Interlanguage-stretching activities within a task-based empirical pedagogy Tom Means, [email protected] Marlboro College, Vermont, USA Larry Selinker, [email protected] New York University, NYC Researchproductionassociates.com Definition of Interlanguage (IL)stretching in this study • IL-stretching of Accuracy • IL-stretching of Delivery (Fluency) Definition of Empirical Pedagogy • Following Han and Selinker (1999) • Input-Heavy TBI (Means, 2006) • Underpinned by Cognitive Approach (Skehan, 1998) Input-heavy TBI (of Italian as a FL in U.S.A) (Needs Analysis) • Pre-task – Input floods – Task-relevant production activities • During-task – Time pressure – Recording • Post-task - Self-transcript - Grammar activities Input-Heavy TBI as Empirical Pedagogy • Research base • Institutional implementation – Large US university (3rd year) – Small US university (2nd year) • TBI as a vehicle for: – Copious input, interaction, output, feedback The Study • Central Hypothesis of the study: TBI will better promote accuracy and fluency than TI of Italian as a FL. The Study • Ecologically valid – Representative – Intended audience: researchers, teachers, administrators, students • Fine-grained • 9-week method comparison study involving TBI and TI of Italian as a FL The Study: Who/When/Where • Participants • Research Setting(s) • Subject recruitment and selection • Treatment calendar The Study: Treatment of TBI Group • Research-based – Primacy of input – Psycholinguistic merits of output • Three-phase cycle • Types of tasks The Study: Treatment of TI Group • Empirical base • 4-phase cycle How were the independent variables manipulated consistently? • Methodologies of language instruction – Task-based Instruction (TBI) • Unit of analysis = 3 task cycles – Traditional Instruction (TI) • Unit of analysis = 3 textbook chapters Dependent variables • Accuracy – Grammatical gender agreement in the noun phrase • Determiner-noun agreement – un fratello; *uno fratello – Il mese; *i mese • Noun-adjective agreement – piatto giallo; *giallo piatto – Il mese scorso; *la mesa scorsa • Fluency – Oral fluency (spoken syllables per minute) – Written fluency (written words per minute) Measurements • Battery of 4 tests – Controlled-production tests • Test 1: Create Complete Sentences • Test 2: Narrative Rewriting – Spontaneous-production tests • Test 3: Narrative Retelling • Test 4: Family Tree • Data collection Results of statistical analyses Accuracy Results – In controlled tests • Marginal advantage for TBI group – In spontaneous tests • Marginal advantage for TBI group Accuracy results Test 4: Mean Scores of D-N GG Agreement Accuracy in Family Tree Test 100 80 60 40 20 0 Pretest Posttest TBI Delayed PT TI No statistical significant differences. Fluency Results Oral Fluency • Significant advantage for TBI group Written Fluency • Significant advantage for TBI group (at posttest only) Oral fluency gains Test 4: Family Tree Mean Scores for Oral Fluency 120.0 100.0 80.0 60.0 40.0 20.0 0.0 Pretest Posttest TBI Delayed PT TI Statistically significant difference for TBI at both times; none for TI. Written fluency gains 100 90 80 70 60 TBI TI 50 40 30 20 10 0 Pretest Posttest Delayed Statistical significance for TBI group at both points. For TI, only at delayed Examples from the data • Analysis of accuracy from Test #1 • TI subject # 5080 Pretest la forma Posttest la forma Delayed Posttest *forma • TBI subject #3262 Pretest *il spettacolo Posttest *il spettacolo Delayed Posttest lo spettacolo Examples from the data • Analysis of oral fluency from Test #3 • TBI subject # 6211 – Pretest performance: 58 syllables p/m – Delayed posttest performance: 107 • TI subject # 5080 – Pretest performance: 80 syllables p/m – Delayed posttest performance: 82 Discussion • Central Hypothesis of the study: TBI will better promote accuracy and fluency than TI of Italian as a FL. • Was the central Hypothesis retained? – Fluency mode – Accuracy mode Why was IL better delivered by Input-Heavy TBI? • Teacher is decentralized • Collaborative, structured input • Meets head-on the challenge of producing spontaneous oral language Why was IL better promoted with Input-Heavy TBI? (at a marginal level) • Language as vehicle and object • Engagement of rule-based mode • Problems with the language feature in question, grammatical gender (Swain, 2005) Implications • For language instruction in foreign language settings – a “usable” pedagogy that is empirical • For research in Second Language Acquisition Conclusions • Input-Heavy TBI as an Empirical Pedagogy – Empirical method comparison study • Analyzing fluency and accuracy improvement (IL-stretching) – Contributes some evidence of TBI’s effectiveness – Creation of replicable TBI model, “Input-Heavy” that is flexible and adaptable – Initiation of a TBI line of study that is fundamentally an educationally-relevant endeavor Bibliography • Han, Z and Selinker, L. 1999. ‘Error Resistance: Towards an Empirical Pedagogy.’ Language Teaching Research, 3.3, 248-275. • Means, T. 2006. A Comparative Study of Task-based and Traditional Instruction of Intermediate Italian: Findings on Accuracy and Fluency. Unpublished Ph.D. dissertation, Rutgers University, New Jersey • Skehan, P. 1998. A Cognitive Approach to Language Learning. Oxford: OUP • Swain, M. 2005. ‘The Output Hypothesis: Theory and Research.’ in E. Hinkel (ed.), Handbook of Research in Second Language Teaching and Learning. Mahway, New Jersey: Lawrence Erlbaum.