion is assessed some time AFTER the measure

Question 27

Convergent Validity

Answer 27

Test does indeed correlate with other similar tests or variables as it should

Question 28

Discriminant (Divergent) Validity

Answer 28

Test does not correlate with other tests or variables as it should not

Question 29

Beck Depression Inventory (BDI-II)

Answer 29

Cognitive-Affective Factor Guilt, self-criticism, pessimism Attention and concentration problems Loss of interest in enjoyable activities Somatic Factors Tired everyday, loss of energy/motivation Eating and sleeping problems Difficulty completing everyday tasks

Question 30

Rapport

Answer 30

Testing environment of mutual respect and understanding is crucial to good test scores. Low rapport can cause anxiety, hostility

Question 31

Examiner Factos

Answer 31

Sex, Race, Experience. Less important than rapport

Question 32

Rosenthal Effect (Pygmalion Effect)

Answer 32

Lofty expectations of Examiner leads to improved examinee scores Low expectations result in lower scores Expectancy very subtle/unintentional Tester/examiner may subtly convey expectancy to examinee

Question 33

Examinee Motivation (Response Bias)

Answer 33

Test scores are unreliable if examinee willingly and purposely alters his or her responses during testing

Question 34

Stereotype Threat

Answer 34

Threat of conforming to a negative stereotype about one's group

Question 35

Yerkes-Dodson Law (1908)

Answer 35

Principle that moderate levels of arousal leads to optimal level of arousal