Resources: Dynamic Assessment

I am officially in my first month of my clinical fellowship year and ready for more learning ahead. I am starting out in a school district with a primarily low-income, multilingual population of students, who also often experience high-risk situations and trauma at home. There are few classes or clinical experiences that can prepare you for how to assess and analyze the assessments of children in these situations. Dynamic assessment, at least in my program, seemed so untouchable to me when I was in graduate school--I knew its definition, but not how to actually do it. Yet the more I have used dynamic assessment tools, the more I see why they are so vital to understanding the nuances of children who do not fit into the middle class, English-speaking standardized tests that so many of our assessments are based upon.

I do not want to ignore the great standardized assessments that have come out for Spanish speakers, such as the BAPA (Bilingual Articulation and Phonology Assessment), and CELF-5, Spanish Edition, and for non-mainstream American English speakers, such as the DELV (Diagnostic Evaluation for Language Variation) Assessment. I have used the first two, and they are very helpful in making comparisons and getting a clearer picture of a child's articulation and language.

I started 5 months ago in this district on a student waiver, so I am still a novice in every way, but I feel I've already learned about two resources in particular I'd like to share.

Firstly, if you haven't gotten a chance to work with SALT 20 (Systematic Analysis of Language Transcripts), try to get your hands on a copy. They have clinical, instructional, research, and student versions. They are expensive, but worth the cost, especially if you take the time to train yourself or can find someone to help get you started. They have free online trainings as well.

If you haven't heard of SALT, it is a computer program that helps you analyze language samples and compares them to databases of other children's language samples that are bilingual (Spanish-English, French-English) or monolingual. They are working on more languages and have an Australian/New Zealand version as well.

How do you do this? You have options for how you elicit a language sample: story retell, unique story tell, or exposition. There are different wordless picture books for the story retells for different ages K-8. You can have the child explain a game, have a conversation with you, tell you a story, or retell a story you told them, record the sample, transcribe the sample, add the transcription codes, and then run it through SALT. The program then breaks down the sample into discrete categories (verbal facility, semantics, syntax/morphology, utterance length, and errors in a sample) and compares those categories to other children based on criteria you set (gender, age, grade, etc) and 1-2 standard deviations from the mean of the group. It takes some time to learn the transcription codes, but once you learn them and start using it, it is SO HELPFUL in understanding your student's language. You can also add code schemes to see where your students fall amongst other students in terms of their narrative abilities and semantic units.

My advice if it seems overwhelming? Just try it! Learn it little by little. It's worth the time.

Secondly, I highly recommend The Leaders Project https://www.leadersproject.org/. Their videos, free materials, research "nuggets" (they break down research AND tell you how to you can use it clinically) https://www.leadersproject.org/law-policy-directory/relevant-research/, and intervention resources are incredibly helpful for learning more about dynamic assessment, including how to perform it and tell others about its importance--you could even get to a point where you replace standardized assessments with dynamic assessments if you're thorough. The research is there and it is often better for our diverse country.

I've included a table I made from the bibliography presented in a video that the Leaders Project founder, Catherine Crowley, posted in November 2019. Comment or ask questions about it if you have them!

