Textbook Notes (270,000)
CA (160,000)
UTSC (20,000)
Psychology (10,000)
PSYC37H3 (100)
Chapter 5

PSYC37H3 Chapter Notes - Chapter 5: Discriminant


Department
Psychology
Course Code
PSYC37H3
Professor
Bouffard
Chapter
5

This preview shows pages 1-2. to view the full 8 pages of the document.
Validity
Validity: the agreement between a test score and the quality it is believed to measure
Does the test measure what is supposed to measure?
Three types of evidence include construct-related, criterion-related, and content-related
Unitary concept that represents all of the evidence that supports the representation of a measure
The consensus document cautions against separating validity into sub-categories
The use of categories does not mean that there are distinct forms of validity
Aspects of Validity
Face Validity
Face validity: the mere appearance that a measure has validity
Not a technical form of validity, but is used in testing literature
Exists if the items seem to be reasonably related to the purpose of the test
Not really validity because it does not offer evidence to support conclusions drawn from test
scores
In many settings it is ideal to have a test that looks like” it is valid, because these appearances
can help motivate test takers since they can see that the test is relevant
Content-Related Validity
Content-related validity: considers the adequacy of representation of the domain the test is designed to
over
If you are being tested on the first five chapters of the book, then content-related evidence is
provided by the correspondence between the items on the test and the information in the chapters
The only type of evidence (except face validity) that is logical rather than statistical
www.notesolution.com

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

Attempt to determine whether a test has been constructed adequately
The wording and the appropriateness of the reading level must be considered
Multiple judgers rate each item in terms of its match or relevance to the content
Factor analysis has been used to determine whether items fit into conceptual domains
Construct underrepresentation: describes the failure to capture important components of a
construct
If a test of mathematical knowledge included algebra but not geometry, then the validity is low
Construct irrelevant variance: occurs when scores are influenced by factors irrelevant to the
construct
A test of intelligence might be influenced by reading comprehension, test anxiety, or illness
Criterion-Related Validity
Criterion-related: tells us just how well a test corresponds with a particular criterion
Occurs when there is a high correlation between a test and a well-defined criterion measure
Used when we want to know how well someone will do on a job
Determined because the test is to serve as a stand-in” for the measure we are interested in
Predictive validity: the forecasting function of tests
A measurement of how well a test predicts future performance
The validity of a test is established by measuring it against known criteria
When students apply to colleges, they are required to write the SAT (the predictor). The belief is
that the test scores can predict how well a student will perform in college
One measure occurs earlier, and is meant to predict some later measure
Concurrent validity: when a test correlates well with a measure that has previously been validated
The two measures may be for the same construct, or different
www.notesolution.com
You're Reading a Preview

Unlock to view full version

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

The two measures are taken at the same time, unlike predictive validity
If a test measuring job satisfaction gives similar results to those gathered using a job satisfaction
measure which has been validated in the past, then the new measurement has concurrent validity
Validity Coefficient
Validity coefficient: the relationship between a test and a criterion
The coefficient tells the extent to which the test is valid for making statements about the criterion
A coefficient of 0.60 of more is rare, but a coefficient ranging from 0.30 to 0.40 are commonly
considered high
The squared result is the percentage of variation in the criterion that we can expect to know in
advance because of our knowledge of the test
Example: We know 0.40 squared, or 16%, of the variation is college performance because of the
information we have from the SAT test
Why the University of California Rejected the SAT-I
SAT-I is the traditional test that evaluates reasoning ability, while the SAT-II is an achievement
test that evaluates student knowledge in particular areas
SAT-II was a better predictor, and the SAT-I contributed little or nothing to the prediction of first
year grades
SAT-I scores were more sensitive to the socioeconomic background, and even when controlled,
this test was still inferior to SAT-I
The Cholesterol Test: Predictive Validity Evidence
High levels of cholesterol in the blood can help predict early death from heart disease and stroke
The level of association is actually low
Those with high levels of blood cholesterol are told to eat foods low in saturated fats and
cholesterol
Some studies have failed to find a strong relationship between dietary habits and mortality rates
www.notesolution.com
You're Reading a Preview

Unlock to view full version