CHAPTER 6 SUMARRY for exam 2 Flashcards
There are three ways to measure reliability?
(a)
the test-retest method, which measures temporal stability;
(b)
the alternate-forms method, which measures forms stability; and
(c)
the internal consistency method (split-half, K-R 20, and coefficient alpha), which measures item homogeneity.
Split-Half Reliability:
A method to assess the consistency of a test by splitting it into two halves (e.g., odd vs. even questions) and comparing the scores from both halves. If the results are similar, the test is considered reliable.
Kuder-Richardson Formula 20 (K-R 20)/ true or question
It’s a way to check how consistent a test is when it has true/false questions. Higher values (closer to 1) mean the test is more reliable.
Coefficient Alpha (Cronbach’s Alpha):
A statistic used to measure the internal consistency or reliability of a scale or test, especially when there are multiple items.
Tests can be validated using five approaches:
content, criterion, construct, known-group, and face.
Content
Content: Refers to the specific elements being measured, like aspects of job satisfaction (e.g., pay, work environment).
Criterion/ hint outcomes
measure used to assess outcomes, such as job performance or employee satisfaction
Construct/ Actions
It’s an idea or trait, that measured based on actions or behaviors.
Known-group:
It’s a way to check if a test can distinguish between groups that are expected to be different, like high and low performers.
Face:
The apparent validity of a test—how much it seems to measure what it’s supposed to, based on a casual look.
Information about tests can be obtained from such sources as
Mental Measurements Yearbook (MMY) \
Tests in Print (TIP)
Mental Measurements Yearbook (MMY)/ hint reviews
it’s a detailed resource that reviews commercially available of standardized tests.
Tests in Print (TIP)
it’s a reference guide that lists all the tests published in the United States, including details like the test publisher and its purpose. Unlike MMY, it doesn’t include detailed reviews, just basic information
Taylor-Russell Tables: (hint predicts)
It estimates the value of a test by considering how well it predicts successful employees
Lawshe Tables/(hint content)
It checks how well a test covers the content it’s meant to measure and if the test items are relevant to that content.
Proportion of Correct Decisions:
Measures how often the test correctly classifies individuals (true positives and true negatives).
Expectancy Tables:
Predict the potential success of individuals based on test scores and help forecast future performance.
Utility Formulas:
Mathematical formulas used to quantify the cost-benefit analysis of using a test, considering factors like improvement in job performance and hiring costs
Adverse Impact:
Tests if a particular group (e.g., based on race or gender) is unfairly disadvantaged by the test compared to others.
Single-group Validity:
Assesses whether the test accurately measures what it’s intended to within a single group (e.g., a specific demographic).
Differential Validity:
Checks if the test predicts outcomes differently for different groups, ensuring it works equally well for all.
Selection decisions can be made in four ways:
top-down, rule of three, top-down with banding, or passing scores.
Top-down:
Candidates are ranked, and the top ones are selected based on their scores.
Rule of three:
Only the top three candidates are considered for selection, even if there are more qualified individuals.
Top-down with banding
Candidates within a certain score range (band) are treated equally, and selections are made from the top-ranked band.
Passing scores:
Candidates who meet a minimum threshold are selected, regardless of how they rank compared to others.