edTPA Summary Report Now Available

The Stanford Center for Assessment, Learning, and Equity (SCALE) has announced the public release of the edTPA Summary Report for the 2012-13 field test. The report describes edTPA design, development efforts by the profession for the profession, data informing validity and reliability studies, and field test and standard-setting results.

edTPA is the first nationally available, educator-designed performance assessment for teachers entering the profession. It provides a measure of teacher candidates’ readiness to teach that can inform program completion, licensure, and accreditation decisions, while supporting candidate learning and preparation program improvement.

Subject-Specific and Performance-Based:

edTPA is a subject-specific assessment, which includes versions for 27 different teaching fields. The assessment systematically examines an authentic cycle of teaching aimed at specific learning goals, using evidence about 1) planning, 2) instruction, and 3) student assessment derived from candidates’ practice in their student teaching or internship placement. This evidence includes lesson plans, instructional materials, student assignments and assessments, feedback on student work, and unedited video recordings of instruction. Also assessed through these three tasks are candidates’ abilities to develop academic language and to analyze teaching.

Aligned to Standards:

edTPA is aligned with the newly revised Interstate Teacher Assessment and Support Consortium (InTASC) standards for beginning teacher licensing, as well as the Common Core State Standards. edTPA also shares key points of alignment with the Council for Accreditation of Educator Preparation (CAEP) standards.

Developed By the Profession:

Modeled after the National Board for Professional Teaching Standards’ assessments of accomplished veteran teachers, edTPA was created with input from teachers and teacher educators across the country in a process led by Stanford University’s Center for Assessment, Learning, and Equity (SCALE) and supported by the American Association of Colleges for Teacher Education. More than 1,000 educators from 29 states and the District of Columbia and more than 430 institutions of higher education (IHEs) participated in the design, development, piloting, and field testing of edTPA.

Scored by the Profession:

An additional 650 teachers and teacher educators have served as scorers of the assessments for the 12,000 plus candidates who participated in the field tests. Scorers must be P-12 or higher education educators with significant pedagogical content knowledge in the field in which they score, as well as experience in working as instructors or mentors for novice teachers. In the field test, about half of the scorers were teacher educators and about half were practicing classroom teachers, many of whom are National Board certified.

Scorer training and certification is a rigorous process that takes about 20 hours to complete. Scoring is conducted using 15 analytic rubrics applied to the three major tasks in the assessment, each evaluated on a five-point scale. Thus total scores on the assessment can range from 15-to-75. The scoring process and reporting is coordinated by Evaluation Systems, a division of Pearson, which distributes the assessments and manages the electronic platform on which they sit.

Score Distributions:

This report analyzes more than 4,000 submissions from the 2013 field test by content field, area of teaching, and candidate group. Score differentials across fields were relatively small. The scores were generally higher in secondary teaching fields than in most elementary and middle childhood fields. Candidates varied in their mastery of different teaching skills. As might be expected of pre-service teachers, candidates performed most highly on the planning task, followed by the instruction task, and then the assessment task. This conforms with other studies that have found that learning to evaluate and respond to students’ learning is one of the more challenging elements of teaching, one that eludes even some veteran teachers.

No noticeable differences were found across different genders or for the small number of candidates whose primary language was not English. Comparatively small differences were obtained across ethnic subgroups, given the standard deviation of mean scores. In other words, there was greater variation in scores within subgroups than there was variation across subgroups. Because many of these comparisons are based on unequal sample sizes, and some subgroup sample sizes are quite small, differences should be interpreted cautiously.

Standard-Setting:

Standard-setting was conducted in August 2013 through a data-informed, standards-based consensus process featuring three separate panels of experts. These included teachers and teacher educators with relevant subject matter expertise, education organization representatives, and state education policy makers from across the nation. The three panels independently agreed on a score target of no more than 42 to which developers applied lower bounds based on the standard error of measurement in order to minimize erroneous decisions. Using a half standard error of measurement results in a cut-score band ranging from a total score of 39-to-42.

States set their own passing scores based on state standard-setting that takes into account state-specific data, measurement data, and policy considerations. As discussed by the standard-setting committee members, states may consider setting their initial cut score lower than the committee-suggested cut score to give programs time as they learn to deliver and support edTPA activities and to support candidates’ preparation of their submissions. As warranted, the cut score can be raised over time.

Based on the 2013 field test data, at a cut score of 42, 58% of candidates would have “passed” edTPA on their first attempt. Field test data, of course, do not fully reflect operational results for several reasons:

1) Candidates are a voluntary sample of prospective teachers who take the assessment in a non-consequential context;

2) Many programs using the assessment for the first time have little experience designing support systems, coursework or clinical experiences to enable students to demonstrate the targeted skills. Under actual testing circumstances with programs that have become experienced in supporting the assessment, pass rates would be expected to increase.

In the operational phase, all scores on or near the cut score will be read and scored by two independent certified scorers. If there are differences in the scores, a third reader independently scores the submission to adjudicate discrepancies in scores. Candidates who do not pass on their initial attempt may retake one edTPA task or the entire edTPA assessment, with advice and counsel of faculty from their teacher preparation program.