Home  Michael  want talk  help talk  AAC: testing: bb learn  pec'd  My Talker Vocab  Role Models

PAPER PRESENTED AT THE CM2006 NATIONAL SYMPOSIUM, UNIVERSITY OF LEICESTER. SEPTEMBER 2006

Observation of Learning and Learning Style as an Assessment Technique in Augmentative
and Alternative Communication
BRUCE R BAKER
Email: minspeak@minspeak.com Website: www.minspeak.com

A NEW FRAMEWORK FOR AAC ASSESSMENT

It is difficult to assess the communication skills and needs of children and adults with significant speech and multiple impair­ments. New procedures are especially critical for learners with physical and com­munication limitations. Traditional methods often underestimate growth and develop­ment potential for people who could benefit from augmentative communica­tion. Many current approaches are not helpful in developing intervention strate­gies. Documenting progress is often difficult in current standardized tests be­cause of high basal criteria. Also, adaptations for physical disabilities invali­date standardization.

The criticisms of formal testing for learn­ers, particularly those with physical and communication limitations, is well docu­mented in the literature and fall into three general categories (Cullen & Pratt, 1992). First, standardized tests tend to give false information regarding the status of learn­ing. Second, standardized tests are biased against some students. Third, standardized tests examine simple skills that are easily tested and overlook more complex think­ing processes.

This paper is going to propose a new framework for testing in AAC. The frame­work can generally be described as the observation of learning and learning style. It may provide greater insights into the real current status of an augmented communi­cator as well as a delineation of a potential next steps in the therapeutic and educa­tional process. The following is a comparison of a traditional baseline status study versus an assessment of learning study.

TRADITIONAL BASELINE STATUS STUDY OF AMY AND ITS RESULTS

Amy was referred at the age of 13 years 10 months. Her diagnoses included severe mental retardation, microcephaly, seizure disorder and cerebral palsy. Amy was am­bulatory and functionally non-verbal. She had received speech, physical, and occu­pational therapies since the age of three. She was medicated with Tegretol for sei­zures and Cogentin for drooling. Her mother sought the least restrictive educa­tional placement for her.

Amy's education profile is described in terms of mental age versus chronological age (Figure 1). These examples are taken from a presentation at the Minspeak Con­ference in Birmingham, Alabama , USA (Clippard & Rice, 1993). The scales and tests are typical of those given in the United States. Current examples are easy to find on the Internet.

Additional anecdotal information from school records suggested that Amy did not know colours consistently, had few if any number concepts, could not sequence, and had difficulty following directions. Her sen­tence structure was incomplete and she did not identify the concepts of 'blond' and 'brunette'.

Amy brought a language board used in some fashion at a previous school. She reportedly used it appropriately for 1 of 8 responses during school assessment. She was very social. Her academic skills suggested she had not benefitted from previous years of traditional developmen­tal instruction. The school district recommended placement in the state school (which is now closed) for individu­als with severe mental retardation

.
ASSESSMENT OF AMY'S LEARNING AND LEARNING STYLE

Clippard & Rice (1993), at the Rush Reha­bilitation Center in Missouri, were asked to do an assessment of Amy's current sta­tus and her potential for growth. The method they adopted was different from the Alabama school district. Rather than test Amy on a mental age versus chrono­logical age baseline study, they decided to work with her for two days and observe how she performed in instructional tasks based on learning in augmentative commu­nication. The language system selected was a 128-location Words Strategy® software with their own customizations for Amy's age and experience level. Rather than use her previous standardized test scores or re-administer them, they watched closely what Amy did whilst they interacted with her in teaching the language software.

The procedures for learning evaluation extended over a multi-day period. Amy was given a 128-location Prentke Romich Company (PRC) device without icon prediction. A Qwerty overlay was used and some vocabulary was stored. The vocabulary included nouns, names, a protest, judgments, descriptors, and emotional statements. Access to some of this vocabulary required symbol sequencing. Amy was engaged in both conversation and in structured interactions. Clippard & Rice closely observed Amy's social, linguistic; and learning behaviours during this multi-day assessment. They found Amy to be very social and used this aspect of her personality to drive their teaching interactions. The results of Amy's second assessment reveals a significantly different child (Figure 2).

