Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


What is a good test? Reliability, Objectivity, and Feasibility

No description

Jason Langley

on 29 August 2018

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of What is a good test? Reliability, Objectivity, and Feasibility

What is a good Test? Reliability, Objectivity, and Feasability
Jason G. Langley Ph.D. NSCA-CPT, CSCS
EXSC 394
Fall 2018

Test should obtain approximately the same results regardless of the number of times it is given
Test given to same group on one day should yield same results on the second day
Factors effecting test repetition: fatigue, motivation, environmental conditions, and measurement error
For a test to be valid it must be highly reliable
Validity can only be claimed if the test is highly reliable
Example: Driving in a golf class
Hitting a ball off the tee is a valid measurement of golf skill, but the inconsistency makes it less reliable
Highly reliable tests are not necessarily valid
Reliability and Validity
Several methods that all are reported like correlation coefficients from -1.00- +1.00
Closer to +1.00 means greater reliability
Performance tests most often have higher reliability than skill tests
Criterion referenced tests and reliability
Determined by how consistently masters and nonmasters are classified
Procedure is to compare mastery on two separate days
Determining Reliability
Three methods
Requires two administrations of the same test
Use product -moment or ANOVA to calculate the coefficient
Time interval must be considered
Learning effect, motivation, environment
Parallel Forms Method
Give equivalent tests at the same time
Same calculation
Difficult to construct two test of the same content
Reliability of Norm-Referenced Tests
Split-Half Method
Split the test in half and correlate those scores
Only need one test, most often used with cognitive tests by comparing even and odd numbered questions, could also be used for skill tests
Reliability will increase with number of test items
Reliability of whole = 2 x reliability of half divided by 1 + reliability of half
Reliability of Norm-Reference- Part 2
Reliability Factors to Remember
Method of scoring
Objective versus subjective
Heterogeneity of subjects
Reliability coefficients are based on certain groups, the smaller the better
Test length, longer = higher reliability
Administrative procedures
Test by having two individuals administer the test to the same group
Similar results equals high objectivity
Form of reliability between testers
Multiple choice, T/F, matching if there is a key is highly objective; vertical, shooting accuracy
Essays, gymnastics, diving, figure skating highly subjective
Clear instructions
Trained testers administer the test
Simple measurements
Use of technology to limit error
Need scores not phrases
Used when there are multiple tests available that are valid, reliable, and objective
Ease of administration
Use a game specific test it is more of an authentic assessment
Full transcript