Hello all,
I am working on an RCT where we hope to measure 5-6-yr-old children's executive functioning (memory, cognitive flexibility, inhibitory control, attention, etc... exact domains not exactly nailed down yet) 3 times, 2 months apart (baseline, end of month 2, and end of month 4). I'm not finding much literature on the test re-test reliability of such tests, nor am I finding anything addressing testing effects. I know many tests (such as DCCS) have ways to change the task to avoid such effects (e.g., switch which rule you start with), but only for two measurements, not three. If anyone can point to any resources regarding this, I would be most appreciative. Also important to note; we are conducting these tests in West Africa, so would prefer to avoid screen based measures (as they might be too far out of the realm of kids' normal experiences).
Thank you in advance!
This is trickier than one would expect, isn't it? What you really want are neuropsychological tests of executive functions that have parallel (a.k.a. equivalent) forms. Nothing will have three forms, but with two, you could start with "A", then uses "B", then go back to "A" again without worrying about effects of repeated testing.
In general, psychometric data on executive function tests tend to be scanty (and often a bit disappointing when available). Luckily, you don't need to worry too much about test-retest reliability since you are (presumably) looking for change over time due to some intervention or other. But a few measures will report test-retest findings in the Manual.
As far as domains go, you're probably best advised to break it down like this:
(1) working memory
(2) fluency
(3) response inhibition
(4) set shifting
(5) higher-order concept formation/problem-solving
Naturally, many tests will tap into more than one of these (and often other cognitive abilities as well).
The fact that you are working with very young children adds another layer of complexity, since many tests will be beyond their ken. Even a simple, useful measure like the Stroop (functions (3) and (4) above) works only after reading becomes an automatic process.
Well, anyway, here are a few possibilities:
(1) The Delis-Kaplan Executive Function Scales (D-KEFS) is an extensive executive function battery (the first one ever developed, I think). A few of the subtests even have alternate forms. Be careful extrapolating from U.S. norms, naturally, but at least the norms are based on a large, well-constructed sample. A larger problem is that the norms extend only to age 8, so it isn't clear what will happen if you administer to younger kids. It is 100% okay to use just selected subtests. It's a sophisticated measure, and will yield rich data, but your examiners will need to be very well trained.
(2) The CANTAB is probably out, since it's a computer-based measure.
(3) The Test of Everyday Attention for Children (TEA-Ch) is more limited in scope, but includes measures of inhibition and set shifting. The norms are weak, but do extend down to age 6.
(4) Behavioural Assessment of the Dysexecutive Syndrome in Children mostly measures planning-related functions. A nice feature is that the subtests resemble "real world" situations, enhancing ecological validity. Again, the norms are weak, but they extend down to age 7 or so.
(5) The NEPSY-II is an extensive battery of cognitive tests that includes several executive function measures. The norms, which reach down to age 3, are reasonably good (as with every other measure, they're not based on African samples, though!).
My sense is that the D-KEFS and the NEPSY-II are probably your best bets, but that the BADS or TEA-Ch could work as well. As a final thought, you may want to use these tests to construct your own measures that are geared for the population you are studying. For example, one subtests on a couple of these batteries is Animal Sorting. Well, you'd want to use animals with which the kids in question are familiar! Similarly, if you're doing a fluency task (like the F-A-S), you'd want the initial letters to make sense in the language spoken by the kids.
Thank you so much - this is extremely helpful! I very much appreciate your thorough and informative response.
We did three studies (oben with Patients After strok, ohne with Patients with addictive diseases and one with Patients with Rheumatische diseases) measuring Memory, Executive Functions with Word-fluency and cognitive speed. To get effects of Memory functions we needed Training once a week for at least three month. To get effects in Executive functions and word fluency we needed Training three times a week for 4 weeks with Socio-Communicative cognitive Training in groups.
- Badges
- Science topic