ASSESSMENT OF ACTUAL LEARNING VS. BASELINE STATUS STUDY: WHY THE DIFFERENCE?

To the outside observer, the Alabama evaluation seems to be focusing on a different child from the Missouri evaluation. The first assessment gives us a child with severe disabilities who had benefited little from years of clinical intervention and teaching. Her oral motor skills were "totally inadequate," her chronological age, when measured against her skills indicated a child at the lower end of severe cognitive disability in several domains. At the age of 13, her auditory comprehension skills were approximately 2 5 percent of her chronological age.

The second assessment used different procedures and painted a different picture. How does a clinician used these disparate results? Was one (or perhaps both?) assessment simply wrong? Do we merge their different data? But, how?

I contend that both assessments can be reconciled and employed clinically by two methods. The first is not to look toward the mental age/chronological ages ratios but rather focus on actual skill levels es­tablished by the tests. For instance, in the test of Auditory Comprehension, Amy's ability to use grammatical morphemes was been 28 to 30 months. On the Brown's Stages, this would put her between Stages II and III. This would imply the ability to un­derstand subject, verb, object word order and to process grammatical morphemes to modulate meaning. She would be compre­ hending different modalities of the simple sentence - yes/no questions, wh- ques­tions, negatives, etc. Her overall score placed her at 37 to 38 months - the begin­ning of Brown's Stage IV. Rather than compare these skills to the skills deemed appropriate, why not start Amy with the developmental issues of Stages II and III - generating her own two and three word utterances, grammatical morphemes, and sentence modalities.

In second language acquisition, one often sees the phenomenon of a learner who is able to understand utterances in the tar­get language and yet be unable to produce them. Amy seems to be at the "understand­ing" level. Now she needs access to simple words, so she can build her own utter­ances. Her social skills and interest will help her fit these utterances into the informa­tion flow. As Amy builds her own short utterances and fits them into the informa­tion flow, she will be able to test and compare her language output with that of other people.

The three foregoing activities - putting utterances together (synthesis), fitting them into the information flow (embedding), and testing against target performance (matching) are the three classic activities of second language learning (Klein, 1994).

So, one way to use the results of the baseline study was to focus on skill level rather than chronological/mental age ratios. The second method is then to focus on interpersonal language use in playful discourse. Amy's social skills helped her in the playful discourse and kept her involved in the language activity.

Thus baseline studies on standardized tests can be exploited for their skill-based material and observation of learning style can be used as a tool for focusing on lan­guage. A more complete picture of Amy can be derived by focusing on her actual skills rather than by focusing on mental age/cognitive age ratios. These skills can supply a starting place for language inter­vention. Observation of learning and learning style as an assessment technique in augmentative and alternative communi­cation will show how and why an 'engaged' Amy can learn.

There are other problems with standard­ized tests beyond focusing on past learning rather than the ability to learn. At least in the United States, standardized tests in the field of speech and language exert pressure on professionals to empha­size nouns. Tests for young children focus on noun identification. Basal or entry scores often require five or more consecu­tive noun identifications. Functional core vocabulary accounts for 75 percent of what children and adults actually utter. This functional core of approximately 300 to 400 words is not noun rich, yet it is the backbone of language. It is composed of pronouns, determiners, prepositions, con­junctions, simple verbs, helping verbs, etc.

Clinicians are systematically directed to­ward noun teaching as opposed to language teaching to prepare students for success on standardized tests. Core words are rarely used as stimuli or re­sponses. A noun focus may be appropriate for typically developing children but may ignore the language needs and abilities of augmented communicators.

Analysis of toddler and preschool language reveals that 80 percent of a child's speech is made up of core vocabulary. This vo­cabulary has limited noun usage. Pronouns (I, me, mine, you, your, it) and demonstratives (this, that) perform diverse pragmatic and semantic functions including social control, affirmation, and establishing joint atten­tion. Over 90 percent of toddler vocabulary is represented by fewer than 30 core words (Banajee, 2003). One hundred core functional words comprise 73 percent of preschool language usage (Beukelman, 1989). Recorded conversations of adoles­cents about food use only 2.2 percent unique fringe vocabulary (Balandin, 1997).

