What is Computer Adaptive Testing (CAT)?
Computer Adaptive Testing (CAT) uses Item Response Theory statistical models to select items from a large item pool that are most appropriate for an individual. This minimizes the number of items needed to ensure an accurate score, thus providing an individualized and meaningful assessment. For each assessment, all respondents begin with the same item in the middle of the scoring range and the response to that item then dictates which item will appear next (a harder or easier item), tailoring the items to the individual and avoiding irrelevant items. With each new response, the score and confidence interval are re-estimated and the CAT software determines whether the pre-set stopping rule (the desired level of precision or a pre-set number of items), has been satisfied. If satisfied, the assessment ends and a score is provided. If not satisfied, new items are administered until the stopping rule is satisfied.
Is the PEDI-CAT an electronic version of the PEDI?
No, the PEDI-CAT is not just a computer version of the paper-pencil version of the Pediatric Evaluation of Disability Inventory (PEDI). Completed using pre-installed software on a computer or iPad, the PEDI-CAT uses a statistical algorithm that selects questions from the PEDI-CAT’s large set of items, based on the respondent’s age, gender, mobility device use, and previous responses. The PEDI-CAT items and content were developed to capture a broader age range and skill level than the original PEDI. While the PEDI-CAT continues to have 3 Functional Skill domains, there is no longer a Caregiver Assistance Scale. The Caregiver Assistance Scale has been replaced with the Responsibility domain. Also, the response scales have been changed from the PEDI 2-point “Capable” and “Unable” scale to a 4-point difficulty scale for the 3 Functional Skill domains (Daily Activities, Mobility, and Social/Cognitive). A 5-point response scale is used for the Responsibility domain.
When purchasing the PEDI-CAT, will the download be useable on only one computer, or can the license extend to the department or district? Are there multiple license discounts if more than one license is needed for more than one computer?
- PEDI-CAT for Windows – The license is for one computer for 1 year. However, multiple users can use the same computer.. For instructions on how multiple users can use the same computer contact email@example.com
- PEDI-CAT for iPad – The PEDI-CAT app has a perpetual license and includes 5 downloads per Apple ID.
Is there a version of the PEDI-CAT for Mac users?
No, the PEDI-CAT is only available for Windows and for the iPad. It will run on a Mac using Boot Camp or an equivalent program.
I checked power wheelchair, so why didn’t I get Mobility items using a power wheelchair?
Items to assess mobility using a power wheelchair are not yet available for the PEDI-CAT. The option to check power wheelchair on the demographics page is to provide an accurate description of the child’s devices.
Has the PEDI-CAT been translated into other languages?
The current version of the PEDI-CAT is available in Spanish and comes with the English version. Anyone interested in translating the PEDI-CAT should contact firstname.lastname@example.org to obtain translation permission and guidelines. As of November 2016, the following PEDI-CAT translations are in progress: Portuguese, Italian, German and Dutch.
Administration and Scoring
If I do a Speedy assessment do I have to do a Speedy every time I assess the patient?
No. Scores on the Speedy and Content-Balanced CAT, though not identical, have been shown to be within the margin of error suggested by the standard error.
Can I choose the items to be administered?
You may choose the domains (e.g. Daily Activities, Mobility etc.) to be administered and the version of the PEDI-CAT (Speedy or Content-Balanced) to be used. The PEDI-CAT software then provides an individualized assessment for each child because the computer algorithm selects items from a large item pool that are most appropriate based on age, gender, mobility device use, and responses provided for each item throughout the assessment.
Can older children complete the assessment themselves?
The PEDI-CAT can be administered by professional judgment of clinicians who are familiar with the child or by parent/caregiver report, but has not yet been validated as a self-report measure for children.
Because the items are selected by the software program, some skills that I would like to document don’t show up. For example, I might not get any items for tooth brushing but I want to document the child’s ability to brush his/her teeth.
The PEDI-CAT is not intended to be used to document level of difficulty for specific skills or tasks but provides an overall description of functional activity performance. Computer Adaptive Tests use a computer algorithm to pre-select the items that will be administered to an individual based on responses to previous items (not randomly). This approach minimizes the number of items that are administered to obtain an estimate of functioning in any particular content area (e.g. Daily Activities). Items that are not relevant or are not needed for a precise score estimate of overall functional performance are filtered out.
Why do I get different questions every time I do an assessment?
The PEDI-CAT’s computer algorithm selects items from the large item pool that are most appropriate for that child and minimizes the number of items needed to ensure an accurate and precise score of overall functional activity performance.
If I do an initial assessment and then a discharge assessment and the same questions don’t come up, how can I compare the results?
Results from assessment (e.g. initial) to assessment (e.g. discharge) can be compared because the PEDI-CAT provides scores which indicate an overall functional activity performance. Even if different items are administered, all items have been calibrated and placed along an ability scale of easier to harder items so as children improve their functional abilities, harder items will be administered.
How should a clinician read and interpret the PEDI-CAT scores? Is there an age equivalent?
