Validity and Reliability

Get Started. It's Free
or sign up with your email address
Rocket clouds
Validity and Reliability by Mind Map: Validity and Reliability

1. Validity: Checking to ensure a test measures what it is intended to measure.

1.1. Content Validity

1.1.1. determined through inspection of the test questions as a means to determine whether or not the correspond to what the material the user is being tested on.

1.2. Criterion-Related Validity

1.2.1. validity based on the correlation with external criterion. Two types: concurrent and predictive.

1.2.1.1. concurrent

1.2.1.1.1. validity determined by if and how well the results concur with other similar tests.

1.2.1.2. predictive

1.2.1.2.1. validity accuracy determined by how well the evidence or results predict the future outcomes of those being examined.

1.3. Construct Validity

1.3.1. validity determined when the relationship to other information is correctly corresponding to the theory, or rationale accounting for the 'interrelationship' to a set of variables.

2. Reliability: Consistency of results for individuals taking the test more than one time.

2.1. Test-Retest or Stability

2.1.1. A means of estimating reliability through testing and then retesting and correlating the two test scores.

2.2. Alternate Forms or Equivalence

2.2.1. Reliability tested through the use of two forms of the test to estimate the reliabiity of the test scores.

2.3. Internal Consistency

2.3.1. Two types of internal tests to measure a single concept: split-halves and odd-even reliability.

2.3.1.1. Split-Halves

2.3.1.1.1. Split-halves determined by splitting the test questions in half, equally to determine the correlation between the two halves.

2.3.1.2. Odd-Even

2.3.1.2.1. Odd-even reliability, similar to split-halves, differing by determining the correlation of the questions, not by half, but dividing the questions by even and odd number questions.

3. Validity and Reliability are both important factors in creating tests. When tests are not reliable or valid, then their results will not be able to clearly indicate whether or not the questions being asked are important, appropriate, necessary or an adequate means of evaluating the students being tested.

3.1. Reference: Kubiszyn, T. & Borich, G. (2010). Educational testing & measurement: Classroom application and practice (9th ed.). John Wiley & Sons, Inc., Hoboken, NJ.