Speech Pathologists (SLPs) are trained to work with individuals who demonstrate a delay in language development or disor­ders of language. Depending on the work environment, an SLP can work with a vari­ety of clients from young age to older clients, from developing language skills to traumatic brain injuries to stroke patients. In working with preschool children, many SLPs might work in tandem with a school or early start program, utilizing vocabulary and concepts directed by the teaching staff.

Typically, standardized assessments are used not only to qualify a student or cli­ent for therapy, but to direct therapeutic intervention and goal areas. Again, a review of standardized assessments in the field of speech and language reveal that there is a strong emphasis on noun labelling as an early stage of language development, e.g., Peabody Picture Vocabulary Test (Dunn, 1997) and Preschool Language Scale (Zimmerman, 2002). Despite the fact that core vocabulary often comprises a large portion of a pre-schooler's vocabulary, many commonly used assessment tools for young children focus almost exclusively on noun identification and labelling. The references contain many other commonly used standardized tested which exhibit a strong emphasis on nouns. Core words are rarely used as stimuli or responses. In fact, use of circumlocution and describing a noun - important language skills - often results in a decrease in a student's test score.

The noun emphasis in standardized tests does more than fail to measure the lan­guage skills of an augmented communicator. It also directs the attention of clinicians and teachers toward noun in­struction and away from core vocabulary. Core words are essential in the mastery of semantic roles, early syntax (1 and 2 word phrases), basic morphemes, and question structures — Brown's Stages I through III (Brown, 1973).

A clinician working with an augmented communicator must be aware that assess­ments may not reveal a student's full language potential and instead may misdi­rect instructional goals toward context-specific nouns. Noun development cannot be ignored, but it is crucial to ad­dress a student's need for core vocabulary, an area that may not be re­flected in commonly used assessment batteries. *

Bruce R Baker, Linguist & inventor of Minspeak®

REFERENCES
American Guidance Service, Inc. (1997). £x-pressive Vocabulary Test, AGS, 4201 Woodland Road, Circle Pines, MN.
Balandin, S. & Iacono, T. (1999). Adolescent and Young Adult Vocabulary Usage, Aug­mentative and Alternative Communication (AAC), Volume 14, No. 3, September.
Banajee, M., Dicarlo, C, & Stricklin, S. B. (2003). Core vocabulary determination
for toddlers. Augmentative and Alternative Communication (AAC), Vol. 19, No. 1
Beukelman, D., Jones, R., Rowan, M. (1989). Augmentative and Alternative Communica­tion (AAC), Vol. 5, No. 4.
Brown, R. (1973). A First Language, The Early Stages, Harvard University Press, Cam­bridge Massachusetts.

Clippard, D. & Rice, G. (1993). Eighth Annual Minspeak Conference, Proceedings, Use of Augmentative Communication Systems as an Alternative to Traditional Assessment Techniques, Birmingham, Alabama, 1993.
Coplan, J. (1993). Early Language Milestone Scale, PRO-ED, Shoal Creek Blvd., Austin, TX.

Learning for All Students. Stainback, S, & Stainback, W., eds. Eaiamore: Paul H. Brooks Publishing.
Dunn, L. M. & Dunn, D. M. (1997). Peabody Pic­ture Vocabulary Test - Third Edition, AGS 4201 Woodland Road, Circle Pines, MN.
Klein, W. (1994). Second Language Acquisition, Cambridge University Press, Cambridge, UK 1994.
Newcomer, P. L. & Hammill, D. D. (1991). Test of Language Development, PRO-ED, 8700 Shoal Creek Blvd., Austin, TX, USA.

Rossetti, L. (1990). The Rossetti Infant-Toddler Language Scale, LinguiSystems, Inc., 3100 4th Ave. East Moline, IL.
Wiig, E. H., Secord, W., & Semel, E. (1992). Clini­cal Evaluation of Language Fundamentals -Preschool, The Psychological Corporation, 19500 Bulverde Road, San Antonio, TX.

Zimmerman, I. L., Steiner, V. G., & Pond, R. E. (2002). Preschool Language Scale - Fourth Edition, The Psychological Corporation, 19S00 Bulverde Road, San Antonio, TX

COMMUNICATION MATTERS, VOL 20 NO 2, AUGUST 2006, pp - 25