Appraisal Flashcards
Assessment
Process of accessing or estimating attributes
Subjective Test
No right or wrong. Scoring is based on scorer’s opinion
Projective Test
Rorschach test or picture/story telling test, gives the interpreter a better
understanding of the test taker’s attributes or thought processes, free association (when I say
____ what do you think of), completion (finish a sentence), and construction (picture drawing)
Free Choice Test
No options given (short answer)
Forced Choice Test
Answer options given (multiple choice)
Difficulty Index
Percentage of people that answered the question correctly
Recognition Testing
Multiple choice, options given, test taker doesn’t need to fully generate the answer
Ipsative
Compares traits in an individual, compares the test takers scores to their own standard of behavior (how does the test taker’s level of achievement compare to the test taker’s level of motivation)
Normative
compares traits to other’s traits (how does the test taker compare to the population)
Speed Test
No one should be able to finish, difficulty is in the time limitation not the subject matter (multiplication tables)
Power Test
Evaluates mastery, has no time limit (final exam)
Achievement Test
Looks at maximum performance
Personality Test/Interest Inventory
Measures typical performance
Spiral Test
Gets more difficult as the test continues
Cyclical Test
Has sections built in the test that get progressively harder and then go back to the baseline
Test Battery
Several tests given to access a trait or person as a whole
Parallel or Equivalent Forms
Tests ask different questions but have the same mean, standard error, and test the same thing
Validity
Number one factor in test construction, does the test measure what it is supposed to measure? A VALID TEST IS ALWAYS RELIABLE
Content Validity
Does the test look at the theoretical construct specifically from different angles?
Construct Validity
Does the test measure a theoretical construct (achievement, interest, etc.) traits that you cannot observe?
Concurrent Validity
Does this test compare to other tests that measure the same thing?
Predictive/Empirical Validity
Does the test predict future behavior?
Consequential Validity
Does using the test have social implications?
Face Validity
Does the test look like it measures the intended attribute or trait
Convergent Validity
Used to establish construct/criterion validity by comparing test scores to an outside source (like a different test that measures the same thing)
Discriminant Validity
The test should not measure unrelated variables (if you are using a test for stress it should not also measure IQ)
Reliability
Second most important factor in test construction, do the measurements remain consistent over repeated tests? If you give someone the same test again and hold all things constant, will they get the same score? A reliability coefficient of 1.00 is perfect. A TEST CAN BE RELIABLE AND NOT VALID
Equivalent/Alternate Forms Reliability
One group of test takers take multiple forms of the same test
Test-Retest Reliability
One group of test takers take the same test at least twice
Split-Half Method
One test is split in half and each half is given to test takers (test takers do not take both halves of the test)
Interrater/Interobserver Reliability
Used with non-forced choice tests, multiple raters give the same score to the same test
Internal Consistency Reliability
Does each item measure the same thing as every other item? Can be established using the Kuder-Richardson coefficient of equivalence if you don’t want to split the test in half
Coefficient of Determination
This is correlation squared, If the correlation between the first administration and the second administration is .5 then the coefficient of determination is .25 which means the true variance (or shared variance between both) is 25%
IQ
Intelligence quotient, formula is mental age/chronological age X 100, found to be normally distributed by Francis Galton, first IQ test created by Binet and Simon,
Stanford-Binet IQ test
A standardized measure, used from 2-adulthood, uses the standard age score instead of the IQ formula, mean of 100 and standard deviation of 16
Wechsler Adult IQ Test (WAIS-III)
Uses 7 verbal scales and 7 performance scales, for adults, mean of 100 and standard deviation of 15
WPPSI-III
Used for ages 2 years 6 months to 7 years 3 months
WISC-IV
Used for kids 6-16 years 11 months
J.P. Guilford
Discovered 120 factors that make up intellegence
BITCH Test
Black Intelligence Test of Cultural Homogeneity, created by Robert Williams inresponse to an article written by Arthur Jenson that accused IQ tests of scientific racism
MMPI-2
Standardized Personality Test
Psychometric
Any type of mental testing
16 PF
Developed by Raymond Cattell, measures personality
MBTI
Developed by Carl Jung
Mental Measurements Yearbook
Explaination of psychometric tests first compiled by Buros
Aptitude
A Person’s potential, natural ability to do something
Acheivement
What has been learned or mastered
Standard Error of Measurement
How accurate a test score is (SEM of 5 and a score of 126 means the person’s true score range is 121 – 131)
