The language on page 19 is a slippery slope as it builds in too much room for bias and subjectivity from the testing adults:

"Careful consideration of bias and diagnostic accuracy or error rates is suggested when examining performance on norm referenced or standardized tests. Research shows that cultural and linguistic differences may result in an impact of up to 35 standard score points depending on the particular test and individual student’s cultural background and language skills. Teams should discuss the impact of regional dialectal differences, common family or cultural customs, lack of practice, and other factors that while appropriate for the individual student, may result in a lower score due to inappropriate comparison with the test norming population. 

Rather than examining the student for cultural background and language skills - the better approach is to examine the testing instrument being used to assess the student.  The test instrument needs to be assessed for cultural and linguistic appropriateness.  

It is more appropriate to consider the language spoken and/or dialect used by the child before selecting a standardized assessment tool and make sure that the assessment tool is appropriate.  We should not be looking at the child’s differences to explain testing results but rather look at the test, not the child.  

Educators should refrain from creating assumptions about students or their families based on general cultural, ethnic, or racial information.

When you allow the adults to subjectively look at the ethnic and regional dialectal differences of the family and the child, then you are opening a huge pathway to the dark side of testing that Virginia has exploited for racist purposes in the past.