Home
Subjects
Create
Search
Log in
Sign up
Upgrade to remove ads
Only $2.99/month
Social Science
Psychology
Psychometrics
Test 2: Measurment
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (50)
Measurement
the process of assigning numbers according to some rule or convention to aspects of people, jobs, job success, or aspects of the staffing system
The measures enables improvement of the staffing system by ?
identifying patterns useful for understanding and predicting relevant processes and outcomes
Types of Measures
-Jobs, people, and performance (Attributes)
-Predictors (KSAs, interviews, tests, biodata)
-Criteria (objective and subjective measures)
Measurement Scales (4)
Nominal
Ordinal
Interval
Ratio
Nominal
Categorical
(Gender, religion, national origin, etc)
Ordinal
Rank Order, distance between scores doesn't matter, just the order of them
(Rankings, some HR measurement. Example like winning first place, second place, third place)
Interval
Equal Intervals
(Intelligence scores, personality scores,scoring keys for interviews)
Ratio
Zero has meaning
(Ex. salary, weight, typing speed, sales per month)
Scoring
The process of assigning numerical values during measurement
Raw Scores
the unadjusted scores on a measure
Raw Scores: Criterion Referenced Measures vs. Norm-Referenced Measures
The scores have meaning in and of themselves
The scores have meaning only in comparison to the scores of other respondents
Normal Curve
a symmetrical, bell shaped curve representing the distribution of a characteristic
Ex. height, weight, typings skill, math ability
Percentile Score
a raw score that has been converted into an expression of the percentage of people whose score falls at or below that score
Ex. 84th percentile means that 84 percent of people scored at or below that person
Central Tendancy
describes the midpoint or center of data
mean, median, mode
Variability
describes the spread of the data around the midpoint
range, variance, standard deviation
Standard deviation
how much your data set varies from the average
so how far it is away from the middle of the bell curve or the average
Standard Scores
converted raw scores that indicate where a person's score lies in comparison to a referent group
A standard score is ________ when the target individual's raw score is below the referent group's mean, and ________ when the target individual's raw score is above the referent group's mean
negative
positive
Correlation Coefficient
o The strongest is 1 or negative 1
o The sign tells us the direction of the relationship
o How far from zero tells us the strength of the relationship
tells us the relationship between 2 values
(if it is the same then the correlation will be 1. The closer to 1, the more reliable)
Reliability vs. Validity
Refers to how dependably or consistently a test measures a characteristic
-Deals with the inferences made from the measure
-Whether the measure is a measure of what you think it is, and/or whether it predicts what you think it does
Validity is ________ by reliability
limited
Reliability of measurement
how consistently a measure assesses a particular characteristic
-A measure is reliable to the degree it is free of measurement error
Reasons for differing scores on a test or assessment (4)
-Temporary physical or psychological state
-Environmental factors
-Version, or form, of the measure
-Different evaluators
Types of Errors (4)
-Random Error
-Systematic Error
-Deficiency Error
-Contamination Error
Random Error
error that is not due to any consistent cause
Ex. running into traffic on the way to work or experiencing bad weather can cause an employee productivity to randomly fluctuate
Systematic Error
error that occurs because of consistent and predictable factors
Ex. an employees productivity may go down every tuesday because he or she works late at a second job monday nights
Deficiency Error
error that occurs when you fail to measure important aspects of the attribute you would like to measure
Ex. if you wanted to measure the applicant's ability to use calculus but the test you used focused on algebra
Contamination Error
error that occurs when other factors unrelated to whatever is being assessed affect the observed scores
Ex. if the calculus test had many complex word problems then language could influence the results
Classical Test Theory
-Obtained measure is an indicator of an underlying construct
-Obtained score consists of at least two components (True score,
Error (systematic and random))
-The less error, the more accurate and reliable
Estimating Reliability (4)
-Test retest
-alternate forms
-internal consistency
-interrater
Test-Retest Reliability Estimates
