split-half correlation
A researcher calculates the split-half correlation to assess test reliability.
Noun: A statistical measure of reliability, specifically a correlation coefficient calculated between the scores obtained on two halves (e.g., odd-numbered items vs. even-numbered items) of a single test or assessment. This coefficient is interpreted as an estimate of the test's internal consistency or reliability.
This term is used primarily in the fields of psychometrics, educational testing, and psychological assessment. It describes a specific methodological technique for evaluating how consistently a test measures a construct. A high split-half correlation suggests the test is reliable.
- The researcher reported a split-half correlation of 0.85, indicating strong internal consistency for the new personality inventory.
- To estimate reliability, we calculated the split-half correlation between the odd and even items of the exam.
- A low split-half correlation may suggest that the test items are not measuring the same underlying trait.
- The split-half correlation is often adjusted using the Spearman-Brown prophecy formula to estimate the reliability of the full-length test.
- This method assumes the two halves of the test are parallel, meaning they have equal difficulty and variance.
- Reliability Coefficient (n): A general term for any statistical index of a test's consistency, which includes the split-half correlation.
- Internal Consistency (n): The degree to which items on a test measure the same concept; the split-half correlation is one method to assess it.
- Spearman-Brown Formula (n): A formula used to correct or adjust a split-half correlation to estimate the reliability of the entire test.
- Split-half reliability coefficient
- Internal consistency reliability (when estimated via the split-half method)
- To calculate a split-half correlation: The action of computing this specific statistic.
- The first step in our analysis was to calculate a split-half correlation for the survey data.
A researcher calculates the split-half correlation to assess test reliability.
- a correlation coefficient calculated between scores on two halves of a test; taken as an indication of the reliability of the test