Research

Front 1

Classical test theory

Back 1

Observed variability in test scores as reflecting two components: true differences betn examinees on the attribute measured by the test and the effects of measurement (error).

Front 2

Coefficient alpha KR 20

Back 2

Method for assessing internal consistency reliability that provides an index of aveage inter item consistency. Kuder Richardson Formula 20 can be used as a substitute for coefficient alpha when test items are scored dichotomously.

Front 3

convergent and discriminant validity - construct validity

Back 3

Extent to which a test measures the hypothetical trait it is intended to measure. Methods for establishing construct validity invlude correlating test scores with scores on measures that do and do not measure the same trait.

Front 4

Content validity

Back 4

Extent to which a test adequately samples the domain of information, knowledge of or skill that it purports to measure. determined primarily by "Expert judgment". Most important for achievement and job samples.

Front 5

Criterion Contamination

Back 5

refers to bias introduced in to a person's criterion score as a result of the knowledge of the scorer about his or her performance on the predictor.

Front 6

Criterion referenced interpreation

Back 6

Interpretation of a test score in terms of a prespecified standard, that is, in terms of percent of content correct or of predicted performance on an external criterion.

Front 7

Criterion related validity - concurrent and predictive validity

Back 7

involves determining the relationship between the predictor and the criterion. It can be either concurrent (predictor and criterion scores obtained at about the same time) or predictive (predictor scores obtained before criterion scores.)

Front 8

Cross validation

Back 8

Process of re assessing a test's criterion related validity on a new sample to check the generalizability of the original validity coefficient.

Front 9

Shrinkage

Back 9

the validity coefficient usually shrinks on cross validation because the chance factors in operating in the original sample are not all present in the cross validation sample.

Front 10

Factor Analysis

Back 10

Multivariate statistical technique used to determine how many factors are needed to account for the intercorrelations among a set of tests, subtests. It can be used to assess a test's construct validity by indicating the extent to which the test correlates with factos that it would and would not be expected to correlate with. True score variability consists of communality and specificity. Factors identified in a factor analysis can be either orthogonal or oblique.

Front 11

Factor Loadings

Back 11

Correlation between a test and a factor.

Front 12

Communality

Back 12

total amount of variability in scores on the test that is accounted for by teh factor analysis.

Front 13

Types of Dyslexia (Reading Disorder)

Back 13

• Surface dyslexia – reads without comprehension due to an inability to read words that are irregularly spelled (e.g. light, sword)
• Deep dyslexia – several reading errors including semantic paralexia = substitution of words with similar meanings (e.g. cold for hot)
• Pure Alexia – “word blindness” an inability to read words, even if they were just written by the person.
• Literal Alexia – can read whole words, but not individual letters
Phonological Alexia - be unable to read (pronounce) pseudo-words

Front 14

Reactivity (reactive arrangements)

Back 14

Research participants can respond to an IV in a particular way simply because they know their behavior is being observed, Results can then, not be generalized.

Front 15

Demand characteristics

Back 15

Cues in teh experimental setting that inform subjects of the purpose of the study or sugges what behaviors are expected of them.

Front 16

Multiple treatment interference

Back 16

Whena study involves exposing each subject to two or more levels of an IV, effects of one level of the IV can be affected by previous exposure to another level (Order effects)

This can be controlled using a counterbalanced design in which different subjects or groups rceive the levels of the IV in a different order. e.g. Teh Latin Square design - involves administering each level fo the IV so that it appears the same nuber of times in each position.

Front 17

Experimenter Expectancy

Back 17

E unintentionally provides cue’s as to what is being looked for. Computational errors are frequently made in the direction supporting the hypothesis.

Front 18

Randomized block design

Back 18

at least one of the IV's is a blocking variable and subjects within each block are randomly assigned to levels of another IV.

Front 19

Split plot (mixed) design

Back 19

at least one between groups variable and one within-groups variable.

Front 20

Latin Square

Back 20

a type of within-subjects design in which the order of the different levels of the independent variables are counterbalanced (applied to different subjects or subgroups)

Front 21

Solomon – Four group design

Back 21

used to study the effects of pretesting

Front 22

Continuous Variable

Back 22

can take on an infinite number of values on the measurement scales.

Front 23

Discrete Variable

Back 23

Can only assume a finite number of values

Front 24

Nominal Scale

Back 24

divides variables into unordered categories (i.e.; sex).

The only mathematical operation is frequency.

(use Mode)

Front 25

Ordinal Scale

Back 25

divides observations into categories and provides information on the order of the categories. They do not provide information on how far apart the scores are in the categories. Associated with ranks.

(use Mode or Median)

Front 26

Interval Scale

Back 26

has the property of order and equal interval.

Addition and subtraction is possible for mean and SD.

(use Mode, Median or mean)

Front 27

Ratio Scale

Back 27

most mathematically complex.

It has, order, equal intervals, and an absolute zero point. One can add, subtract, multiply and divide ratio scores.

Temperature, calories consumed, number of correct items on a test.

(use Mode, Median or mean)