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? Part 2

Validity, Relability

Jason Langley

on 31 August 2018

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of What is a good test? Part 2

What is a good test? Part 2
Jason G. Langley Ph.D. NSCA-CPT, CSCS
EXSC 394
Fall 2018

Most important criterion to consider when evaluating a test (remember this when we start to evaluate actual tests)
Actually it is more the agreement between what the test measures and the performance, skill, or behavior the test is designed to measure
Example: CRF and tennis
Population specific
Validity of tests vary, so choosing the best is based on statistical evidence using the validity coefficient
Similar to the correlation techniques we have already discussed (range from -1.00 to +1.00, +1.00 is most valid)
If a test is being compared to a previously validated test, a new test will need >=.80 coefficient
Directly related to behavioral objectives
Objectives should be clear, in the exact manner expected, and limited to small sections of instruction
Example: knowledge test, questions should be directly tied to objectives, several items to each objective
2nd method- pre/post, a significant improvement after instruction or a certain number of students master the objectives
The standard for mastery is key
In domain referenced validity, if the item represent the criterion behavior the test has logical validity
Decision validity is used to classify proficiency, above a cut score proficient, below nonproficient
Validity of Criterion-Referenced Tests
Three types of validity evidence are reported
The best scenario is if there is strong evidence for more than one type
Content (AKA, logical or face)
Does the content of the test cover the objectives required for the skill, sport, exam?
Experts may identify the skills necessary
Claimed often in physical tests of agility, balance, and flexibility
If possible, use in conjunction with another type of validity because logic can be debated
No statistic to calculate
Validity of Norm-Referenced Test
Evidence is indicated by how well a score correlated with a specific criterion, +1.00 is best
Two types
Predictive (future performance)
Talent identification
College entrance exams
Concurrent (current performance)
Measures current level of skill compared to current criterion not a future one
Used to estimate validity
Criterion Validity
An important consideration is the criterion choice
Three are most often used
Expert ratings
Use a player of experts to rank the players from best to worst during play
Correlate rankings to skills test
Tournament play
Round robin tournament
The best test takers will finish higher if the test is valid
Previously validated test
Most often what we do in exercise science with estimations and alternative methods
For example, maximal oxygen consumption and body fat analysis
A few of the measurements are "Gold Standards" to which all other measurements are compared for validation
Criterion Measure
Third form of validity in which evidence refers to the degree to which an individual possesses a trait (construct)
Examples are anxiety, intelligence, and motivation, they can't be seen but are expected to behave in one way
Cardiorespiratory fitness is a construct
Can also be demonstrated by comparing higher-skilled and lower-skilled individuals
Construct Validity
Characteristics of test takers: only valid for same:
The criterion measure
Administrative procedures
Pre-test instructions
Factors Affecting Validity
Full transcript