Click Here To Download the PDF Version of This TEA Announcement- Jan 2014
The development of the State of Texas Assessments of Academic Readiness (STAAR®) English I and English II reading and writing assessments included extensive public scrutiny and input from Texas teachers, administrators, curriculum staff, faculty and staff at Texas colleges and universities, education service center personnel, and national content-area experts. Educators recommended which Texas Essential Knowledge and Skills (TEKS) should be eligible for assessment and which TEKS should be considered readiness as opposed to supporting standards. Educators also provided valuable input on the reporting categories, the test design and blueprint, the prototype questions, the reading and writing rubrics, and the reading and writing selections and questions developed each year.
To reflect the intent of House Bill (HB) 5, the STAAR English I and II assessments have been redesigned to combine reading and writing into a single measure with a single test score, but the goal has been to maintain the fidelity of the original design since it was developed in close collaboration with Texas educators. To accomplish this, Texas Education Agency (TEA) staff made a proportional reduction of the blueprint across the six reporting categories to ensure that the reading and writing strands were equally represented, that no reporting category was unduly impacted (i.e., either overrepresented or underrepresented on the test), and that the combined assessments could be administered in spring 2014, as required by HB 5. No changes were made to the TEKS eligible for assessment, the readiness and supporting standards, the reporting categories, the rubrics, or the types of selections and questions included on the assessment. Both the rigor of the assessments and the level of performance required for student success remain the same.
5-hour time limit in relation to the blueprint reduction
In 2012 and 2013, there was a 4-hour time limit for reading and a 4-hour time limit for writing. The assessments were given on two different days. In determining an appropriate time limit for the redesigned English assessments, TEA staff reviewed two pieces of data related to the testing time needed to complete the separate English I and English II reading and writing assessments.
In 2012, districts that were representative of the state completed a survey related to the time needed for students to complete the English I assessments on paper. These data represented approximately 5% of the students who tested. This survey showed that 99.7% of students completed the English I reading test within the 4-hour time limit, with 77% of students needing 3 hours or less. For English I writing in 2012, 98.7% of students completed the English I writing test within 4 hours, with 59% completing the test within 3 hours.
In spring 2013, approximately 10% of students who took the English I and II reading and writing assessments tested online, and TEA received data for these students related to testing time. On average, students needed approximately 2 hours to complete the English I and II reading tests and 2 hours and 45 minutes to complete the writing tests.
The length of the English I and II assessments has been reduced by eliminating the following:
an entire reading selection and the questions associated with it
an entire writing selection and the questions associated with it
one essay (the literary essay for English I and the expository essay for English II)
19 multiple-choice questions (across reading and writing)
The new blueprint for English I and II reflects a reduction of 50% of the essays and 26.5% of the multiple-choice questions that count toward a student’s score. When combined with the reduction in multiple-choice questions, reading two fewer selections and writing one less essay significantly decreases the testing burden on students and the time they need to complete the test. The reason for this is that the amount of actual reading and writing required on the assessments greatly affects testing time because these tasks represent the most time-consuming parts of the test.
In 2012 and 2013, the English I reading and writing assessments took the average student 43⁄4 hours to complete, and the English II reading and writing assessments took the average student slightly less than 5 hours to complete. With the reductions listed above, TEA staff estimate that most students will take approximately 31⁄2 hours to complete the combined English tests, less than the testing time needed when the assessments were given as separate tests on separate days. It is anticipated that students who take the combined tests will need approximately 11⁄2 hours for the reading section (1 hour to read the selections and answer the multiple-choice questions and 30 minutes to respond to the short answer questions) and 2 hours for the writing section (11⁄2 hours to write the essays and 30 minutes to answer the multiple-choice questions). However, to address the needs of students who might work more slowly than the average general education student, TEA has established a 5-hour time limit, although only a small number of students are likely to require this amount of time to complete the test.
For students served by special education, 504 students (e.g., those who have been identified with dyslexia), and English language learners, TEA will continue to allow extended time beyond the established time limit. Students who have multiple or severe disabilities may be allowed to complete the test over two days (which was done for the TAKS grade 10 and exit level ELA assessments) if their district submits an accommodation request form that is approved.
The number of field-test questions
Two needs have to be balanced when field testing is conducted: minimizing burden on students and districts and administering tests that meet recommended industry standards. Practically, it is important to minimize stand-alone field testing so that time can be spent on instruction. Statistically, it is imperative that Texas collect enough field-test data from a representative sample of students to have reliable information to evaluate the quality, difficulty, and fairness of questions. Embedding the field test in the actual assessment provides the most accurate data. The reason for this is that these data reflect students’ authentic test performance, as students are unable to distinguish the field-test questions from the actual questions. Having accurate performance data is critical to building high-quality tests and ensuring that these tests are legally defensible as graduation requirements.
Thirteen field-test multiple-choice questions (7 in reading and 6 in writing), one field-test short answer question, and one field-test prompt are embedded in the English I and English II spring tests, but they are not part of the calculation of a student’s score. This number is necessary due to the legislative requirement to release the spring form of STAAR in 2013, 2014, 2015, and 2016. In addition, three tests must be built each year since there are three administrations (spring, summer, and fall). This requires annual development and field testing of a significant number of questions. However, field-test selections, questions, and prompts are embedded only on the spring tests; no field testing is done during the summer and fall administrations of English I and II.
New Weighting of Components for the Redesigned English Assessments
The English I and English II tests have been redesigned so that writing and reading each contribute equally (50%) to the total test score. Each multiple-choice question, whether it assesses reading or writing, counts one point toward the score—a total of 50 points (of 92 points). The two short answer reading responses and the essay are each weighted by three so that they contribute 18 points (20% of total score) and 24 points (26% of total score) respectively. While more than half the total test score is based on multiple-choice questions, assessment of reading and writing skills in the context of actual performance tasks is also considered extremely important at the high school level.
The STAAR English I and II rubrics for expository and persuasive writing require students to write well-crafted essays to achieve high scores. Scoring criteria for the essays include a strong match between structure/form and purpose, the inclusion of an explicit thesis or position statement, sustained focus, a logical progression of sentences and ideas, specific development that adds substance to the essay, specific use of language and appropriate tone for the purpose, and strong conventions. However, the short answer reading section of each assessment is not scored on these same expectations for writing. On the short answer section of the English I and II tests, students are required to answer two questions: one on a single reading selection and one on a thematically linked pair of selections. For each question, students must generate a reasonable idea and confirm the validity of that idea by using specific evidence from the text. Students are scored on the content of their answers and the text evidence they use, not on the quality of their writing. This section of the reading test is actually assessing the degree to which students can critically read on-grade-level texts.
Because only one score is reported for each assessment, performance in reading can compensate to a small degree for performance in writing or vice versa. In addition, although the test blueprint for each assessment has been reduced, the assessments continue to support broad coverage of and alignment to the TEKS curriculum, and the length of the tests remains psychometrically sufficient to maintain their validity and reliability as a graduation requirement.