Florida State University Libraries

Follow this and additional works at the FSU Digital Library. For more information, please contact lib-ir@fsu.edu

2016

Words Per Minute: Narrative LanguageAssessment of English Language LearnersMae Espinosa, Eric Luberto, Claire Wofford and Carla Wood

Words Per Minute: Narrative Language Assessment of English Language Learners

Mae Espinosa, Eric Luberto, Claire Wofford, and Professor Carla Wood

Florida State University ޮ School of Communication Science and Disorders

English language-learners (ELLs) are a rapidly growing populationfacing many challenges in education, due to difficulty with comprehension andexpression (Kieffer, 2008; Hammer, Hoff, Uchikoshi, Gillanders, Castro, &Sandilos, 2014). Thus, research in how ELLs communicate is imperative inorder to aid in their learning. This project seeks to understand how ELLsprocess information in the language they are learning and how they can sharethat same information; its purpose is to describe words per minute (WPM) andnumber of different words (NDW) performance of young Spanish-Englishbilingual ELLs of low socioeconomic background .

Language sample analysis (LSA) provides a less biased assessment thanstandardized testing for children from linguistically diverse backgrounds(Miller & Iglesias, 2010). Kindergarten and first-grade children performed anarrative retell of a short cartoon with little to no verbal cues from examiners.Audio samples were transcribed and analyzed using the Systematic Analysis ofLanguage Transcripts (SALT) software in order to obtain WPM for eachtranscript. These data will be used to describe the typical language productivityand lexical diversity used by Spanish-English ELLs on a narrative retell taskand to observe the relationship between these LSA measures during a narrativeretell and children’s performance on standardized language assessments.

The first step was to collect the audio samples from the children. This wasdone at various elementary schools around the state of Florida. Childrenwatched a short cartoon using headphones and were then asked to retell themovie to a stuffed animal. The stuffed animal was used to make children feelmore comfortable while retelling the movie, and thus elicit a longer narrative.

Children received minimal verbal cues from examiners, unless theexaminer deemed prompting was necessary in order to elicit a narrative fromthe child. Examiners recorded children’s narrative retells. These audio sampleswere then transcribed, using C-units to divide examiner and child utterances. AC-unit is defined as an independent clause and its modifiers. The audio sampleswere also coded as they were transcribed, in order to signal pauses,interruptions for either the examiner or child, or abandoned utterances. Oncethe audio samples were transcribed, the transcripts were entered into theSALT software in order to generate the WPM for each child.

This project is still a work in progress. Currently, we are working on generating the quantitative data using the SALT software. Once we compile the WPM for all the transcripts, we will be able to compare these data to results from three other tests:! Peabody Picture Vocabulary Test (PPVT)

! Measures receptive, or hearing, vocabulary of children and their achievement in acquiring vocabulary.

! Children are shown pictures and are asked to point to the one that corresponds to a certain description.

! Bilingual English Spanish Assessment (BESA)! Designed as a comprehensive language assessment that evaluates

overall language skills for children who are bilingual in English and Spanish

! Contains three subtests. ! Woodcock Reading Mastery Test, Third Edition (WRMT-III)

! Comprehensive set of nine tests that evaluate reading readiness and reading achievement.

Comparing the WPM from the audio transcripts to the results of these tests will reveal more information about how ELLs communicate and process language. We expect the results from all tests to correlate: children that have low WPM will likely score poorer on the PPVT, BESA, and WRMT-III, whereas children that have high WPM will likely score higher in the other tests. This may be due to different levels of fluency in English.

Thuy-Linh PhamShauna Gillooly

Darrell KilpatrickMitchell D'sa

Allison Richard Amanda Yoss

References

The method chosen and used to collect language samples during thisproject was having the child watch a movie and tell a narrative about it back tothe recorder with minimal, if any, prompts. Many researchers regardconversation as the most natural and representative context in which to collecta sample and has been found to elicit more complex language than that elicitedfrom a picture description task (Longhurst & File, 1977; Kemp & Klee, 1997).

Once the samples have been collected, they proceed to be transcribed inorder to evaluate valuable data. Number of different words (NDW) and wordsper minute (WPM) have been traditionally used to evaluate language samplesand both have become well-respected measures in recent years. NDW isreportedly the most commonly used language sample measures to assess lexicaldiversity (Price & Jackson, 2015) and to monitor oral language proficiency indual language learners (Jacobson & Walden, 2013). In recent years the numberof words per minute (WPM), a fluency measure, has increased in recognitionfor Spanish-English language sample analysis (Heilmann, Nockerts, & Miller,2010; Miller, Andriacchi, & Nockerts, 2015; Price, Hendricks, & Cook, 2010;Tilstra & McMaster, 2007). For young English language learners, WPM isthought to represent overall integration of the multiple demands of formulationand production (Rojas & Iglesias, 2013); and demonstrates a positivecorrelation with age of ELLs (Miller & Heilmann, 2004).

In the future, more research will be needed to better inspect anddetermine the relationship between WPM and performance on standardizedlanguage and literacy tests.

Abstract

Background

Methods Discussion

Acknowledgments

Heilmann, John, Ann Nockerts, and Jon F. Miller. “Language Sampling: Does the Length of the Transcript Matter?” Language, Speech, and Services in Schools. 41. (2010): 393-404.Heilmann, John J. “Myths and Realities of Language Sample Analysis.” Perspectives on Language Learning and Education. (2010): 4-8.Heilmann, John J., Jon F. Miller, and Ann Nockerts. “Using Language Sample Databases.” Language, Speech, and Services in Schools. 41. (2010): 84-95.Price, Lisa Hammett, Sean Hendricks, and Colleen Cook. “Incorporating Computer-Aided Language Sample Analysis Into Clinical Practice.” Language, Speech, and Services in Schools. 41. (2010): 206-222.Guo, Ling-Yu, and Sarita Eisenberg. “Sample Length Affects the Reliability of Language Sample Measures in 3-Year-Olds: Evidence From Parent-Elicited Conversational Samples.” Language, Speech, and Services in Schools. 46 (2015): 141-153.McMaster, Kristen, and Janet Tilstra. "Productivity, fluency, and grammaticality measures from narratives: potential indicators of language proficiency?" Communication Disorders Quarterly. 29.1 (2007): 43+. Expanded Academic ASAP. Miller, John F., John Heilmann, Ann Nockerts, Aquiles Iglesias, Leah Fabiano, and David J. Francis. “Oral Language and Reading in Bilingual Children.” Learning Disabilities Research & Practice. 21.1 (2006): 30–43.Allen, Melissa M., Teresa A. Ukrainetz,a and Alisa L. Carswell. “The Narrative Language Performance of Three Types of At-Risk First-Grade Readers.” Language, Speech, and Services in Schools. 43 (2012): 205-221.Sturm, Jennifer A., Carol H. Seery. “Speech and Articulatory Rates of School-Age Children in Conversation and Narrative Contexts.” Language, Speech, and Services in Schools. 38 (2007): 47-59