UMNOU CHAPTER 4  The idea that a test-taker's score on the

targeted ability or trait is predicted to

Test and Testing
increase when they react more correctly or
consistently with a given trait as indicated
by the test manual.
Assumption 1: Psychological Traits and
States Exist Assumption 3: Test-Related Behavior Predicts
Non-Test-Related Behavior
Trait
▪ Some tests are used to postdict behavior rather
 “Any distinguishable, relatively enduring than predict it in the future. to comprehend actions
way in which one individual varies from. that have already been taken.
“
 tends to be a characteristic or pattern of ▪ The test's goal is to give some indication of other
conduct that is more persistent and conditions.
steady.
▪ the examinee's behavior in some aspects.
States
Assumption 4: Tests and Other Measurement
 Distinguish one person from another Techniques Have Strengths and Weaknesses
but are relatively less enduring (Chaplin
▪ The test user should be aware of the test they will
et al., 1988)
be using.
 A temporary way of being, including
one's thoughts, feelings, actions, and
Of gr Assumption 5: The assessment process
interpersonal relationships, is referred includes a variety of sources of error
to as a state.
▪ The term "error" refers to the widely-held belief
Construct that factors other than those that a test seeks to
measure would affect performance on the test.
 A well-informed, scientific idea created
or put out to characterize or ▪. Traditionally, the term "error" refers to something
explain behavior. that exceeds expectations; In actuality, it is a step
 are invisible, audible, and tactile, but we in the measurement process.
can deduce their presence from their
outward behavior. Assumption 6: Testing and Assessment Can Be
Conducted in a Fair and
Overt Behavior
Unbiased Manner
 Refers to an observable action or the
result of an action that can be seen, ▪. The test user who tries to administer a particular
including answers from tests or test to individuals whose background and
assessments. experience differ from those of the individuals for
whom the exam was meant is one cause of
Assumption 2: Psychological Traits and unfairness-related issues.
States Can Be Quantified and Measured
Assumption 7: Testing and Assessment Benefit
▪ Defined constructs that need to be measured Society
▪ Consider the test object that is intended to be ▪ Imagine a world without tests. How would we
representative of the construct, qualities, or determine whether officials like Bong-bong Marcos
stats being measured. and others possess the knowledge, abilities, and
intelligence necessary to serve as president?
Cumulative Scoring
Good Test
 Represent the strength of the targeted
ability or trait or state.  Logically, the criteria for a good test would
include clear instructions for administration,
scoring, and interpretation.
UMNOU
Reliability Sample
 The criterion of reliability involves the  a percentage of the population considered
consistency of the measuring tool: the to be representative the total population.
precision with which the test measures and
the extent to which error is present in Sampling
measurements.  The method of determining the area of the
 We want to have some degree of assurance universe that is representing the entire
that the test or measurement device we are population.
utilizing is reliable.
Types of Sampling
Validity
1. Stratified Sampling
 A test is considered valid for a particular
purpose if it does, in fact, measure what it  A technique for selecting samples from a
claims to measure. partitionable population in to smaller
populations.
Example: If an intelligence test actually does
measure intelligence, it is valid. 2. Purposive Sampling

Norms  When choosing population members to take

part in the test, participants were
 Norm-referenced testing and assessment purposefully chosen using a sampling
as a technique for assessing a test taker's technique that is based on a criterion.
performance and a strategy to interpret test
results that compares their results to those 3. Convenience Sample/Sampling
of other test takers.
 One that is convenient or available for use.
 The test performance data of a certain set
of test takers that are intended to be used Developing Norms for Standardized Test
as a guide when analyzing or interpreting a
single test result are known as norms in a ▪ The test developer administers the test according
psychometric context. to the standard set of instructions that will be used
with the test.
Normalized Sample
▪The test developer additionally outlines the
 Individuals whose performance on a specific suggested configuration for making the test. It must
test is examined to use as a guide for be continuous.
assessing how well each test-taker did.
Types of Norms
Norming
1. Percentiles
 The method of obtaining norms.
 An expression of the percentage of people
Race norming whose score on a test or measure falls
below a particular raw score
 The divisive practice of enforcing norms
 a percentage of test takers represented by
based on race or ethnicity background.
a converted score.
Program Norms/ User
 Which "include descriptive statistics based
on a set of test results" takers over a
specific time period rather than norms
determined by formal techniques for
sampling.
2.UMNOU
Age Noems Fixed Reference Group Scoring Systems
 Describe the typical results of several ▪ The dispersion of test results from one group of
samples of test takers who were at various students'
ages at the time the test was administered.
Test takers
3. Grade Norms
▪ Serve as the foundation for calculating test scores
 Created to show the typical test for upcoming administrations of the exam.
performance of test takers in a specific
academic grade. Norm-Referenced Versus Criterion-Referenced
Evaluation
4. Developmental Norms
Norm-Referenced
 a term applied broadly to norms developed
on the basis of any trait, ability, skill, or other ▪ One way to derive meaning from a test score is to
characteristic that is presumed to develop, evaluate the test score in relation to other scores
deteriorate, or otherwise be affected by on the same test.
chronological age, school grade, or stage of Criterion-Referenced Evaluation
life.
 Defined as a method of evaluation and a
5. National Norms way of deriving meaning from test scores by
 Derived from a normative sample that was evaluating an individual’s score with
nationally representative of the population at reference to a set standard.
the time the norming study was conducted. Domain/Content Referenced
6. National Anchor Norms  Mastery of a subject matter is measured.
 give test results some consistency by tying Correlation
them to other test results.
 is an expression of the degree and direction
7. Equipercentile Method of correspondence between two thing.
 The calculation used to determine the Graphic Representations of Correlation
equivalent scores on various tests is a
comparison of matching percentile values. Scatterplot

8. Subgroup Norms  is a simple graphing of the coordinate

points for values of the X -variable (placed
 Any of the criterion can be used to segment along the graph’s horizontal axis) and the Y
a normative sample. initially employed to -variable (placed along the graph’s vertical
choose sample individuals axis).
9. Local customs  The advantage of scatterplots is that they
quickly show the direction and strength of
 offer normative details regarding the local any link between the two variables.
population’s performance on some test
performance of a population on a test. Curvilinearity
 Usually created by test subjects themselves  in this context refers to an “eyeball gauge”
of how curved a graph is.
Outlier
 is an extremely atypical point located at a
relatively long distance—an outlying
distance—from the rest of the coordinate
points in a scatterplot
  a percentage of test takers represented by a
converted score.
UMNOU

Standard error of the estimate.

 This error in the prediction of Y rom X is
represented by the standard error of the
estimate.
Standard Deviation
8
 Descriptive, typical or average
difference between the data points from
their means.
Multiple regression
 The use of more than one score to
predict Y requires the use of a multiple
regression equation.
Inference from Measurement
Meta-Analysis
 The term "meta-analysis" refers to a
group of methods for statistically
combining data from various research to
create a single estimate of the statistics
under consideration.
Culture and Inference
 Responsible test users must keep in
mind that culture affects how tests are
administered, scored, and interpreted.
Therefore, when choosing a test to
employ, a responsible test user does
some preliminary study on the test's
available norms to see whether or not
they are appropriate for usage with the
intended test taker demographic.
UMNOU