Take measures of the same characteristics (or people) using the same instrument at two different times
compute simple correlation between the two measures
Ex. scale that measures your weight and then do it again a month later
Alternate Forms Reliability
Take measures twice
Indicates how consistent scores are likely to be if a person completes two or more forms of the same measure
Internal Consistency Reliability
Extent to which items measure the same construct (unidimensional) and share variance
Slit Half: split test and correlate 2 halves
Cronbach's Alpha: the average of all split half correlations; estimates reliability of using the average of items
Interrater Reliability
how consistent scores are likely to be if the responses are scored by two or more raters using the same item, scale, and measurement
Interrater Agreement (agreement is absolute) vs. reliability (reliability focuses on pattern)
2 Rater Case --> simple correlation between raters' rating
More than 2 raters --> intraclass correlation
Standard Error of Measurement (SEM)
margin of error that you should expect in an individual score because of the imperfect reliability of the measure
Represents the spread of scores you might have observed had you tested the same person repeatedly
The lower the standard error the more __________ the measurements
accurate
The Confidence Interval
represents the degree of confidence that a person's "true" score (zero) lies within their earned score plus or minus the SEM, given some level of desired confidence
Validity of Selection Devices
How useful a measure is
The degree to which the devices are job-related (if used as a selection device)
To establish job relatedness you must conduct a valid job analysis that identifies the essential job functions and job specifications (knowledge, skills, and abilities required)
Measurement Validity: Criterion Related
Is the measure related to the criterion of interest?
Statistical Relationship between scores on the device and job performance
Concurrent: taken concurrently from a current sample of employees
Predictive: taken from applicants at a later time
Measurement Validity: Content
Is the measure representative of job-related behaviors?
Representative of tasks, behaviors, and knowledge required for the job
Job analysis info and judgement provide evidence of validity
Measurement Validity: Construct
Does the measure actually measure what you think it measures?
Your selection device must adequately measure the construct
Validity Generalization
the degree to which evidence of validity obtained in one situation can be generalized to another situation without further study
What is Face Validity
Whether the measure seems to measure what it is supposed to measure
a subjective assessment of how well times seem to be related to the requirements of the job
often important to job applicants who tend to react negatively to assessment methods if they perceive them to be unrelated to the job or not face valid
Even if a measure seems face valid, if it does not predict job performance, then it should not be used
A standardized score is a _______________
z score
it describes where you are in relation to the mean
Statistical Significance
The degree to which the relationship is not likely due to a sampling error.
Tells us if it is significantly different from zero
Practical Significance
The relationship is large enough to be of value in a practical sense
Small correlations might not be practical but could be statistical
T or F: A correlation that is statistically significant is not necessarily large enough to be of practical significance
true
T or F: For a correlation to be practically significant, it HAS to be significantly significant
True
Can something be more valid than it is reliable?`
NO
Reliability sets the limit on validity
Test Retest
o A coefficient of stability
o you make sure the ratings are consistent, you want it to be close to 1 (.7 or better when rating the same thing)
o we would like the ratings of 2 interviewers to one interviewee be close to the same
This set is often in folders with...
Week 1 Budgeting
20 terms
Week 2 Forecasting- Part 1
20 terms
Week 1 Sustainability
12 terms
Staffing Overview
26 terms
You might also like...
Measurement Personnel Psych: Test 3
70 terms
Chapter 8 PowerPoints
58 terms
Chapter 8 - Staffing
52 terms
Organizational Staffing Chapter 8
87 terms
Other sets by this creator
Chapter 12: Quiz Questions
6 terms
Chapter 11: Quiz Questions
15 terms
Ch. 12: Channels and Distribution
31 terms
Ch. 11: Pricing
47 terms
Other Quizlet sets
SC Pre-Licensing A2
152 terms
ECN 202 : exam 2
34 terms
ems 9
33 terms
SYG3325 Exam One
42 terms