Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
90 Cards in this Set
- Front
- Back
test
|
measurement device that quantifies behavior
|
|
states
|
situation-specific characteristics
|
|
traits
|
enduring characteristics of an indiviual independent of the situation
|
|
psychological test
|
measure overt and covert characteristics of humans past, present, and future
|
|
psychological testing
|
use of psychological tests and all its applications, uses, and underlying concepts
|
|
Item
|
specific questions or problems in a test, or a stimulus that elicits an overt response
|
|
Achievement Testing
|
test of previous learning (the kind you take for a class)
|
|
Aptitude Testing
|
Test for a person's ability to learn a new skill
|
|
Intelligence Testing
|
problem solving, adapting to change. potential INDEPENDENT of previous learning
|
|
Psychological Test
|
measures characteristics pertaining to behavior, both covert and observable, in past, present, and future human behavior
|
|
Psychological Assessment
|
a procedure used to evaluate an individual to describe current and future functioning of a person. Can be tests, interviews, case studies, observations, etc.
|
|
Psychological Testing
|
use of psychological tests and all possible uses, applications, and underlying concepts of psychological tests
|
|
Results if a test is reliable
|
free from measurement error
|
|
Reliability
|
the extent to which a score is free from measurement error; ratio of true score variance to observed score variance, estimated using correlational methods
|
|
Standardization sample
|
comparison group of individuals administered a test under standard conditions. Also called a normative sample. Used to calculate percentile rank.
|
|
Factor Analysis
|
a. A set of multivariate data analysis methods for reducing large matrixes of correlations to fewer variables; the variables are linear combinations of the variables that were in the original correlation matrix (or, finding patterns in the data that strengthen correlations by making them more specific)
|
|
Hypothetical Construct
|
The theoretical ideas or processes on which operational definitions are based
|
|
Operational Definition
|
Sample of the construct's effects, defined to be measurable in the real world
|
|
Measurable Phenomenon
|
all phenomena the construct generates (like the real world demonstrations of the idea of "love")
|
|
Veil of Measurability
|
the point that separates the world of constructs from the world of measurable phenomena; creates a slight disjunction between constructs and operational definitions, but its effects can't be measured.
|
|
Structured Personality Test
|
tests that provide a statement, usually of the self-report variety (I like Rock and Roll music), and require the subject to choose between two or more alternative responses (true or false, for example); also called objective personality test
|
|
Projective Personality Test
|
b. Projective: tests in which the stimulus or the required response or both are ambiguous; the general idea behind projective tests is that a person’s interpretation of an ambiguous stimulus reflects his or her unique characteristics
|
|
Psychometry
|
the study, creation, and evaluation of Psychological Tests: includes the major properties of Reliability and Validity
|
|
Norm-referenced Test
|
Uses a standardization sample to reference the tester's performance in comparison to others
|
|
Criterion-referenced Test
|
Uses the test to predict performance outcome outside the test (i.e.: the ACT is used to detect how a student will do in college)
|
|
Types of psychological questions answered through assessment
|
a. Diagnosis and treatment planning; monitor treatment progress; help clients make more effective life choices/changes—career planning and deciding to seek therapy or to work harder in therapy; program evaluation; helping third parties make informed decisions—custody evaluations, employers, police academies, college admissions, personality styles
|
|
Purpose of assessments in a hospital setting
|
Diagnose and treat patients, monitor progress
|
|
Purpose of assessments in a school setting
|
Diagnose learning disabilities, giftedness, behavioral problems
|
|
Purpose of assessments in Forensic settings
|
Measure sanity, psychopathology, lawsuit justifications, court-ordered evaluations, competency to stand trial, jury selection
|
|
Purpose of assessments in Employment setting
|
aptitude tests, intelligence tests, competence of job (ie customer service evaluations), psychological profiling, applicant selection
|
|
Purpose of assessments in a counseling setting
|
track progress, evaluate therapy effectiveness, profiling, aptitude, compatibility
|
|
Magnitude
|
scaling property of "moreness"--a person can have more or less of a given quality (ie John is taller than Fred--good example!)
|
|
Equal Intervals
|
Scaling property that gives the same "distance" between two points in one part of the scale than in another; personality tests can't really do this
|
|
Absolute Zero
|
Scaling property that allows for ratio calculation (you can't calculate ratio for Celsius or Farenheit but you can for Kelvin)
|
|
Nominal
|
weakest scale of measurement: mutually exclusive or exhaustive categories (Nike wearers vs. Adidas wearers). No magnitude, equal intervals, or Absolute 0)
|
|
Ordinal Scale
|
Rank-ordered scaling. Exclusive and exhaustive categories. Has magnitude, but not equal interval or absolute zero data)
|
|
Interval Scale
|
Intervals can be measured in the data. Has magnitude and equal interval, but not absolute zero data. T-tests can still be performed here.
|
|
Ratio Scale
|
measured on a scale with an absolute zero; rarely exists in psychology. Has magnitude, equal interval, AND absolute zero. T-tests can be performed.
|
|
Frequency Distribution
|
Shows ALL possible scores and how frequently each possible value was obtained
|
|
Histogram
|
A bar graph using equal intervals and no spaces in between the bars
|
|
Stem-and-Leaf Plots
|
Groups data by tens digits (stems) and ones digits (leaves) to give a quick feel to the shape of the frequency distribution
|
|
Percentile Rank
|
percentage of scores that fall below a certain percentile
|
|
Mean
|
best used for interval and ratio data, but most sensitive to skews and outliers. The average of all scores
|
|
Median
|
Ordinal-level data: best for situations with outliers or markedly skewed data. Ignores all points but the 50th percentile. The middle score in a rank-ordered set
|
|
Mode
|
Nominal data; can draw attention to outliers. Gives a bad sense of central tendency. Most commonly occuring value in the set. Dr. South thinks it's useless.
