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;
42 Cards in this Set
- Front
- Back
standard error of measurement
|
the standard deviation of an individual’s observed scores from repeated administrations of a test (or parallel forms of a test) under identical conditions; characterizes the accuracy of an individual score
|
|
standard error of the estimate
|
the standard deviation of the errors of prediction; a measure of the average deviation of the prediction errors about the regression line; a measure of prediction accuracy
|
|
standard error of a statistic
|
the standard deviation of the sampling distribution of that statistic; depends on the sample size, the larger the sample size the smaller the standard error
|
|
squared correlation coefficient
|
r2 (r squared) is the Coefficient of Determination – the proportion of the total variability in Y that is accounted for by X; the proportion of variance in common
|
|
observed score model (classical test theory)
|
an individual’s observed score on a test is the sum of a true score component for the test, plus an independent measurement error component.
X = T + E X = person’s observed score on the test; T = true score; E = error (E = X – T) |
|
Assumptions made about the Observed Score Model’s Components
|
X = T + E, where (mean of the error = 0), rTE = 0 (correlation of the true score and error = 0), rEE’ = 0 (correlation of the error and the predicted error = 0)
|
|
Relationship between Validity Coefficients and Reliability Coefficients
|
the correlation between the predictor and criterion cannot exceed the square root of the reliability coefficient of the predictor
|
|
Effect of the Test Length on the Validity
|
the general formula for the correlation of any two sums may also be utilized to determine the effect of test length upon test validity
|
|
base rate
|
The true proportion of a population having some condition, attribute, or disease; proportion of individuals that would be successful (ex: at a job).
|
|
selection ratio
|
the number of individuals you select from your applicant pool to fill the [job] position.
|
|
Taylor & Russell Tables
|
the values in the table are the conditional probabilities of success given that you have been selected. Use the Base Rate, Selection Ratio, and Correlation between the Criterion and Predictor to get your Conditional Probability
|
|
Spearman-Brown prophecy formula
|
refers to the psychometric reliability to test length; used to predict the reliability of a test after changing the test length
|
|
multiple regression (correlation)
|
allows the simultaneous testing and modeling of multiple independent variables; only has one dependent variable
|
|
true positive
|
A test diagnoses you as having a disorder and you actually have it
|
|
true negative
|
A test does not diagnose you as having a disorder and you do not have it.
|
|
false positive
|
A test diagnoses you as having a disorder when you do not actually have it.
|
|
false negative
|
A test does not diagnose you as having a disorder and you have it
|
|
scales of measurement
|
Nominal Scale – categorizes items without order; ex: gender; Ordinal Scale – classifies and ranks elements/scores; ex: letter grades; Interval Scale – aka: equal unit scale; orders and establishes and equal unit in the scales so that distances between two scores are of known magnitude; ex: performance on a standardized achievement test; Ratio Scale – has equal units and also has an absolute zero in the scale; ex: height
|
|
reliability of measurement
|
consistency or repeatability of the measures used
|
|
parallel forms reliability
|
correlation of scores on 2 alternate forms of a test. Reflects consistency across items. (coefficient of equivalence)
|
|
split-half reliability
|
an internal consistency reliability measure; the correlation between the scores on these two half-tests, adjusted via Spearman-Brown; provides an estimate of the alternate-form reliability of the total test
|
|
Kuder-Richardson Formula 20 & 21 Reliabilities
|
These are alternative formulas for calculating how consistent subject responses are among the questions on an instrument. Items on the instrument must be dichotomously scored (0 for incorrect and 1 for correct). All items are compared with each other, rather than half of the items with the other half of the items. It is actually the mean of all split-half coefficients resulting from different splittings of a test. K-R 21 assumes that all of the questions are equally difficult. K-R 20 does not assume that.
|
|
Cronbach's coefficient alpha
|
An internal consistency reliability coefficient based on the number of parts into which the test is partitioned, the interrelationships of the parts, and the total test score variance
|
|
reliability coefficient
|
the proportion of observed score variance that is due to true score variance
|
|
validity of measurement
|
The greater the degree of validity of the data collection device, the more confident you will be that the results you achieve reflect true differences in the scores of your subjects and not some random or constant error, the degree of validity will reflect the degree to which we are controlling accounting for constant error. Methods of establishing validity of the measurement technique fall into one of three categories: self-evident measures, pragmatic measures, and construct validity.
|
|
validity coefficient
|
calculated as a correlation between the two items being compared; the square root of the reliability
|
|
Criterion-Related/Concurrent Validity
|
the extent to which test scores predict or correlate with performance on an external criterion
|
|
predictive validity
|
a variant because the criterion is in the future
|
|
concurrent validity
|
the extent to which test scores predict or correlate with performance on an external criterion
|
|
construct-related validity
|
indicates that the test scores are to be interpreted as indicating the test taker’s standing on the psychological construct measured by the test
|
|
content-related validity
|
items on the test measure what they are supposed to fully
|
|
discriminant validity
|
testing statistically whether two constructs differ
|
|
convergent validity
|
testing two constructs are similar
|
|
Likert scale
|
a type of psychometric response scale often used in questionnaires, and is the most widely used scale in survey research; the sum of responses on several Likert items
|
|
semantic differential
|
measures the connotative meaning of concepts on a bipolar scale; perform factor analyses on the responses
|
|
analysis of variance (ANOVA)
|
a hypothesis-testing procedure that is used to evaluate mean differences between two or more treatments or populations
|
|
F-distribution
|
all the possible F values, varies with degrees of freedom (both numerator and denominator); always positive, positively skewed, median F value = approximately 1; used most commonly with ANOVA
|
|
contingency table
|
the array into which a set of numeration data may be grouped according to two or more classification variables
|
|
analysis of covariance (ANCOVA)
|
statistical control of variability when experimental control cannot be used; a statistical method for reducing experimental error or for removing the effect of an extraneous variable; removes variability of the pretest in the posttest
|
|
bivariate normal distribution
|
describes the joint probability distribution of two variables, say X and Y, that both obey the normal distribution; zero correlation (r=0) necessarily means that X and Y are independent random variables
|
|
error suppressor
|
A predictor variable with a very small correlation with the criterion can still increase the multiple correlation coefficient if it has a large correlation with another predictor. It removes the part of the variable that does not correlate with the predictor, leaving behind that which does correlate with it. ex: Reading Comprehension Tests use speed as a factor. If we do not want speed as a factor in our analysis, we adjust for it and therefore suppress it.
|
|
isodensity contours of the bivariate normal distribution
|
the set of points for which the values of x1 and x2 give the same value for the density function ; looking at the picture of the bivariate normal distribution like a topography map, it is the concentric ellipses, the height is the same all the way around the ellipse for each line
|