What is a psychological test?
An objective and standardized measure of a sample of behavior (Anastasi & Urbina) A systematic procedure for observing behavior and describing it with the aid of numerical scales or fixed categories (Cronbach)
The technical term for the science behind psychological testing
Define Psychological assessment
a process that involves the integration of information from multiple sources, such as tests of normal and abnormal personality, tests of ability or intelligence, tests of interests or attitudes, as well as information from personal interviews.
Define The first responsibility of test users as well as additional responsibilities
to “Define the purpose for testing, the content and skills to be tested, and the intended test takers.
When is The use of formal testing is most productive
when it answers a client’s question and is seamlessly integrated into the therapeutic process.
After identifying the potential tests, then you must evaluate the tests to see which ones will be most effective for the identified purpose based on:
- Appropriate selection of norming group or criterion
- Interpretation and scoring materials
- User qualifications
What are the
Four Types of Measurement Scales and their definitions/examples?
- Nominal-Classification by names (e.g., colors, political parties, ethnicities, etc.)
- Ordinal-Rank ordering by magnitude (e.g., pain, depression, sickness scale ratings)
- Interval-Units of measurement are in equal intervals (e.g., measures of IQ, time, weight, etc.)
- Ratio- Similar to interval scale, but “zero” is meaningful (e.g., miles/hour, temperature)
Most statistical techniques used to evaluate differences in test scores assume what?
that the scales being used are interval scales
Define Norm-referenced tests
tests where the individual’s performance is compared to scores of other individuals who have taken the same test. A norming group typically is quite large, or it can be relatively small, e.g., when an instructor grades “on the curve.”
Define criterion-referenced tests
the individual’s score is compared with an established standard or criterion. Criterion-referenced tests are also sometimes called domain- or objective-referenced. The individual is judged as to whether his/her performance reaches a certain standard of performance within a certain domain of knowledge (e.g., licensing exams, BAR exam, CPA exam, etc.)
What are the difficulties with criterion-referenced testing
- Test has to appropriately assess the domain of knowledge. However, experts often have difficulty agreeing upon what information comprises their core domain of knowledge
- Another difficulty is determining the criterion for mastery level. For example, if mastery is set at 90%, has a person not mastered the material if they score 89%?
Define the Measures of Central Tendency
- The mode is the most frequent score in a distribution. To calculate the mode, you merely count the number of people who received each score, and the score with the highest number of people is the mode
- Median-point in the distribution half the scores are higher and half are lower
- Mean-arithmetic average
Define the Measures of Variability
- Range-The range provides a measure of the spread of scores and indicates the variability between the highest and the lowest scores.
- Variance and Standard Deviation-Indicate how scores vary around the mean.
- The variance is the mean squared deviation
- The standard deviation is the square root of the variance
Define Normal Distribution
- The normal distribution or normal curve is the distribution that characterizes a number of our human traits (e.g., weight, height, intelligence) and has mathematical properties that make it very useful in interpreting norm-referenced tests.
- The normal curve is bell-shaped with a single peak and the mode, median and mean are the same score.
What are the values of Standard deviation in the normal distribution or normal curve
- 68% of the cases fall between one S.D. below and one S.D. above the mean
- 95% of cases fall between + 2 S.D.
- 99.5% of cases fall between + 3 S.D.
Define Percentile scores or percentile ranks
indicate the percentage of people in the norming group who had a score at or below a given raw score.
Why are Standard Scores useful?
They can be used with all types of tests (e.g., intelligence tests, personality tests, career assessments). Standard scores provide a quick, easy way to know a client’s relative position on a test because standard scores describe how many standard deviations a client’s score is from the mean
There are a number of different types of standard scores including:
- z Scores-to convert a raw score to a z-score, subtract the mean of the test from the client’s raw score and divide by the standard deviation of the test. Mean=0 and SD=1
- T Scores-T scores have a fixed mean of 50 and a standard deviation of 10. A z score can be converted to a T score by multiplying the z score by 10 and adding or subtracting the number from 50.
- Stanines-range from 1-9, M=5 and SD =2. They represent a range of percentile scores
What does Reliability provide
an estimate of the amount of true variance to observed variance.
What is the coefficient of determination used for?
to measure the degree to the extent of shared variance between two variables.
What is correlation used for?
The statistical technique of correlation is used to examine the relationship between scores.
Types of Reliability
- Test-Retest Reliability-Correlating the performance of the first test administration and the second.
- Challenges: The trait that is being measured with test-re-test must be stable over time.
- Challenges: There should be no differential in practice effect.
- Challenges: There should be no differential in learning
- Alternate or Parallel Forms-requires that there be two forms of a test. Persons are given one form and then administered the alternative form and a correlation between the two forms is calculated and the estimated reliability is determined. Care should be taken to assure that the two forms are truly parallel.
- Challenge: The effort to develop two parallel forms of a test is monumental – not a lot of parallel forms tests.
- Internal Consistency Measures of Reliability-these methods of estimating reliability use one administration and a single form of the test.
- Split-half reliability-the test is administered once and then split in half to determine reliability
- Kuder-Richardson 20 and 21
- Coefficient alpha (Cronbach’s Alpha)-estimation of reliability when the scoring of the test is not dichotomous (scales that have range of responses)
Define Standard Error of Measurement
Standard error of measurement provides an estimation of the range of scores that would be obtained if someone took the test over and over again. The SEM can be used to provide a client with an expected range of where his/her true score would fall.
- Does the test measure what it purports to measure.
- The validation of a test is the gradual accumulation of evidence that indicates whether the test does indeed measures what it is intended to measure.
- It is not the test that is validated, rather, it is the uses of the test that are validated.
- Reliability is a prerequisite to validity. If a test has too much unsystematic error, then it cannot measure anything consistently.
Types of validity
Content validity-do the test items measure the intended behavioral domain
Criterion-related validity- refers to the extent to which the test is a good predictor of a certain criterion. When we want a test to predict future behavior, the validation evidence must focus on how well the test predicts future performance (e.g., the SAT)
- Concurrent validity-we use to make immediate predictions
- Predictive validity-we are making predictions in the future
Construct validity-does the test measure a theoretical or hypothetical construct or trait.
Mental Status Examination generally provides an overall impression of a person in a number of areas including:
- General Appearance
- Feeling ( Affect & Mood)
- Thinking (Intellectual functioning, memory, attention & concentration, insight and judgment)
- Thought content
What has The most extensively developed component of the mental status examination process has focused on
What is the most extensively studied and most frequently used Mental Status Exam?
The mini mental status exam
Who is is credited with launching the first testing movement
Who was the first to use the term, mental test
Who revised the Binet-Simon Test and what did it become known as
- L.M. Terman
- Stanford-Binet Intelligence Test
What is the difference between the Army Alpha and Army Beta tests?
- The Army Alpha- routine administration
- Army Beta was non-language test designed for persons who were illiterate and non-English speaking recruits