By Komal Sidhu, MS, CCC-SLP.

As speech-language pathologists, quick and accurate assessment is at the heart of our best work. A good assessment not only helps us identify the presence of a disorder but also highlights the strengths and needs of our patients. Truly great assessments also serve to inform our treatment plans. In this sense, accurate assessment is absolutely vital to effectively helping individuals with communication disorders thrive. However, a persistent issue looms over this topic: the overidentification of language impairments in culturally and linguistically diverse (CLD) populations. Overidentification occurs when individuals from CLD populations are identified as having a language impairment when their perceived difficulties may actually be due to cultural or linguistic differences rather than a true disorder. Overidentification can lead to misdiagnoses, inappropriate intervention, and most importantly, the perpetuation of educational and societal inequities.

Our traditional testing tools have a history of incorrectly diagnosing minority students with language impairment. Not only is this racist and classist, this is contrary to our ethical and professional responsibilities as speech-language pathologists. Barragan et al. (2018) found that one in every three Latino children from a low SES community was incorrectly identified with language impairment using the CELF-4 Spanish. Overidentification of a third of the population is ethically and professionally inexcusable because accurate identification of disorders sets the foundation for our work. The consequences of overidentification may include unnecessary therapy sessions that reduce time and access to needed supports and the stigma of being labeled as language impaired. These factors may have far-reaching impacts on a patient’s self-esteem and success.

Much of the issue is due to the biases embedded in traditional standardized testing tools. Populations particularly vulnerable to overidentification include bilinguals and multilinguals, dialect users, individuals from a low socioeconomic status, and cultural minorities. Traditional standardized testing tools are often normed on monolingual standard English speakers from middle SES populations and mainstream cultures, skewing standardized assessment results. In addition to this, traditional standardized testing tools lack the capacity to account for breadth of knowledge across languages and cultures and require specific vocabulary knowledge, putting multilinguals and individuals from minority cultures at an immediate disadvantage. Similarly, these tools often rely on experiential history, or assumed common experiences, as a basis for responses. In addition to multilinguals and minority cultures, this is heavily biased against individuals from a low socioeconomic status and can result in artificially low scores leading to overidentification.

There are many reasons SLPs continue to rely on these flawed measures. For one, monolingual SLPs report a specific lack of confidence in assessing bilinguals (Parveen & Santhanam, 2020), but even bilingual SLPs are bound to encounter a language they don’t speak on their caseload. In many cases, there is a lack of appropriate tests or interpreters in the patient’s language. In these cases, it is understandable that SLPs rely on familiar and available tools; however, it doesn’t have to be this way.

There are several evidence-based, freely available tools that can reduce testing bias in CLD populations. Consider language processing tasks to avoid experiential bias. The nonword repetition task (Dollaghan & Campbell, 1998) is a tried-and-true measure which is quick and informative. Family collaboration can also play an important role, as it provides valuable insight into a patient’s culture and language. One of my favorite tools, the Alberta Language Development Questionnaire (Paradis et al., 2010) is an invaluable and evidence-supported resource for patients with a primary language other than English. There are several language sampling and dynamic assessment probes readily available to help SLPs examine naturalistic use of language and learning ability without relying on biased testing items. All of these tools are powerful pieces of an authentic assessment. Thinking critically about the use of these measures in strategic combination can reduce bias and significantly improve the accuracy of assessment in CLD populations. 

It is crucial for SLPs to engage in culturally and linguistically responsive practices, both ethically and professionally. In addition to selecting appropriate tools, it is vital to pursue ongoing professional development and cultural competence training to become more effective assessors. By shedding light on this issue, I hope to promote awareness of the importance of recognizing testing bias and its consequences as well as encourage the use of evidence-based tools that accurately examine the knowledge and skills of CLD populations. By reducing overidentification in CLD populations, we can help ensure equity in the assessment and treatment of communication disorders.

References

Barragan, B., Castilla-Earls, A., Martinez-Nieto, L., Restrepo, M. A., & Gray, S. (2018). 

     Performance of low-income dual language learners attending English-only schools on the 

     clinical evaluation of Language Fundamentals–Fourth Edition, Spanish. Language, Speech, 

     and Hearing Services in Schools, 49(2), 292–305.
     https://doi.org/10.1044/2017_lshss-17-0013 

Dollaghan, C., & Campbell, T. F. (1998). Nonword repetition and child language impairment. 

     Journal of Speech, Language, and Hearing Research, 41(5), 1136–1146.   

     https://doi.org/10.1044/jslhr.4105.1136 

Paradis, J., Emmerzael, K., & Duncan, T. S. (2010). Assessment of english language learners: 

     Using parent report on first language development. Journal of Communication Disorders, 

     43(6), 474–497. https://doi.org/10.1016/j.jcomdis.2010.01.002 

Parveen, S., & Santhanam, S. priya. (2020). Speech-language pathologists’ perceived 

     competence in working with culturally and linguistically diverse clients in the United States. 

     Communication Disorders Quarterly, 42(3), 166–176. 

     https://doi.org/10.1177/1525740120915205