Home > Learning Difficulties > Special Education Community

Question about the Woodcock Johnson Tests of Achievement


SaturnV May 25, 2013

My son was administered the WJ-III ACH test before 2nd grade in the following areas (standard scores and percentiles are listed)
Broad Reading - 94 33rd %ile
Academic Skills - 102 55th %ile
Letter-Word Identification - 98 45th
Reading Fluency - 96 39th
Passage Comprehension - 89 24th
Calculation - 110 74th
Applied Problems - 97 42nd
Spelling - 102 56th

I'm thinking of getting him retested with a different test along with a test of cognitive ability rather than achievement. He was reading since he was 3 or 4 so I don't see how his passage comprehension and letter-word identification skills could only be average. How much cultural bias is on this test and how many questions are there per grade level? The Reading Fluency test requires the examinee to read a sentence and answer yes or no to it in a given time of 3 minutes. Is the passage comprehension test timed as well (this test requires the examinee to fill in missing words in a sentence or paragraph)? My son is very meticulous and careful when it comes to testing so I'm wondering if this could deflate his scores. Another thing that confuses me is there is a separate "grade band profile" that shows higher standard scores as follows...

Broad Reading - 102
Academic Skills - 113
Letter-Word Identification - 107
Reading Fluency - 102
Passage Comprehension - 95
Calculation - 123
Applied Problems - 101
Spelling - 107

The scores also indicate that his proficiency levels were different on this score profile. Can someone explain this?

So my main interrogative points are as follows.

-How much cultural bias is on this test and how many questions are there per grade level in applied problems?
-Is the passage comprehension test timed? I've also heard teachers complain that the passage comprehension test is culturally biased so how can it be an effective measurement of reading skills.
-Why were his scores in the grade band profile higher than the age band profile if the scores are from the same test? His relative proficiency levels were different too, but it was the same test.

Post a reply
Facebook  Digg 


Sort by:  Oldest first |  Newest first 


Aybee27 August 10, 2013

First, I'll say that I am a school psychologist and use the Woodcock Johnson regularly.

I would not say that it is accurate to say that there are a certain number of questions per grade level. There are entry points for various grade levels. The examinee answers questions until they either miss a certain number of questions in a row (the ceiling) or they reach the end of the subtest. If they miss one of the first three items (generally), they test backwards from the start point until they have three in a row correct (the basal). This assumes that they did not start at the first item, which first and second graders usually do.

The Passage Comprehension test is not timed. Examinees have as much time as they need (within reason) to read and respond to passages. The first few items have picture prompts to help the reader determine potential responses. Later items do not contain pictures, are somewhat longer, and contain progressively more difficult vocabulary. To reach the ceiling, you have to respond incorrectly (or not respond, or say, "I don't know") to 6 consecutive items. As far as timed tests go, ones that are measures of fluency are timed, while all others are not timed. So, Reading Fluency, Math Fluency, and Writing Fluency are all timed. No other tests on the achievement battery are timed. There are some on the cognitive battery that are timed.

The difference in scores is due to the normative sample used. Think about it: not all kids that are in the spring of first grade are the same age. Thus, if you use grade norms, you are comparing against the normative sample of other children in your child's grade. They can range anywhere from the same age as your child to one year older or younger than your child. If you are using age based norms, you are comparing against other kids who are 7 years and 'x' months old. Those children could be first graders, or second graders.

If your child's age based scores are lower than his grade based scores, I would guess it is because he is older than the average child in his grade (but not by much). For what it's worth, I think it is a mistake for a psychologist (or whoever is giving the test) to report both age and grade based scores as it can only be confusing for parents, and has no benefit. That's really just my opinion though. Some folks think that age-based norms are good for making eligibility decisions (because most IQ tests that they would be compared against only have aged-based norms), while grade based norms are better for making instructional decisions. By the way, just FYI, the Woodcock Johnson Cognitive I think is the only cognitive test that has the option for grade based norms. I will say that I have a bias toward age based norms, probably because I am a psychologist as opposed to a special educator, who would be more likely to rely upon grade based norms.

By the way, in your case it doesn't much matter which norms you use, as the interpretation would be pretty much the same (all the scores fall within average limits and would not be cause for concern). Granted, it is not good practice to rely upon one set of scores to make any decisions or interpretations grading a child's abilities. You have to look at the bigger picture. If all other information says something totally different than...say...a score on the Passage Comprehension subtest, then you might decide that the Passage Comprehension test was not the best measure for that person.

As far as cultural bias goes, most if not all tests of academic achievement contain some level of cultural bias, since they measure academic knowledge that is based on the American curriculum, which is biased--necessarily--toward American culture. Generally, tests of cognitive abilities are the ones that have varying degrees of cultural bias. As far as that goes, the Woodcock Johnson does contain some tests that have a degree of cultural bias, but overall is one of the less culturally biased batteries, as the developers of the test are very conscious of cultural and linguistic variables in testing.

Search Community