Evaluation Considerations – Low Average ≠ ‘Okay’

May 2018

The abilities of children with hearing loss, whether they are exiting from early intervention or are already school-aged, are typically evaluated to identify overall delays or learning disorders.Since children with hearing loss have access issues learning language due to barriers caused by the hearing loss, they often score ‘low-average’ on norm-referenced language tests. Rather than having overall delays, the access issues caused by hearing loss often result in ‘spotty skills’ or learning gaps that are not identified by typically used evaluation instruments. Because these needs are not identified by typical measures, our students are often denied eligibility for specialized instruction and supports. The specialist in education of students with hearing loss needs to be a member of the evaluation team to help tailor the assessment process to identify the unique needs of these children.

Research has consistently revealed that a ‘good’ result of early intervention for children with hearing loss is a standard score of -1 SD to -1.5 SD on norm-referenced language tests (standard score 78-88 range). All too often teachers of the deaf/hard of hearing have sat in meetings where the evaluation team has described these results as ‘normal’ and ‘he will be okay.’ After all, special education is not preventative, it is for children who have identifiable disabilities. ‘Low-normal’ does not equal a disability. Yet professionals who work with these students realize that there ARE language issues, including ‘Swiss cheese language’ which influences comprehension, delays in syntax learning, and in early literacy skills.

Using Norm-Referenced Tests to Determine Eligibility

The purpose of the testing is to identify an educational disability or adverse educational effect on educational performance. For children with hearing loss, assessment needs to be sufficient in scope and intensity to identify gaps in auditory (or sign language development), language, narrative discourse, academic, literacy, and social language skills. Information needs to be collected that reflects the student’s ability to function in situations similar to the school setting, including typical use of amplification.

Continue Reading the May 2018 Update

Share This!

Evaluation Considerations – Low Average ≠ ‘Okay’

The abilities of children with hearing loss, whether they are exiting from early intervention or are already school-aged, are typically evaluated to identify overall delays or learning disorders. Since children with hearing loss have access issues learning language due to barriers caused by the hearing loss, they often score ‘low-average’ on norm-referenced language tests. Rather than having overall delays, the access issues caused by hearing loss often result in ‘spotty skills’ or learning gaps that are not identified by typically used evaluation instruments. Because these needs are not identified by typical measures, our students are often denied eligibility for specialized instruction and supports. The specialist in education of students with hearing loss needs to be a member of the evaluation team to help tailor the assessment process to identify the unique needs of these children.

Research has consistently revealed that a ‘good’ result of early intervention for children with hearing loss is a standard score of -1 SD to -1.5 SD on norm-referenced language tests (standard score 78-88 range). All too often teachers of the deaf/hard of hearing have sat in meetings where the evaluation team has described these results as ‘normal’ and ‘he will be okay.’ After all, special education is not preventative, it is for children who have identifiable disabilities. ‘Low-normal’ does not equal a disability. Yet professionals who work with these students realize that there ARE language issues, including ‘Swiss cheese language’ which influences comprehension, delays in syntax learning, and in early literacy skills.

Using Norm-Referenced Tests to Determine Eligibility

The purpose of the testing is to identify an educational disability or adverse educational effect on educational performance. For children with hearing loss, assessment needs to be sufficient in scope and intensity to identify gaps in auditory (or sign language development), language, narrative discourse, academic, literacy, and social language skills. Information needs to be collected that reflects the student’s ability to function in situations similar to the school setting, including typical use of amplification.

Norm-referenced tests typically have various subtests, each of which assesses one type of ability area. These subtest scores are rolled together into an overall or total score for the norm-referenced test. It is often claimed that only the overall score from the norm-referenced test can be used to determine eligibility. This is frustrating to persons who work with students with hearing loss as there are often one or more subtests that show areas of need, but the overall score average is within the acceptable range.  Section 300.304(b)(2) of IDEA states that a single measure or assessment cannot be used as a sole criterion for eligibility. So yes, a single subtest score cannot be used to determine a child as eligible. This is misleading. If the area of need identified by the low performing subtest(s) was also demonstrated by other norm-referenced and/or functional measures, then the child’s area of need would have been demonstrated with more than one measure. If there is a substantial need identified, go the extra step to verify it. This can make the difference between eligibility for supports and services or no specialized help for the child. For more discussion on using subtest results to determine the need for further testing, read here.

