Script-ing Teachers in the language of the T-TESS Evaluation System

• In Event: Knowledge mobilization practices and problems from the global to the local level

Proposal

Purpose
In this paper, we look to the schooling system of Texas, USA, as an illustration of how teachers and teaching are being thoroughly (re)made by data technicians, statisticians and algorithm developers. We show how the Texas Teacher Evaluation Support System (T-TESS), an ensemble of digital services, applications and analytic reporting that rely on various modes of statistical calculation, prediction and computational processing, are superseding conventional ‘knowledges’ about teacher performance and practices in Texas schools.

Theoretical Framework
In recent years, there has been a significant increase in data collection and use in education globally, reflecting the relevance of numbers (Ozga, 2016), data (Selwyn, 2015) and data visualization (Williamson, 2016b) for purposes of measuring, comparing and governing schooling performance. Accordingly, teachers now face increased pressures to employ numerical data (e.g., standardized literacy and numeracy tests), evaluative tools (e.g., observation rubrics) and prescriptive definitions of ‘what works’ to guide their pedagogical decisions and classroom practices (Bradbury & Roberts-Holmes, 2017; Hardy, 2018; Author(s) 2018). This emphasis on using (largely numerical) data to inform teacher practice and schooling policy is concurrent with moves towards what have been described as digital or algorithmic modes of governance (Williamson 2015; 2016a; 2016b), reflecting the ‘apparent power, agential capacity and control that algorithms command of our lives’ (Neyland 2015, 199).

Data and Methods
We collected and analyzed all publicly available resources associated with T-TESS (e.g., implementation guides, appraiser training materials, observation rubrics, online training videos). Using a combination of discursive strategies, such as open coding and analytic memoing (Saldaña, 2013), and informed by our analytic themes of datafication and algorithmic ideology, our analysis explicates (1) the actors and agencies comprising the T-TESS policy network; (2) the ensemble of digital services, applications and analytic reporting practices associated with the T-TESS system; and (3) how T-TESS ‘feeds back’ into Texas schools, classrooms and teachers to shape teacher subjectivities and practices.

Results
In the highly technical infrastructure of T-TESS and associated processes, teachers are seemingly ‘programmed’ with scripts developed by external technicians who have been authorized to determine 1) what counts as teacher effectiveness, 2) how it gets counted, and 3) how it should be used to shape teacher practice. It is thus the programming language that determines what can be known and understood about teachers and teaching (epistemology), as well as what teacher subjectivity can possibly be under the T-TESS regime (ontology). In this way, T-TESS constitutes what we describe as an onto-epistemic regime that defines and shapes both teacher subjectivity and practice.

Significance
We find that T-TESS, although clearly informed by data and data-driven modes of governance, is much more all-encompassing, constituting a ‘cradle to the grave’ approach that constantly modulates teacher thought and practice, thereby constraining their ability to imagine teaching and even themselves as teachers, otherwise. Operating at multiple levels, including teacher preparation programs and ongoing professional development, we argue T-TESS serves to (re)define every aspect of teacher being and practice in Texas, which presents significant challenges for teacher professionalism and pluralism.

Authors

• Jessica Holloway, Deakin University

• Steven Lewis, Research for Educational Impact (REDI) Centre, Deakin University