For each of the 4 domains, in addition to scaled scores, normative standard scores (provided as T-scores and age percentiles) are calculated. Please note however, that the PEDI-CAT does not provide an age-equivalent. Normative scores describe the child’s performance in comparison to other children of the same age (in one year intervals). For T-scores, the mean for each age group is 50, with a standard deviation of 10 (same format used for normative scores in the original PEDI). Typically, T-scores between 30 and 70 (i.e. mean ± 2 standard deviations) are considered within the expected range for age. Scores below 30 indicate decreased functional ability compared to what is typically expected for that age range. Scores above 70 indicate scores above what is typically expected for that age range.
The age percentile range (i.e. 5~25) indicates the percentage of children of the same age group whose scores were as high as or higher than the child being assessed. If a child scores below the 5th percentile (<5), it can be considered ‘below average” while 5th~25th can be considered “low average”. When using PEDI-CAT scores for service eligibility decisions, we recommend that the child should be identified as eligible if EITHER the T-score OR the percentile range is below the criterion.
When testing a child with hemiplegia, is it OK to score test items for the involved side so you can see progress?
The PEDI-CAT is not a capacity-based assessment but rather a measure of overall functional activity performance. PEDI-CAT items were written to focus on the outcome of activity performance and purposely allow a variety of methods to be used. Items focus on the child’s ability to perform each functional activity in a manner that is effective given their abilities and challenges. Items do not require the child to perform the activity in a standardized manner. The items on the PEDI-CAT were carefully worded to allow children to complete activities using alternative methods. This lessens the extent to which children with physical or communication difficulties are penalized in scoring due to use of adaptations or technology (e.g. augmentative communication devices, switches).
Is it possible to see significant changes from month to month with the PEDI-CAT?
The PEDI-CAT can be used on multiple occasions for the same child (e.g. initial, interim, discharge and follow-up) and there is no minimum time that must pass between assessments. Responsiveness testing has been completed for children and youth in inpatient hospital rehabilitation.
Can you say “I don’t know” 5 or 6 times in a row?
Yes. However, each “I don’t know” removes an item as deemed appropriate for use by the software and another comparable item is used in its place.
I have had times where inappropriate questions were generated when I used the Content- Balanced assessment. I completed a PEDI-CAT Content-Balanced assessment for a 15 year old and it asked if he pumps his feet on a swing. For a Content-balanced assessment that I did for a 2.5 year old, it asked if she could remove a dollar bill from a wallet. Why did this occur?
Occasionally, a question which may seem irrelevant, is presented on the Content-Balanced assessment. The software algorithm requires that a certain number of items be administered in each content area of the domain being assessed. If almost all other items in a content area have been administered, the program will be forced to administers whatever item(s) remain, including some that may seem less relevant based on a child’s age or abilities. The selection of items is not based on age but rather on the level of performance indicated by previous responses. So, if a very young child showed generally good manipulation skills, then an item like removing a bill from a wallet would be appropriate from a performance perspective even though it is not a likely functional skill for this age group.
The possibility of irrelevant items is reduced with the Speedy CAT. It is quicker and doesn’t require content balance among the items.The Content-Balanced CAT is most appropriate for children whose estimated functional abilities are in the middle to higher end of the scale. When completed on very young or children with lower level functional abilities, the Content-Balanced CAT may present questions that seem irrelevant as the software tries to meet the requirement for content balance.
Are there specific scoring criteria for each of the PEDI-CAT items?
There are no specific scoring criteria for each item. Responses should be indicative of the child’s typical performance for that activity. During the development of the PEDI-CAT, each item was reviewed by parents of children with and without disabilities for clarity to represent a specific task. Pictures are included for the Daily Activities and Mobility domains to help respondents understand the intent of each item.
Why doesn’t the assessment report show the Standard Error (SE) of the T-scores?
The SE of the T-score is the same within each domain because it is an estimate of the SE for that domain. An estimate for each domain is provided in the manual in Chapter 2. As with all measures, the SEs are likely to be higher at the extreme ends of the score ranges.
In the original PEDI, the score profile visually shows the changes and whether the confidence intervals are overlapping for multiple assessments. Do you think that a score profile could be incorporated in the assessment report?
Rather than score profiles, the PEDI-CAT provides item maps for each domain. The PEDI-CAT item maps depict the location of individual item scores along the continuum of difficulty for each domain. Scaled scores are estimates of the placement of an individual child along this continuum. When you print the item maps from either the Content-Balanced or Speedy PEDI-CAT, this vertical line is drawn in the map for you, and the confidence interval around the line is shaded gray.
Why is the PEDI-CAT on a 20-80 scale metric instead of a 0-100 scale?
This was done to allow new items to be calibrated and added to enlarge the item bank and improve the psychometric properties of the assessment. This work is not yet complete.
Can I print a score report?
Yes, score reports can be printed from either the Windows or iPad versions. A print option is available on the score report screen and can be exported to Microsoft Excel, detailed instructions available in the PEDI-CAT manual in Chapter 2.
Is it possible to save a case and return to fill out new information on separate occasions when administering/scoring the PEDI-CAT?
No, you must complete the assessment before closing the PEDI-CAT program.