Another aspect of norm-referenced testing is that the measures are designed to identify areas of delay or disorder. To use an analogy, we can compare the knowledge and skills learned each year to a row of 12 bricks to build a wall. Every year there would be a row built up, starting from infancy. Think of each brick as collectively representing vocabulary and concepts learned during one month of exposure. Consider our students who did not consistently use amplification, or were not consistently exposed to fluent sign, during early childhood and how that would impact their ‘row of bricks development’ as compared to typically developing hearing peers. Norm-referenced tests, specifically the overall scores, consider development as ‘How high is the wall?’ as compared to typical learners. Children who have cognitive delays or learning disorders would have shorter walls. Children with hearing loss are typically found to have walls almost as high as their age peers (low average) but what is NOT identified is the gaps in learning that are typical of children with hearing loss due to communication access limitations that vary over time.

To combat this, the person specializing in the education of children with hearing loss needs to be part of the evaluation team (IDEA Section 300.321(4)(i)(5)). We need to press for evaluation in the areas that we know are at highest risk for issues due to the impact of hearing loss. In looking at the list below, it is clear that the test battery typically used by the school team to evaluate children suspected of learning disorders will not capture the most likely areas of need for students with hearing loss. Moreover, we need to work with the team to discuss evaluation measures that can provide the norm-referenced results of the specific skills needed to tailor the assessment to specific areas of educational need, as required by IDEA (Section 300.304(c)(2)). Refer to Steps to Assessment for information on specific measures.

Areas of learning most likely to be impacted by hearing loss:

  • Understanding group discussions or participating in small group work due to distance/noise in class and socially
  • Vocabulary:  Gaps due to decreased ability to overhear incidental language (‘Swiss cheese language’)
  • Syntax: Incomplete understanding of rules (i.e., cannot hear /s/ or /ed/; do not understand plurals, possessives, past tense)
  • Working Memory: ability to retain fragmented parts of words heard and new spoken/signed vocabulary
  • Listening skills:  Can be challenges with simple discrimination of sounds, phrases or comprehension of conversation or verbal instruction in class (they may hear but not process the full meaning)
  • Attention: Periodic inattention due to listening fatigue and gaps in understanding; ‘tuning out’ when it is challenging
  • Early reading: Phonology/phonemic awareness issues related to not distinctly hearing speech sounds
  • Language processing: due to fragmented hearing, vocabulary gaps, syntax and morphology gaps, slower listening rate, reduced understanding words in context
  • Viewing information from different perspectives, understanding emotions of others, critical thinking
  • Social language: Socially awkward, delays in pragmatic language, nonverbal social cues, and appropriate peer interactions
  • Passive or immature skills in responding when they do not understand what was said; lack of self-advocacy

Influence of CONSISTENT use of hearing devices on language outcomes: Consider our students who did not consistently use amplification, or were not consistently exposed to fluent sign, during early childhood and how that would impact their ‘row of bricks development’ as compared to typically developing hearing peers. Children with more consistent DAILY hearing aid use have better language and auditory outcomes than children with less consistent use, averaging 2/3 of 1 standard deviation difference2. This is especially true for children with hearing loss of 41-70 dB. If children with hearing loss already perform in the low average range for language, an additional 2/3 of 1SD delay can make a lifelong difference in school outcomes. Consistent hearing use is a big key to protecting against language delay and catching up or keeping up with language/learning.

A Predictable Downslide in School Performance: With exposure to a dynamic language environment in a structured classroom setting, many typically hearing children who have low-average language ability can begin to catch up to their more average peers. This assumption cannot be applied to children with hearing loss. A dynamic classroom language environment typically provides less access to communication than what the child experienced in early childhood. It is typical for our students to have their learning trajectory decrease once they enter school, meaning their rate of learning actually declines due to increased issues clearly accessing communication. If a child was not made eligible for specialized instruction, be sure that a 504 Plan is developed and include periodic monitoring by a DHH specialist as part of the necessary auxiliary aids and services (education in regular classes with supplementary services, read FAQ 4).

An evaluation team has the responsibility to appropriate assess students to identify areas of need that will interfere with educational performance. To this end, IDEA Section 300.304 requires that teams gather relevant functional, developmental, and academic information about the child. Academic information is only one part of educational performance. This is especially important for students with hearing loss who often have functional performance issues related to decreased access, such as challenges following directions, participating in group work, listening comprehension, and fatigue. These are all relevant functional performance issues that must be considered when evaluating a student’s need for specialized instruction and support.

