Data-Driven Instruction Improves Spoken Language Skills
Michael Douglas · The Hearing Journal · 2016
Figure 1: Data Driven Instruction ModelData Collection: Teachers gathered spontaneous language samples. Documentation: Results were transcribed onto a criterion-reference form: The Teacher Assessment of Spoken Language (TASL). Analysis: Skills not demonstrated were systematically targeted for instruction. Reflection and Planning: Experienced classroom coaches worked with teachers to analyze data and develop activities. Implementation: Lessons were conducted under the guidance of instructional coaches. (Figures from Douglas. Otol Neurotol 2016;37[2]:e13 http://ow.ly/rsGm300QSYv)Figure: Michael Douglas, MAAlthough modern day hearing technology is continuously improving, nothing exists that provides a signal complete enough for most children with severe and profound hearing loss to learn spoken language without specific instruction (Ambrose. J Speech Lang Hear Res 2012;55[3]:811 http://ow.ly/R8SU300QMa7). Typical cochlear implants provide envelope-based signals for the spectral range spanning approximately 100 to 8000 Hz; whereas, the young, normal-hearing cochlea processes fine spectral and temporal detail of incoming sound in the range of 20 to 20,000 Hz. Despite this difference in fine spectrotemporal processing and absolute frequency range, some children with significant hearing loss who receive cochlear implants have demonstrated the ability to acquire age-appropriate spoken language skills (e.g., Niparko. JAMA 2010:21;303[15]:1498 http://ow.ly/vRjJ300QKVx; Tobey. Int J Audiol 2013;52[4]:219 http://ow.ly/a11T300QLkI). The majority of these children, however, require specialized and individualized teaching by qualified professionals (Geers. J Deaf Stud Deaf Educ 2009;14[3]:371 http://ow.ly/v7lL300QN52). EFFECTIVE INTERVENTIONS In general, it has been the consensus for decades that the most effective interventions for students with disabilities—whether in special education or general education settings—have employed intensive and reasonably individualized instruction, combined with careful, frequent monitoring of student progress (Hocutt. Future Child 1996 Spring;6[1]:77 http://ow.ly/rcnE300QNoP). Such data-driven instruction (DDI) ensures that teachers and/or clinicians are aware of students’ ongoing progress and accelerates learning to levels comparable to children with typical hearing (Bambrick-Santoyo. Leverage leadership. San Francisco: Jossey-Bass, 2012; Dornan. Volta Rev 2010;110[3]:361 http://ow.ly/P5iq300QPtB; McLead, 2014; Moog. Otol Neurotol 2010;31[8]:1315 http://ow.ly/THQz300QX3L). Acceptable achievement relies on teachers’ and/or clinicians’ accuracy in diagnosing student skills and their ability to: Prescribe appropriate teaching activities, Structure the lesson to ensure success, Give clear direction for what is expected, and Provide feedback on the child's responses (Moog. CICSD 2008;35:133-42). A NEW LOOK AT DATA-DRIVEN INSTRUCTIONFigure 2: Data-Driven Instruction Group ResultsBoth mean raw score of TASL results and mean standard scores of the speech and language test battery for the data-driven instruction group are plotted for baseline, year one, and year two.Until recently, there has been no mention in the literature on the effects of DDI for children with hearing loss compared to those who do not receive such instruction in spoken language programs. In a two-year retrospective study, Douglas (Otol Neurotol 2016;37[2]:e13 http://ow.ly/rsGm300QSYv) compared 11 matched pairs of children with cochlear implants who attended the same spoken language preschool. Groups were matched with no statistically significant differences for age of hearing device fitting, time in the program, degree of pre-device fitting hearing loss, gender, or age at testing. Daily informal language samples were collected and analyzed over a two-year period, per preschool protocol. Annual informal and formal spoken language assessments in articulation, vocabulary and omnibus language were administered at the end of three time intervals: baseline, end of year one, and end of year two. Utilizing a partnership approach, classroom coaches worked with the deaf education teachers who participated in the study for two years to analyze data, develop activities, and implement lesson plans that would help the child meet the selected spoken language objective(s). The role of the instructional coaches was to mitigate challenges associated with fidelity of instruction and treatment integrity (e.g., helping the teacher document appropriate language samples, assuring the development of activities that obligated the use of the targeted language, and reinforcing the use of modeling and imitation techniques that improved, corrected and/or extended the children's spontaneous utterances). Teachers also had the opportunity to discuss student learning and design intervention on a regular basis in their shared office spaces and during bi-monthly staff meetings scheduled throughout each school year.Figure 3: Treatment and Control Group ComparisonMean standard score results for each speech and language test administered (articulation, total language, expressive vocabulary and receptive vocabulary) is plotted for the control group and the data driven instruction group.The primary outcome measures were total raw score performance of spontaneous utterance sentence types and syntax element use as measured by the Teacher Assessment of Spoken Language (TASL). In addition, standardized assessments (the Clinical Evaluation of Language Fundamentals – Preschool Version 2 [CELF-P2], the Expressive One-Word Picture Vocabulary Test [EOWPVT], the Receptive One-Word Picture Vocabulary Test [ROWPVT] and the Goldman-Fristoe Test of Articulation 2 [GFTA2]) were also administered and compared with the control group. Results indicated that the data-driven instruction group demonstrated significantly higher raw scores on the TASL each year of the study and significantly higher scores for total language on the CELF-P and expressive vocabulary on the EOWPVT compared to the control group, but not for articulation or receptive vocabulary. Post-hoc assessment revealed that 78 percent of the students in the DDI group achieved scores in the average range compared to 59 percent in the control group. IMPROVED OUTCOMES MANAGEMENT DDI is a system of teaching and outcomes management that strives to provide more detailed information about students into the hands of classroom teachers and clinicians. Specifically, the data provided can help drive individualized instruction systematically in a language-rich environment as well as guide educational decisions. In doing so, listening and spoken language programs could potentially increase the trajectory of language and vocabulary development, allowing children with hearing loss to close the language gap that continues to exist at the group level between children with hearing loss and their normal-hearing peers (Niparko; Tobey; Yoshinago-Itano. Otol Neurotol 2010;31[8]:1268 http://ow.ly/5JeB300QRUY; Geers. J Speech Lang Hear Res 2013;56[2]:643 http://ow.ly/RquM300QSoK). This type of instruction may not only be effective, but also an important factor to the improved spoken language success of children with cochlear implants.