èÖľyuzuki

The State Education Department today released the results of the math and English Language Arts (ELA) exams taken by all èÖľyuzuki students in grades 3-8 in May 2011. The data show that the average scale scores on this year’s English exams are slightly lower than last year’s at all grade levels; the average scale scores in math are about the same as last year’s. While the majority of students statewide met or exceeded the state’s proficiency standards in both math and ELA, overall performance remains low and the gaps in achievement persist.Ìı Actions to implement the Regents education reform agenda are underway, with the related goals of closing the gap and preparing all students for college and careers.

Last year, the state raised the “proficiency� standard scores to better reflect the level of achievement needed to indicate that a student is on track to achieve college-ready scores on future state exams (80 or above on a Math Regents and a 75 or above on the ELA Regents exam). The Department also made changes to this year’s exams to make them more comprehensive and better measures of students’ skills.

Regents Chancellor Merryl H. Tisch said, "These results underscore the urgent need for èÖľyuzuki to continue to aggressively move forward with the implementation of the Regents' reform agenda. ÌıThrough aggressive implementation at the district and school level of higher standards, better and more accurate assessments, a more content rich curriculum and a teacher evaluation system aimed at supporting teaching excellence, we can make tremendous strides towards ensuring all of our children succeed."

Education Commissioner John B. King, Jr. said, "Student outcomes have been stubbornly flat over time. The Regents reform agenda is designed to change that, by driving long-term gains in student performance. Better tests are only one part of the reform strategy. We’re also moving forward in our efforts to ensure better training and better support for the teachers and principals in our schools; to provide more transparent and useful data; and to help our lowest performing schools take the necessary steps to turn around their performance or replace them with innovative alternatives. Taken together, these efforts will dramatically improve the likelihood that èÖľyuzuki’s students are well-prepared for college and careers."

Summary of 3-8 Exam Results

• 52.8% of grade 3-8 students across the state met or exceeded the ELA proficiency standard (a decrease from 53.2% last year); 63.3% met or exceeded the standard in math (up from 61% last year).
• 12.6% of English Language Learners (ELLs) statewide met or exceeded the new ELA proficiency standard (down from 14.3% last year) across grades 3-8; 32.3% of ELLs met or exceeded the standard in math (up from 30.7% last year).
• Statewide results for black students reveal the persistence of the achievement gap: 35% of black students across grades 3-8 met or exceeded the ELA proficiency standard (compared with 52.8% for all students and 64.2% for white students); 44% met or exceeded the standard in math (compared with 63.3% for all students and 73.3% for white students).
• Statewide results for Hispanic students reveal the persistence of the achievement gap: 37.2% of Hispanic students across grades 3-8 met or exceeded the ELA proficiency standard (compared with 52.8% for all students and 64.2% for white students); 50.2% met or exceeded the standard in math (compared with 63.3% for all students and 73.3% for white students).
• 14.5% of Students with Disabilities (SWDs) statewide met or exceeded the new ELA proficiency standard (down from 15.2% last year) across grades 3-8; 26.9% of SWDs met or exceeded the standard in math (up from 24.6% last year).
• The percentage of students scoring at Level 4 in both ELA and math decreased statewide. On the ELA exam, 3.5% of students across grades 3-8 combined scored at Level 4 (down from 10.2% last year). In math, 23% scored at Level 4 (down from 24.7% last year).
• Across the Big 5 city school districts, a smaller proportion of students met or exceeded the math and ELA proficiency standards than in the rest of the state:

Ìı

Ìı

• • In Buffalo, 26.9% of students in grades 3-8 met or exceeded the English proficiency standard (down from 27.7% last year); 31% met or exceeded the math standard (up from 29.8% last year).
  • In èÖľyuzuki City, 43.9% of students in grades 3-8 met or exceeded the English proficiency standard (up from 42.4% last year); 57.3% met or exceeded the math standard (up from 54% last year).
  • In Rochester, 24.4% of students in grades 3-8 met or exceeded the English proficiency standard (down from 25.3% last year); 29.4% met or exceeded the math standard (up from 28% last year).
  • In Syracuse, 22.5 of students in grades 3-8 met or exceeded the English proficiency standard (down from 25.5% last year); 25.3% met or exceeded the math standard (down slightly from 25.7% last year).
  • In Yonkers, 37.8% of students in grades 3-8 met or exceeded the English proficiency standard (down from 39.2% last year); 40.4% met or exceeded the math standard (down from 41.5% last year).

Ìı

A Powerpoint presentation summarizing the test results, as well as individual school and district results, is available at .

Background on Cut Scores and Changes to Exams

Recognizing that remediation rates in èÖľyuzuki’s colleges are far too high, and that much more needs to be done to enhance the State’s economic competitiveness, the Board of Regents has begun an in depth discussion of the knowledge, skills, and abilities high school graduates must demonstrate in order to show they are preparedÌıÌı for college and careers.

To inform that discussion, the Regents have been working to determine which data are good indicators of future post-secondary success. Last summer, for example, they reviewed research that analyzed how student performance on the state’s grade 3-8 math and English tests relates to performance on theÌıNAEP exam and on Regents exams, and how performance on Regents exams relates to first year performance in college. The research indicated that some students who had scored “proficientâ€� on state exams found themselves unprepared, without remediation, to do the work required of them when they reached college. Accordingly, the Regents and the Commissioner adjusted the cut scores on the grade 3-8 exams to more accurately indicate proficiency.Ìı

The Department also made a number of significant changes to the math and EnglishÌıexams that students took this past May. These changes, which demonstrate the Department’s commitment to the continuous development of a more robust testing system, include the following:

• The Department is no longer releasing test questions after they have been administered.ÌıDoing thisÌıhelps to ensure that preparation for the tests goes much deeper than simply reviewingÌıpast exams.
• StudentsÌınow have to answer more multiple choice questions on bothÌıthe math and English exams. Doing so enables the Department to test a greater range of performance indicators at varying degrees of difficulty. With a greater number of questions to answer, the length of time each test takes to completeÌıhas increased.
• Students at every grade level must now write at least one full essay on the English exams (i.e., essays were added to the exams in grades 3, 5, and 7).
• At the Department’s request, Daniel Koretz, Henry Lee Shattuck Professor at the Harvard Graduate School of Education, is conducting research to help identify the possibility of “score inflationâ€� in the grade 3-8 test results and to propose actions to mitigate such inflation if it is found to exist.

Changes in the testing system have had several effects on the reported scores. Each year, scores are statistically linked to the scores from the year before, so that any given score, or any performance level, will have the same meaning from one year to the next. For example, a student at the cut score for Level 3 in grade 5 math in 2011 should ideally score 674, which was the cut score in 2010. However, the difficulty of individual test items can vary, in either direction, from one year to the next. Accordingly, the raw score needed to reach that scale score or performance level can change. This happens with all testing programs that change test forms from one year to the next.

The changes made in 2011 had an additional effect on reported scores.Ìı In every grade, the percentage of students classified as Level 4 in ELA dropped in 2011, and in some grades, particularly 3, 5, and 7, this change was large. The reason for these changes is that before 2011, the tests had too few items to measure the Level 4 cut score as accurately as the Department would want. With the additional items included in 2011, the tests could more accurately place the Level 4 cut score. Measurement of the Level 3 cut score was more accurate in past years and, therefore, the percentages of students identified as proficient (Level 3 or 4) in 2011 were more consistent with past results.