|
|
Variance
|
sum of the square of the X values minus the mean, divided by N. (Or, average squared deviations around the mean) For sample, sub n-1 for N.
|
|
Standard Deviation
|
square root of the variance (average squared deviations around the mean)
|
|
Negative skew
|
greatest distribution is offset from normal to the RIGHT
|
|
Positive Skew
|
greatest distribution is offset from normal to the LEFT
|
|
Kurtosis
|
Peakedness of the curve; how steeply does it rise? probability distribution of a real-valued variable
|
|
Leptokurtic
|
Sharp peak, as opposed to the more gradual slope of a normal distribution. Said to have "fat tails" (higher probability of extreme values)
|
|
Playtikurtic
|
very gradual slope of a curve compared to a normal distribution. "Thin tails"; lower probability of extreme values
|
|
Mesokurtic
|
Normal distribution; kurtosis of 0.
|
|
Z score
|
score minus the mean, divided by the standard deviation. Puts scores into easily measurable units that show how extreme they are. Z-score of 0 IS the mean
|
|
T score
|
Like z scores, but in this case the mean=50 and the standard deviation=10 (in T score terms; the scores range from 0-100). Still tells how far from the mean a score falls, but 5sd above to 5sd below are all positive values. Values >70 or <30 are considered clinically significant. T=10z + 50, or T=(10/s)X +(50-10(Xbar/s))
|
|
Norm
|
performance info relative to standardization sample. Used as a reference for an individual's scores
|
|
Distribution of error
|
Error distributed evenly, so all scores are affected
|
|
5 characteristics of a good theory
|
explanatory power (strong ability to explain data), broad scope (external validity), systematic (cohesive statements), fruitful (contributes to existing knowledge), parsimonious (simple explanation)
|
|
Correlation Coefficient
|
degree of variation in one variable estimated by knowing the variable's variation; describes the direction and magnitude of a relationship, not to be confused with causation. (more correlation=the variables are more related)
|
|
Principle of Least Squares
|
squared deviation about the mean; line is obtained by keeping these values as small as possible. Regression line is the line of least squares.
|
|
Covariance
|
degree to which two variables vary together (the other changes when one changes. DOES NOT mean they both change in the same direction!)
|
|
Pearson Product Moment Correlation
|
ratio used to determine degree of variation in one variable that can be estimated from knowledge about variation in other variable. Range: -1-1, where 0=no correlation, -1=perfectly negative correlation, and 1=perfectly positive correlation
|
|
Residual
|
amount of observed Y not accounted for by Y' (predicted Y) in a regression line; sum of residuals always=0 because of least squares equation used.
|
|
Standard Error of Estimate
|
Standard deviation of residuals; gives an estimate of how much error is present in an experiment. If Y' values are close to Y values, it will be small
|
|
Coefficient of Alienation
|
opposite of Correlation Coefficient; how much do the variables NOT influence each other
|
|
Shrinkage
|
Shrinking in the observed strength of a correlation over time due to the decreased probability of a chance correlation
|
|
Restricted Range
|
occurs when range of variables becomes restricted; leads to reduced variance and alter strength of correlation coefficient (floor and ceiling effects)
|
|
Purpose of Discriminate Analysis
|
useful to discriminate between two already-known groups, classify groups into distinct categories
|
|
Coefficient of Determination
|
amount of variance in one variable that can be accounted for by another
|
|
Regression Formula
|
Y'-a+bX. a=intercept (Y value when x=0), b=slope (regression coefficient). Regression tells us how much we can predict.
|
|
Linear Regression
|
X and Y; simple bivariate relationship
|
|
Multiple Regression
|
multiple variables interact to produce Y'
|
|
Classical Test Score Theory
|
X=T + E
(x=observed score, t=true score, e=error) measurement error=difference between true and observed scores |
|
Domain Sampling Model
|
test items are used to sample specific domains. Domain is a large collection of items
|
|
Inter-rater reliability
|
consistency among judges viewing the same behavior.
|
|
Kappa Coefficient
|
measures agreement between multiple judges who rate objects using nominal data
|
|
Cronbach's Alpha
|
measures agreement between judges using interval or ratio data
|
|
How to improve reliability
|
increase number of items or sample size, throw out bad items, and estimate true correlation without measurement error
|
|
Selected Response format
|
multiple choice: dichotomous has two choices, polytomous has multiple
|
|
Distractors
|
incorrect responses in selected response format
|
|
Constructed Response Format
|
essay questions, fill-in-the-blank, short answer
|
|
Summative Scales
|
Likert and Category scales. Create Ordinal data
|
|
Category Format
|
9-10 item max, clearly defined anchor intervals, more options than a Likert scale
|
|
4 questions to have while making an item bank
|
1. What domain should they cover?
2. How many items are needed? 3. What are my demographics? 4. How should I word the items |
|
4 ways to score tests
|
1. cumulative scoring: items summed to form a total score
2. subscale scoring: subscales are scored seperately 3.class of category: test-takers grouped into performance categories (ie: A, B, C, D, F) 4. Ispative: subscale scores are compared against each other |
|
Optimum Difficulty Level
|
point midway between 100% and the level of success expected by chance alone
|
|
Extreme Group Method
|
people with highest scores compared against people with lowest scores on specific items to see if those items have good discrimination
|
|
Biserial Method
|
correlation between item performance and total test performance
|
|
Item Characteristic Curve
|
Item Analysis Graph where X=total test score, and Y=proportion of test takers who pass the specific item
|
|
What percentages fall where on standard deviations for a normal distribution?
|
68% within 1 sd, 95% within 2 sds, 99.7% within 3 sds
|