Test-Retest Minimum Time Advice
Pearson has conducted no general studies addressing the issue of retesting, thus cannot recommend a specific time interval between testing periods. We do, however, suggest that you review the test-retest reliability information published in the manuals. It provides you with an idea of what to expect when a retest is conducted after a short time period.
Professional clinical judgment may deem retesting appropriate when the child matures into a new age or grade-specific item set different from the previously administered set. Again, the decision is based on professional judgment.
Retesting Considerations circa 2016
We often are asked, what is an appropriate or optimal time interval before a child can be retested with the same tool? There isn’t a standard for an "optimal" time interval for retesting. There are some things to consider before retesting a child in a relatively short time frame:
- Has enough time elapsed so that the child isn’t likely to remember his or her answers? Retesting within 0–14 days is not recommended for this reason on language assessments, especially receptive language tasks. We generally find in our test–retest studies that a child's test scores are slightly (1–3 standard score points) higher when he or she is retested the same day or less than 15 days later.
- Has enough time elapsed between tests so that the child has “aged-out” to the next normative group? It is best to wait until enough time has passed so that the child’s chronological age places him or her in the next normative group.
- Is there a reason to believe that the first test administration did not represent the child’s best performance, e.g., was the child was tired or ill? If you think the child may not have been able to perform well during the first test, you should consider retesting the child. However, retesting too soon may result in slightly higher standard scores due to practice effects.
- Has enough time elapsed between tests that the child has made progress? If so, you may want to retest to evaluate progress. If there is a question about prior test results, you may want schedule a retest to obtain a second opinion. If the answer is no to any of these questions, it’s best to wait to retest until one of the questions can be answered yes. Information about test–retest score differences are included in the Technical manual in the Test Reliability, Test Stability or "Test–Retest Study sections.
Some employers may require clinicians to test more frequently than once or twice a year to assess a client's progress. Most standardized assessments were not developed to be administered repeatedly in a short time frames (e.g., every 6 weeks) as progress measures. Standardized tools can be used as progress measures when the answer to any of the above questions is yes, testing at the beginning and end of treatment.