Dynamic Assessment

Crowley (2019) Bibliography

Study Citation	Purpose, Procedure, Findings	Quotes/Important Notes
Dollaghan, C. & Campbell, T. (1989). Nonword repetition and child language impairment. Journal of Speech, Language, and Hearing Research, 41, 1136-1146.	Investigation of processing-dependent, nonword repetition tasks in 20 school-age children enrolled in language intervention and 20 TD age-matched peers to minimize biases in language performance	Children with language impairments perform significantly worse than their non-impaired peers on measures of nonword repetition Nonword repetition tasks can accurately distinguish between children with and without language impairments Nonword repetition tasks tax: Working memory, auditory processing, and organization of articulatory output Remove biases of standardized testing Not technically dynamic assessment as no mediation is involved, but can be part of a larger battery of tests
Henderson, D., Restrepo, M.A., & Aiken, L. (2018). Dynamic assessment of narratives among Navajo preschoolers. Journal of Speech, Language, and Hearing Research, 61(10), 2547-2560.	4- to 5-year old Navajo preschoolers. Predictive Early Assessment of Reading and Language (PEARL) pre-and-post-test distinguished LI from TD with 89% accuracy and modifiability scores with 100% accuracy	Arises from Vygotsky’s (1978) learning theory and Feuerstein’s (1977, 1979, 1981, Feuerstein & Ran, 1974) theory of mediated learning experiences and cognitive modifiability Clearly defined protocols and norms that require slight adjustments at pretest (p. 2557) Adheres to ASHA guidelines and IDEA standards for CLD populations (p. 2557) “In using the PEARL among Navajo preschool children, SLPs should use a cutoff score of 7 for the pretest phase rather than 9; children who score below 7 should then receive the full DA” using a modifiability scale (p. 2557)
Horton-Ikard, R., & Weissmer, S.E. (2007). A preliminary examination of vocabulary and word learning in African American toddlers from middle and low socioeconomic status homes. American Journal of Speech Language Pathology, 16, 381-392.	To examine the effect of SES on early lexical performance in African American toddlers (30-40 months old) Toddlers from low SES performed significantly poorer than those from middle-SES homes on standardized receptive and expressive vocabulary tests and on the number of different words used in spontaneous speech No significant SES group differences were observed in their ability to learn novel word meanings on a fast-mapping tasks The influence of SES on AA children’s lexical semantic tasks varies with the type of measure used	African American children from low SES backgrounds may perform lower or higher on lexical semantic tasks depending on what type of measure is used Fast-mapping tasks may offer an alternative assessment procedure for children, particularly low SES, whose clinical status is difficult to determine (p. 390) Children with language delay or impairment would be expected to perform below average on fast mapping tasks and on other more conventional experience-dependent measures (p. 390)
Kapantzoglou, M., Restrepo, M.A., Thompson, M.S. (2012). Dynamic assessment of word learning skills: Identifying language impairment in bilingual children. Language, Speech, and Hearing Services in Schools, 43, 81-96.	15 4-5 year-predominantly Spanish-speaking children with typical language development (TLD) and 13 SLI participated in 30 to 40 min sessions of DA of word learning skills. TLD children showed ability to make associations between phonological and semantic representations of new words more quickly than SLI children, showing greater modifiability. 78.6% of participants were classified accurately, with 76.9% sensitivity and 80% specificity (p. 90). Study suggests to use with caution—only for screening Spanish-speaking children at risk for SLI as the cognitive demands of the word learning task are highly sensitive to methodological variations and may influence classification accuracy	“In the Ukrainetz et al. (2000) study, based on posttest scores, sensitivity of the expressive categorization subtests from the Assessing Semantic Skills Through Everyday Themes (ASSET; Barret, Zachman, & Huisingh, 1998) was 75% and specificity was 87%. Modifiability correctly classified the stronger and weaker language learners with 87% sensitivity and 100% specificity…” (p. 83). “High neighborhood density [words] increases cognitive demands for recognition…high phonotactic probability facilitates word learning in preschoolers…[thus] brief DA of nonword learning skills in which number of presentations, word neighborhood density, and phonotactic probability are taken into account to control for cognitive demands could give enough support to children to reveal their learning potential” (p. 85).
Peña, E.D., Gillam, R.B., & Bedore, L.M. (2014). Dynamic assessment of narrative ability in English accurately identifies language impairment in English Language Learners. Journal of Speech, Language, and Hearing Research, 57, 2208-2220.	To assess the identification accuracy of DA of narrative ability in English for ELLs 5-year old kindergarteners, 3 DA sessions over 7 to 14 day period using picture books as pre-and-post-tests with mediated learning observation rating responsivity A combination of examiner ratings of modifiability (compliance, metacognition, task orientation), DA story scores, and ungrammaticality classified children with 80.6 to 97.2% accuracy DA conducted in English provides a systematic means for measuring learning processes and learning outcomes, resulting in clinically useful procedure for identifying language impairments in bilingual children in the process of learning a second language	“The story intervention scripts incorporated key instructional elements of MLE (Lidz, 1991, 2002), such as intention to teach, meaning, transcendence, planning, and transfer” (p. 221) Intention to teach: Explain the learning goal to the child (“Today, we’re going to talk about telling complete stories”) Mediation of meaning: Demonstrated that the goal was important (“It’s important to be able to tell good stories”) Transcendence: Helped child relate the goal to everyday activities (“You read and write stories at school”) Planning: Encouraged the child to think about overall goals (“Tell me what the important parts of a story are again?”) Transfer: Encourage child to use the strategies they were taught (“How are you going to remember to include all these parts of the story?”
Peterson, D.B., Chanthongthip, H., Ukrainetz, T.A., Spencer, T.D., & Steeve, R.W. (2017). Journal of Speech, Language, and Hearing Research, 60, 983-998.	42 Spanish-English bilinguals K to 3 children were administered two 25-min DA test-teach-test sessions with narrative retells scored in real time. Examiners taught children missing story grammar elements via a structured approach and then retested the children. Tests were conducted in English. Overall modifiability rating was the best classifier, with 100% sensitivity and 88% specificity after 1 DA session and 100% specificity and sensitivity after 2 sessions.	High focus placed on efficient DA testing in two 20-25 minute sessions and making cut points for identification for placement. See next page for what they did.
Petersen, D.B., Gragg, S.L., & Spencer, T.D. (2018). Predicting reading problems 6 years into the future: Dynamic assessment reduces bias and increases classification accuracy. Language, Speech, and Hearing Services in Schools, 49, 875-888.	Longitudinal study of 370 Caucasian and Hispanic students were administered a 3-minute DA of decoding and static measures of letter identification and phonemic awareness at the beginning of kindergarten	“…a very brief dynamic assessment can predict with approximately 75%-80% accuracy, which kindergarten students will have difficulty in learning to decode up to 6 years in the future” (p. 875). Static measures overidentify whereas DA measures accurately identify children at risk of future reading issues that can be mitigated with early intervention (Ehri et al., 2001) with a short 3-min screener (p. 885)