References

1. Spencer, P., & Marschark, M. (2010). Evidence-Based Practice in Educating Deaf and Hard-of-Hearing Students. “Children who are identified early and receive early intervention have been found to demonstrate language development in the “low average” level compared to hearing children.” (pg 42) Read more about book.

2. McCreery R. W., Walker E. A., Spratford M., et al. (2015). Longitudinal predictors of aided speech audibility in infants and children. Ear Hear. 36:24S–37S Go to article.

Download this Article

Share This!

Advocacy Notes

Grade Retention

Question from the field:

I was wondering if there was any research or insight on retaining kids who are Deaf/HH for one school year. There have been a few small cases where student who were able to repeat one grade level helped to close that wide gap that they had and eventually (with DHH support) catch up to those of their peers. However, I think the first reaction is – do not retain kids with special needs. Any insight pro or con?

In looking up research on this topic, I found a study of Portuguese students comparing 24 with typical hearing to a group of 20 who were Deaf and implanted and 24 who were Deaf without cochlear implants, in grades 4, 7, 9, 12. In my international travels I have found that many countries have no formalized special education services and all cognitively able students are in the general education setting. Where in the US we do not usually ‘fail’ students with disabilities, this is not necessarily true in other countries. In the case of this Portuguese study, the students who were implanted had a higher failure rate than the typically hearing students, and the students who were considered Deaf but not implanted had a higher rate of repeated school years than the implanted students. See Figure.

While this research finding answers the question of whether students who are deaf would be retained at a higher rate if there is a practice of grade retention for lack of performance (yes), it does not answer the question from the field in terms of pros or cons to retention for students with hearing loss. A 2015 study of predictive factors for academic achievement of this population analyzed data of about 500 DHH secondary students. The authors found that having attended regular secondary schools (vs a deaf school or deaf ed program) and having better spoken language were associated with higher test scores. Predictors of poorer academic achievement test scores included having an additional diagnosis of learning disability, having a mild degree of hearing loss, and being African American or Hispanic.

A really fascinating, but old (1999) study looked at the academic achievement of 75 third grade students who failed first grade screening tests, comparing those who did seek medical attention for the hearing issue to those who did not. The author concluded that more children who had not sought medical attention after failing hearing screening, failed either the first or second grade. The typical retention rate between grade 1 and grade 2 was 9% whereas the students who did not seek medical follow up for hearing issues had a retention rate of 28%. It must be noted that in 1999, the field of education had not embraced the ‘do not retain’ policy held by most school districts today.

Thus, the research reinforces that hearing loss can have a very significant adverse effect on academic achievement, but the studies are very limited. Looking more broadly at the practice of retention of students with disabilities has a bigger research base that has been consolidated nicely into an article by the National Association of School Psychologists. In essence, retention in any grade level is associated with later high school dropout, as well as other deleterious long-term effects. Typically, test scores of retained students in the primary grades increase for a couple of years and then decline below those of their equally low-achieving students who had not been retained. NASP declared that retention is a failed intervention which was partly due to a lack of specific remedial strategies to help enhance student social or cognitive competence.

Summary and Karen Anderson’s reply to the field:
In general, we don’t retain kids anymore. I agree that there are exceptions for students with hearing loss.

Example:

Tommy attended kindergarten. He was an inconsistent hearing aid user and upon entry, the school did not see it necessary to provide DHHT support or the specialized instruction/support provided was not appropriately intensive. Now a year has gone by with only a couple of months of progress made during the year. There were just too many access issues that interfered with Tommy’s instructional progress, too little support for the teacher to understand the behaviors she observed were related to hearing loss, too little support with the parents to recognize the 1:1 correlation between lack of hearing aid use and lack of school progress/behavior/social issues. Now it is spring and people have finally gotten around the table to discuss this Tommy’s lack of educational progress and future needs. He is NOT ready to go on to gr 1 as he really never had access to instruction during Kindergarten. The team agrees that Tommy should repeat Kindergarten but only if he has the intensive support he needs along with consistent hearing aid use. They decided that Tommy will receive strong support (i.e. 1 hour/day) from the DHHT, with 1x/week in the classroom and the rest pull out 1:1 services. This time will be devoted to 1) language development 2) social skills 3) self-advocacy/understanding of hearing loss impact/hearing aid use. At the end of the year the school team expects to see the expected annual yearly progress for a student in the class (hearing) and readiness for grade 1.  If this student were to be retained without the intensity of service and consistent hearing aid use, it would be another wasted year from his education.

Note: The Advocacy section does not constitute legal advice and provides information for thoughtful consideration only.

Download this Article

Share This!