ASSUMPTIONS ABOUT PSYCHOLOGICAL STRATIFIED-RANDOM SAMPLING- population

TESTING has an equal opportunity of being included in a

sample.
ERROR- refers to a long-standing assumption
that factors other than what a test attempts to PURPOSIVE SAMPLE- arbitrarily selecting a
measure will influence performance on the test sample that is believed to be representative of
the population who meet certain criteria set.
ERROR VARIANCE- the component of a test
score attributable to sources other than the INCIDENTAL/CONVENIENCE SAMPLE- a sample
trait or ability measured. that is convenient or available for use. May not
be a representative of the problem.
Testing and assessment can be conducted in a
fair DEVELOPING NORMS
- Problems arise if the test is used with - Administering the test with standard set
people for whom it was intended. of instructions
- Some problems are more political than - Recommend a setting for a test
psychometric in nature administration
- Collect and analyze data
Testing and assessment benefit in society - Summarize data using descriptive
- There is a great need for tests, statistics including measures of central
especially good tests, considering the tendency and variability.
many areas of our lives that they TYPES OF NORMS
benefit.
PERCENTILES- percentage of people whose
RELIABILITY- consistency of the measuring tool: score on a test or measure falls below a
the precision with which the test measures and particular raw score.
the extent to which error is present in
measurements. - Popular method for organizing test-
related data because they are easily
VALIDITY- the test measures what it purports to calculated.
measures.
AGE NORMS- average performance of different
OTHER CONSIDERATIONS- administering, samples pf test-takers who were at various ages
scoring, interpretation should be when the test was administered.
straightforward for trained examiners.
GRADE NORMS- average test performance of
- A good test is a useful test that will testtakers in a given school grade.
ultimately benefit individual testtakers
or society at large. NATIONAL NORM- derived from a normative
sample that was nationally representative of
NORMS-REFERENCED TESTING AND the population at the time the norming was
ASSESSMENT- a method of evaluation and a conducted.
way of deriving meaning from a test scores by
evaluating and individual testtaker’s score and NATIONAL ANCHOR NORMS- equivalency table
comparing it to score of a group testtakers. for scores on two different tests. Allows for a
basis of comparison.
INDIVIDUAL TEST SCORES- is understood
relative to other scores on the same test. SUBGROUPS NORMS- normative sample can be
segmented by any of the criteria initially used in
NORMS- test performance data of a particular selecting subjects for the sample.
group of testtakers that are designed for use as
as a reference when evaluating or interpreting LOCAL NORMS- provide normative information
individual test scores. with respect to the local population’s
performance on some test.
NORMATIVE SAMPLE- reference group to
which test-takers are compared. FIXED REFERENCE GROUP SCORING SYSTEMS-
distribution of scores obtained on the test from
STANDARDIZATION- process of administering a one group of testtakers is used as the basis for
test to a representative sample of testtakers for the calculation of test scores for future
the purpose of establishing norms.
administrations of the test.
SAMPLING- test developers select a population, NORM-REFERENCED versus CRITERION-
for which the test is intended that has a least REFERENCED INTERPRETATION- norm
one common, observable characteristic. referenced tests involve comparing individuals
STRATIFIED SAMPLING- includes different to the normative group.
subgroups, or strata from the population. CULTURE AND INFERENCE- responsible test
users should research the test available norms
to check how appropriate they are for use with p= the proportion of the people getting each
the targeted testtaker population. item correct
PROPERTIES OF A STANDARDIZED TEST q= the proportion of people getting each item
incorrect.
RELIABILITY- refers to the consistency of test
scores obtained by the same persons when they 5. COEFFICIENT ALPHA- developed by
are reexamined with the same test on different Cronbach to estimate the internal
occasions, or with different sets of equivalent consistency of tests in which the items
items, or under varying examining conditions. are not scored as 0 or 1.
- Applicable for many personality and
TYPES OF RELIABILITY attitude scales.
1. TEST RE-TEST RELIABILITY- repeating - The SPSS software provides a
the identical test on a second occasion. convenient way of determining the
- The reliability coefficient is simply the coefficient alpha.
correlation between scores obtained by
FACTORS OF AFFECTING TEST RELIABILITY
the same persons on the two
administrations of the test. a. Test format
b. Test difficulty
Two possible negative effects when doing test- c. Test objectivity
retest reliability d. Test administration
CARRYOVER EFFECT- occurs when the first e. Test scoring
testing session influences scores from the f. Test economy
second session. g. Test adequacy

PRACTICE EFFECTS- some skills improve with VALIDITY- test is the extent to which it
practice. measures what it claims to measures. It defines
the meaning of test scores.
- When a test is given a second time, test
takers score better because they have TYPES OF VALIDITY
sharpened their skills by having taken 1. FACE VALIDITY- the least stringent type
the test the first time. of validity, whether a test looks valid to
test users, examiners and examinees.
2. ALTERNATE FORMS RELIABILITY- also - Does not involve statistics.
called “EQUIVALENT FORMS” or 2. CONTENT VALIDITY- the test covers the
“PARELLEL FORMS” reliability. behavior domain to be measured which
- an alternative to test-retest reliability, it is built through the choice of
makes use of alternate or parallel forms appropriate content areas, questions,
of the test. tasks, and items.
LIMITATIONS- can only reduce but eliminate - it is concerned with the extent to which
PRACTICE EFFECTS. the test is representative of a defined
body of content consisting of topics and
- Sometimes, the two forms are processes.
administered to the same group of 3. CRITERION VALIDITY- standard against
people on the same day. When both which is a test or a test score is
forms of the test are given on the same evaluated.
day, the only sources of variation are - can be test score, psychiatric diagnosis
random error and the difference training cost, index of absenteeism,
between the forms of test. amount of time.

3. SPLIT HALF RELIABILITY- test is given CHARACTERISTICS OF A CRITERION

and divided into halves that are scored - relevant
separately. The results of one half of - valid and reliable
the are the compared with the results - uncontaminated: criterion
of the other. contamination occurs if the
4. KR20 FORMULA- also known as a - criterion based on predictor measures;
“KUDER-RICHARDSON 20” it calculates the criterion used is a criterion of what
the reliability of a test in which the is supposed to be the criterion.
items are dichotomous, scored 0 or 1.
4. CRITERION-REALTED VALIDITY
FORMULA: KR20 = N/N-1 {{s2- Σpq)/s2} DEFINED: indicates the test
effectiveness in estimating an
KR20= the reliability estimate individual’s behavior in particular
N= the number of items on the test situation

s2= the variance of the total test score

 - tells how well a test corresponds with a
particular criterion.
TYPES OF CRITERION-REALTED VALIDITY
1. CONCURRENT VALIDITY- test scores
may be used to estimate an individual’s
present standing in a criterion.
2. PREDICTIVE- the scores on a test can
predict future behavior or scores on
another test taken in the future.
3. INCREMENTAL VALIDITY- related to
predictive validity wherein it is defined
as the degree which an additional
predictor explains something about the
criterion measure that is not explained
by predictors already in use.
5. CONSTRUCT VALIDITY- informed scientific
idea developed or hypothesized to describe or
explain a behavior, something built by mental
synthesis.
- it is extent to which the test may be said to
measure a theoretical construct or trait.
FACTORS INFLUENCING TEST VALIDITY
a. Appropriateness of the test
b. Direction/instructions
c. Reading comprehension level
d. Item difficulty
e. Test construction factors
f. Length of test
g. Arrangement of items
h. Patterns of answer