Welcome to the discussion area of the Dyslexia Positive website. The idea is that anyone interested in dyslexia can join in a discussion based on themes initiated by a member of the Dyslexia Positive team. Please participate by commenting on the articles and feel free to ask any questions!
Photograph of Melanie Knight

Posted 8th March, 2011 by Melanie Knight


Significance level .05 .01
Differences between subtest scores  for 13 to 18 years old
Diamonds and Matrices 15 – 18 19 or above
Vocabulary and Verbal analogies 17 – 22 23  or above
Differences between subtests for 19 years and older
Diamonds and Matrices 12 -15 16 or above
Vocabulary and Verbal analogies 14 -18 19 or above

Adapted from table 6.6 from the WRIT Manual (Glutting, Adams and Sheslow, 2000, p.76)

To calculate a learner’s visual and verbal aptitude it is necessary to add the standard scores of the 2 subtests within each domain.  For example to calculate a learner’s visual aptitude score you must add the standard scores for the matrices and diamond subtests together.

However some learners may score significantly better on one subtest than another within a domain.  This poses the question to assessors “Is it safe to combine the learner’s subtest scores to generate a domain score?”

On page 76 of the WRIT manual relevant differences between subtest scores for statistical significance are provided. The differences are listed at 2 levels of statistical significance; p<.05 and p<.01. “P” is an estimate of the probability that the result has occurred by accident.  Therefore the smaller the value of “P”, the greater the statistical significance.  Differences at the .01 level would normally be considered significant by statisticians.

Caution may need to be employed in interpreting domain standard scores when significant differences are found between subtest standard scores ” (WRIT Manual  page 50). However they do not clarify whether it is safe to combine the scores at either the .05 or the .01 level to generate a domain score. Hence it appears to be left to the assessor’s discretion to decide whether to combine statistically significant subtest scores and to comment on the statistically interesting difference.

  • Lesley Wood

    When you read the tables on p76 of the WRIT manual, it does not make it clear (to me) whether you read the lower score from the vertical axis or the horizontal axis.  The differences between the subtests vary, depending on whether you start by looking at the horizontal axis  or the vertical axis.

    • Melanieknight

      Lesley, the scores above the asterisks are significant at the .05 level and the scores below the asterisks are significant at the .01 level. For example if you look at the table 19 years and up a difference of 19 between the verbal analogies and the vocabulary is significant at the .01 because in the vocabulary vertical column the the verbal analogies difference of 19 appears below the asterisk.

      If you then look at the verbal analogies vertical column , the vocabulary difference of 14 appears above the asterisk so it is significant it the .05.

      • Melanieknight

        Since writing the article I attended a Patoss CPD and they recommend you don’t combine scores which are statistically significant at the.05 level

  • Annie

    I’ve only just seen this, I’m glad you’ve added the final comment about
    NOT being able to give a combined score and therefore a General (IQ)
    score as well.
    I went to a PATOSS seminar some time ago, where
    someone explained those WRIT tables at last – they are very unclear. I
    haven’t double checked your table but I’m sure it’s right – and must
    make things a lot clearer for people.

    It would be good to discuss
    perhaps what the implications are about significant differences within
    subtests – with more explanation about what each one shows up and what
    can make a difference. Again the lists in WRIT are rather unhelpful

    • Melanie Knight

      I think you have hit the nail on the head. It is far more important to look at them qualitatively and have a discussion on what they mean for the student.

      Its not only important to consider whether to combine subtests, but whether statistically significant differences, demonstrate a specific strength or weakness within a domain. If they both fall within the average range, qualitatively it is not as significant or interesting.

To receive an email when a new article is published, enter your email address:

NB: you should receive an email asking you to confirm whether you want to subscribe.