Lori Vaughn & Janna B. Oetting,
Louisiana State University
Janna: Welcome to the Ph.D. program Lori. You need to learn SALT.
Lori: My head is about to explode from everything in this Ph.D. program!
Janna: You will find that you can’t live without SALT.
Lori: They tried to teach me SALT in graduate school.
Janna: Now is the perfect time to master SALT. As part of your funding, you will be teaching a weekly computer lab on SALT to undergraduates, and we have many research projects that require it.
Lori: Why didn’t you just start with that? We could have saved 15 minutes.
Janna: Come on. You’re going to love it!
Twelve years ago, first as a graduate student and then as a professional speech-language pathologist, I (Lori) avoided SALT like someone on a low sodium diet. I loved collecting language samples from children, but I didn’t experience any joy when transcribing or analyzing samples. Fortunately, my Ph.D. mentor Janna Oetting knew that with training and practice, I would soon appreciate all that SALT offers, and regular use of the program leads to speed and efficiency. I can now proudly say that I mastered SALT within a few weeks of starting my Ph.D. studies and I have taught a semester-long SALT lab for three years while also completing two research projects with SALT for the analysis. In short, I now consider SALT a mainstay in my professional life as a clinician, future university teacher, and scholar.
As my mother always says, “When you know better, you do better.” As a shout out to my mother, mastering SALT has allowed me to better understand the linguistic profiles of the children I serve and better explain these profiles to families using many of the evidence-based measures SALT generates once a sample is transcribed. The EXPLORE option in SALT also allows me to customize codes for any aspect of language I want to measure, including the local New Orleans dialect of the children I serve. Finally, the normative database of SALT allows me to talk with families about how their child compares to others using objective measures that can be tracked across time.
I (Janna) agree with Lori. SALT is a great tool that can save the clinician and researcher a great deal of time, but it also takes training and practice, as do many complex clinical activities. Within the undergraduate SALT labs, each week is devoted to a specific set of skills. The students first learn to elicit a sample in a uniformed way, and then they learn how to transcribe and analyze their samples. We start with measures of intelligibility that are generated by SALT and end with a pragmatic analysis that exploits the EXPLORE menu. While many students purchase the student version of SALT, others transcribe and code their samples in a word processing program and copy their work to SALT during lab for their analyses. Either approach is valid, and we talk to students about how to make transcription and analysis feasible in the real world.
As a teacher, I love it when students tell me that they will never play, talk, or listen to a child in the same way as they did pre-SALT. I too have a hard time not listening at the morpheme code level when I/’m work/ing AND when I/’m relax/ing with family and friend/s! As I often tell doctoral students who teach the lab and colleagues who ask: SALT is a clinical tool but it is also a great teaching tool. Even if students never become speech-language pathologists or never use SALT in clinical practice (although we hope they all do), a semester of weekly SALT labs changes how a student thinks about language and interacts with others. Post-SALT training, students listen more purposefully and are better able to strategically manipulate their language to elicit language from others.
In 2018, I (Lori) attended my first Symposium for Research in Child Language Disorders and presented a study co-authored with Dr. Oetting. The study examined language samples from 106 children who spoke either African American English (AAE) or Southern White English (SWE). Within each dialect, half of the children were classified as specifically language impaired (SLI) and the other half were classified as age-matched typically developing (TD) controls. The study focused on the children’s relative frequencies of 38 different nonmainstream dialectal forms (e.g., zero copula BE as in She Ø cold; zero verbal -s as in He always runØ to the store; BE leveling as in They was talking). A previous study of 62 children by Oetting and McDonald (2001) showed that the relative frequencies of these nonmainstream forms could accurately classify the dialects of 97% of the children, and within each dialect, > 90% of the children were also accurately classified as either SLI or TD.
In our 2018 study, we also found high levels of classification accuracy for the children’s dialects, and within their dialects, their clinical classifications. In fact, using nothing more than the relative frequencies of the 38 nonmainstream dialectal forms, 88% of the children were accurately classified as a speaker of either AAE or SWE, and within each dialect, > 90% of the children were accurately classified as either SLI or TD. Across the 2001 and 2018 studies, some of the same tense and agreement forms also led to group (SLI > TD) differences. In the 2001 study, these forms included zero BE, DO and irregular past. In our 2018 study, these forms included zero BE, DO, HAVE, regular and irregular past tense, and regular and irregular verbal –s. We would not have found these patterns had our language sample transcription, coding, and analyses not been facilitated by the SALT program.
Our findings have a number of practical implications, but most importantly, they underscore the importance of measuring children’s nonmainstream dialectal forms within assessment. And measurement means not only listing which nonmainstream forms a child produces but also how often the child produces the forms relative to other types of mainstream forms when conversing with others, telling stories, and completing school-based assignments. Since the 1980s, our field has discouraged professionals from including children’s nonmainstream dialect forms in assessment because of concerns that dialect differences would be interpreted as a language disorder. Our findings show that children’s use of these forms serve as both a marker of their nonmainstream dialect, and within their dialect, as an indicator of a disorder when some of these forms are produced at particularly high frequencies. In other words, to find and understand a child’s language disorder WITHIN the context of a child’s dialect, one has to measure the child’s dialect just as one has to measure a child’s use of Spanish, French, or Japanese when finding (and understanding) a language disorder within the context of these languages.