Test Types & Formats
Compares a test taker's score to the performance of a norm group rather than to a fixed standard.
What is a "norm-referenced test"?
A career counselor administers the same interest inventory to a group of clients, waits six weeks, and administers it again to see if scores hold steady. This is the type of reliability being estimated.
What is test-retest reliability? What is Internal consistency reliability?
This measure of central tendency is the arithmetic average of a set of scores.
Whta is the mean?
A counselor gives an assessment to a group of clients. Most score very high, with just a few low outliers pulling the tail toward the left side of the distribution. This distribution is best described as.
What is a normal distribution (normal curve)?
This widely used career interest inventory organizes results around Holland's six RIASEC themes (Realistic, Investigative, Artistic, Social, Enterprising, Conventional) to help clients match interests to occupations.
What is the Strong Interest Inventory?
Measures what a person has already learned or mastered.
What is an "achievement test"?
Two different clinicians independently score the same set of client responses on a behavioral checklist and compare how closely their ratings agree. This is the type of reliability being estimated.
What is inter-rater reliability? What is Standard error of measurement (SEM)?
This measure of central tendency is the middle score in a distribution when all scores are arranged in order.
What is the median?
A researcher wants to compare the same group of clients' anxiety scores before and after an 8-week intervention. This is the most appropriate statistical test to use.
What is a paired-samples (dependent) t-test?
This individually administered intelligence test for adults yields a Full Scale IQ along with index scores for verbal comprehension, perceptual reasoning, working memory, and processing speed.
What is the WAIS (Wechsler Adult Intelligence Scale)?
Administered to multiple test takers simultaneously; more efficient but less observational data.
What is a group assessment?
A 40-item depression scale has a Cronbach's alpha of .91, indicating the items are all measuring a similar underlying construct. This statistic reflects this type of reliability.
What is internal consistency reliability?
This measure of central tendency is the most frequently occurring score in a distribution.
What is the mode?
A distribution of scores has a mean of 82, a median of 88, and a mode of 91. This pattern indicates the distribution is skewed in this direction.
What is negatively (left) skewed?
This projective personality test presents a series of ambiguous inkblots and asks the client to describe what they see, with responses scored for content, location, and determinants to reveal underlying personality dynamics.
What is the Rorschach Inkblot Test?
Presents ambiguous stimuli (e.g., inkblots, incomplete sentences) to reveal underlying personality dynamics.
What is a projective assessment?
A new anxiety measure correlates strongly with an established, well-validated anxiety measure administered at the same time. This is the specific type of validity being demonstrated.
What is concurrent validity (a form of criterion-related validity)?
This statistic measures how spread out or dispersed scores are around the mean.
What is the standard deviation?
A researcher compares outcome scores across three different treatment groups by running three separate t-tests instead of one omnibus test. This is the main statistical problem that approach creates.
What is an inflated Type I error rate (increased chance of a false positive)?
This widely used, empirically based personality inventory uses true/false items and clinical scales (along with validity scales to detect response distortion) to assess psychopathology.
What is the MMPI (Minnesota Multiphasic Personality Inventory)?
Measures whether a test taker has met a predetermined standard or mastered specific content.
What is a Criterion-referenced test
A college admissions test is given to incoming students, and their scores are later compared to their first-year GPA to see how well the test forecasts academic performance. This is the specific type of validity being demonstrated.
What is predictive validity (a form of criterion-related validity)?
This symmetrical, bell-shaped distribution has the mean, median, and mode all falling at the same point in the center.
What is a normal distribution (normal curve)?
A test has excellent internal consistency (alpha = .95) but has never been checked against any outside criterion or shown to predict any real-world outcome. This is the key limitation illustrated here.
What is the fact that reliability does not guarantee validity — a test can be highly reliable without being valid?
A counselor wants a brief, widely used self-report measure to quantify the severity of a client's depressive symptoms over the past two weeks, primarily as a symptom-tracking tool rather than a diagnostic instrument. This is the most likely instrument used.
What is the Beck Depression Inventory (BDI)?