É«ÖÐÉ«

Test scoring is a complex process that derives student scores from the number of questions answered correctly on a test, the level of difficulty of the questions and the skills each question measures.Ìý The final score on most Regents examinations is not a simple percentage or number of correct answers.Ìý Nor is it the same as the raw score – the total number of points a student achieves on a test.

The É«ÖÐÉ« develops Regents examinations in accordance with the procedures for test development that are recognized by the American Education Research Association, the American Psychological Association, and the National Council on Measurement in Testing.Ìý All major test developers, including SAT, NAEP, and AP, use these procedures and insist that all tests must be scaled as described above.

Determining What a Test Will Measure

Before a test is created, teachers and other educators develop a "test map" of the key concepts to be tested -- what students need to know and what skills they need to have in order to pass the test. These skills are based on the Regents standards of what students need to know and be able to do at each grade level.Ìý

Creating Test Questions

Questions for each test are based on the ‘test map’ and measure how well students taking the test can perform these skills.Ìý Some questions are written to be more difficult than others, based on what skills they require students to use.Ìý All questions on the Regents examinations are written by É«ÖÐÉ« teachers under the direction of the Department.

Determining Passing Score

Before a test is administered statewide for the first time, questions written for the test go through several rounds of field-testing in classrooms across the State. A representative sample of questions is then ranked by level of difficulty from the easiest to the hardest.Ìý Committees of teachers then go through several rounds of review, determining which questions reflect the minimum amount of information students must know to pass each exam (65).Ìý The same process is used to determine which questions students must get right to pass with distinction (85).Ìý These are the cut points or anchor points on the scale.

Scale Scores

This information is subjected to statistical analysis that places the questions on a scale of scores according to their level of difficulty.Ìý The State tests and Regents examinations use a scale of 0 to 100.Ìý Raw scores must be converted to the equivalent scale score.

Equating and Ensuring Fairness

Each subsequent test is then equated to the established scale. If there is a greater or lesser number of difficult questions one year, the number of questions that must be answered correctly is adjusted.Ìý This helps ensure that all tests in a subject are equated -- no test is harder or easier to pass from year to year.

Example: If January’s test has a greater number of hard questions than June’s test, a student may need to answer fewer questions right on January’s test to achieve the same grade as a student taking June's test.

In this way, a student getting a passing score on June’s test will be showing the same level of knowledge as a student getting a passing score on January’s test.

Why Does É«ÖÐÉ« Use Scaled Scores Instead of Percentage Correct Scoring?

• In the past, before the É«ÖÐÉ« Learning Standards were adopted, there was no direct relationship between a student’s Regents examination score and achievement of the State Learning Standards in that subject area.
• In É«ÖÐÉ«, State examinations are now used to determine if students have achieved É«ÖÐÉ« Learning Standards.
• Other kinds of scores, such as percentage correct, do not have this built-in reference to achievement and, therefore, are less valid measures of achieving the State Learning Standards.
• É«ÖÐÉ« Regents examinations traditionally used 65 percent of all questions correct as a passing score. This passing standard reflected what was considered to be general competence in the subject matter being tested, not what was needed to achieve any explicit standards.
• By using a scaling method rather than a straight percentage correct, the difficulty of each test’s questions is taken into consideration -- no test is harder or easier to pass from year to year.
• In this way, students can be assured that with the same level of skills and knowledge, their score does not depend on whether this is an easier form of the test or a more difficult form of the test. The same scale score always represents the same level of achievement of the State Learning Standards.
• The overall goal of our scale scoring procedures since this change in the late 1990s has been to ensure that the achievement of a Regents diploma represents the same relative achievement of all of the State Learning Standards in the subject examinations required for graduation.

Does A Regents Examination Score of 65 Indicate That 65 Percent of Questions Were Answered Correctly?

• No. The passing score of 65 is not intended to indicate answering 65 percent of questions correctly.Ìý It indicates achievement of the State Learning Standards, as determined by a committee of É«ÖÐÉ« teachers during standard setting, as described above.
• É«ÖÐÉ« has chosen to use a 100-point scale with 65 as the standard for passing. Though it may look like the scoring of Regents examinations as in the past, it is a scale score, not a raw score or a percentage correct score.
• The Department could have chosen a scale other than the 100 percent scale.Ìý The SAT uses 200 to 800.Ìý Advanced Placement tests use 1 to 5.Ìý We use 1 to 4 on the elementary and intermediate tests.