WJ IV

 Woodcock&Willis1

Richard W. Woodcock and John Willis

Question and Answer on the Use of Grade Norms for the WJ IV by John Willis

July 2, 2015

Re:  WJ IV Scoring *

I just tested a girl who has finished grade 1, and entering grade 2 in mid August. I scored the test designating her in grade 2.0 rather than grade 1.9. Her parent preferred that I use grade 1.9.
So, what is the more accurate way to designate the correct grade for a student when tested in the summer. Appreciate   your help with this.
NAME REDACTED
 
 
Nancytucson gave the official information for the WJ IV: 1.9 as she hasn’t started 2nd grade yet.  
 
The Examiner’s Manual Woodcock-Johnson IV Tests of Achievement (Nancy Mather & Barbara J. Wendling.  Rolling Meadows, IL: Riverside, 2014) states on p. 37: 
 
The online scoring program automatically calculates the examinee’s chronological age and tenth-of-school-year grade placement (based on a standard school year). If the student is enrolled in a year-round school or a school with starting or ending dates that fall more than 2 weeks before or after the default range (i.e., August 16 through June 15), use the option for entering exact starting and ending dates of the school year. Due to the wide variation in starting and ending dates for schools and districts, use this option regularly to increase the precision of the grade norms accessed by the scoring program. After entering the starting and ending dates into the scoring program, it automatically calculates the exact grade placement, in tenths of the school year.
With achievement tests that use seasonal norms (e.g., fall and spring; fall, winter, and spring; or autumnal equinox, winter solstice, vernal equinox, and summer solstice), there can be seasonal assessment disorders when a student’s score drops several points overnight.  See, for an example with the old WIAT-II, http://www.myschoolpsychology.com/WIATII.pdf.  Obviously,the wider the norms band (e.g., half years vs. thirds of years), the greater the overnight changes.  (On some tests, the winter norms are simply interpolated.)  The lower the child’s grade, the greater the effect.  It is also greater for some achievement domains than for others.  (Such jumps can also be also a concern for young children with tests of cognitive ability, especially if the age spans in the norms tables are broad, e.g., 12 or 6 vs. 4 vs. 3 or 2 months.)   If a child’s age or grade placement is close to the border between norm groups, I think it is always prudent to glance at the adjacent page in the norms tables to see if testing a few days earlier or later would have made a substantial change in the scores.
Tests with seasonal norms must set arbitrary dividing points in the summer.  I find it frustrating when I test a child in late summer after the official dividing point for the particular test (e.g., July 1), long after the child has forgotten many of the math skills taught in the past spring and I must use norms for fall of the next grade, which the child has not yet started.  Obviously, I must obey the rules in the particular test’s manual, but I can also offer a second table of what the scores would be by other norms as long as I clearly explain in table and text that those are not official scores.  As Gary Canivez observed, age-based norms would avoid that problem.
One of the many virtues of the Woodcock-Johnson is that norms are based on one-month intervals, so differences in scores between grade 1.9 and 2.0 or age 7:2 and 7:3* would be trivial.  You can check by temporarily changing the test date and rescoring.  The seasonal assessment disorder is not a concern with the WJ IV.
Avner Stern commented that, “Grade norms are often more relevant for teachers, as I understand.”  There is a problem, though, when a child has been red-shirted (such as a “Readiness” or “Transition” year crammed in between kindergarten and first grade, or a repetition of one or more grades in the belief that repeating the same action will yield a different outcome).  It is unreasonable to expect a second grader to do fourth grade math no matter how old or tall the child may be (so grade-based norms might be more appropriate).  However, it is not defensible to make a child repeat several grades because of reading deficiencies and then triumphantly announce that the child is at long last reading almost at grade level (and passed his driver’s license test with oral administration).
 
In some cases, when there is a noteworthy difference between scores based on age norms and scores based on grade norms, it may be best to report both sets of scores.  The whole truth does not reside in either set of scores alone.
 
Similarly, if the test manual says to use norms for grade 1.9, but the parent wants scores by norms for grade 2.0 (or any other comparison, for that matter), I would score the test twice and report both sets of scores (carefully and clearly identifying in both table headings and text which set is based on the test manual requirements and which set is not).  I’d rather switch than fight.
 
Another issue is comparisons of “ability” and “achievement” scores (or, much better “predicted achievement” and “achievement” scores), assuming that is the sort of thing one wants to do for some reason.  It makes no sense to me to compare achievement by one set of norms and ability or predicted achievement by another set of norms.  Consider Johnny, whose IQ standard score was 100, who is of average height and weight for his age, and who has been retained twice in grade.
 
                          Percentile Rank
                            by norms for
                           Age     Grade
 
Intelligence         50          90
Reading                 1          50
Height                  50          90
Weight                 50          90
                          Percentile Rank
                            by norms for
                           Age     Grade
 
Intelligence         50          
Reading                            50
Height                  50          
Weight                              90
 
Conclusions: Johnny is reading just fine and he is obese.
 
Off-hand, I cannot think of an individually-administered cognitive ability test other than the Woodcock-Johnson that offers grade-based norms, so if, for some odd reason, I felt compelled to compare a child’s “ability” scores with the child’s “achievement” scores and I wanted to use grade-based norms, I would have to use the Woodcock-Johnson COG.  If I also use the WJ IV ACH and OL, I have the advantage of scores all based on the same national sample of unsuspecting victims.  I would also have the options of comparing very similar tests of listening comprehension and reading comprehension and of using the General Intellectual Ability, the Gf-Gc Composite, and/or the various specific Scholastic Aptitude Clusters.
 
However, if the child is also undergoing language testing, neuropsychological testing, or other testing scored with age-based norms, I need to provide achievement-test scores also based on age norms.  Some achievement tests offer only age-based or only grade-based norms.  (Again, I often elect to report achievement-test scores by both age norms and grade norms.)
 
John Willis 
 
* You can find very useful tables of average ages for grade placements and average grade placements for ages (and tons of other cool information) in
Mather, N., & Jaffe, L. E. (2004). Woodcock-Johnson III: Reports, recommendations, and strategies (with CD). Hoboken, NJ: Wiley.
and
Mather, N., & Jaffe, L. (in press). Woodcock-Johnson IV: Reports, recommendations, and strategies (with CD). Hoboken, NJ: Wiley.

__._,_.___

Posted by: “John O. Willis” <johnzerowillis@yahoo.com>

Boilerplate (shell) for reporting WJ IV TCA scores by John Willis
WJ IV Report Table Shell *

Approximate percentages of the population expected to obtain differences between various WJ IV subtests
Diff req for sign btw Ability tests All Cog 7.28.14 *

Differences between WJ IV subtests by Age Required for Significance
Appr Percentage of Pop to Obtain Discrep Between Various WJ IV Tests 7.15.14 *

Differences between WJ IV Gf-Gc Cluster Tests by Narrow Abilities and Input and Output Demands
task differences COG 7.28.14 *

How to evaluate the unusualness (base rate) of WJ IV cluster or test score differences: It is a pleasure to use the correct measure – A SlideShare presentation   by Kevin McGrew *

LINKS TO ONLINE REFERENCES *

Kevin McGrew’s Mind Hub.  Scroll down for multiple references to the WJ IV.
The Mind Hub Special Reports and Publications

Riverside Publishing web for the Woodcock Johnson IV
Overview and Ordering Information Links WJ IV

Link to Assessment Service Bulletins (ASB Research)
Assessment Service Bulletins for WJ IV

Scoring Options:   Online only