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. Content Validity Evidence

1.1. Reviews test items to ensure that they are consistent with the instructional objectives.

1.2. Does not inspect whether a test item is poorly written

1.3. Easier to check for achievement tests

1.4. Yields a logical judgement

2. Criterion-Related Validity Evidence

2.1. Types

2.1.1. Concurrent New and established assessments are administered at the same time Checks for a correlation between the distribution of both test scores Yields a validity coefficient

2.1.2. Predictive Administers an assessment to subjects and then measures the predicted outcome over a period of time Checks for correlation between first assessment scores and scores collected after a period of time Yields a predictive validity coefficient

3. Construct Validity

3.1. Determined but correlating assessment scores with scores from a rationale or theory

3.2. No anchor to validate the assessment

4. Test-Retest or Stability

4.1. The same assessment is administered to the same subjects over a period of time

4.2. The reliability coefficient is affected by the time frame

4.3. The greater the time frame, the lower the reliability coeffient

5. Alternate Forms or Equivalence

5.1. Alternative and equivalent assessments are administered to the same subjects over a short period of time

5.2. Assessments must be administered under the same conditions

5.3. Creating two equivalent tests is difficult

6. Internal Consistency

6.1. Methods

6.1.1. Kuder-Richardson methods Determines whether an assessment represents a single and consistent measure of a concept There are two widely used methods KR20 KR21

6.1.2. Split-half methods The assessment is split into two halves and administered to two different subject groups of the same class The Spearman-Brown prophecy formula is used to calculate the internal consisteny

6.2. Problems with internal consistency estimates

6.2.1. Should only be used for assessments measuring a single concept

6.2.2. Yields inflated reliability when used for speeded tests