Above Level Testing
What is above level testing?
Above level testing is when students are given a test that is meant for an older group of students. For example, many talent search programs require elementary and middle school students to take the SAT or ACT. The SAT and ACT are intended for high school students.
Benefits
There are many reasons that using above-level testing is an appropriate way to measure the individual progress of gifted students.
1) Regression towards the mean
-There is a phenomenon in statistics called regression towards the mean. This means that if a student has an extremely high (think 98th-99th percentile) score the first time he/she takes an assessment, then it is likely that the next time the same assessment is taken the score will drop (move closer to the mean). Gifted students are at a higher risk for growth results to be skewed by regression to the mean because they are more likely to obtain the high scores.
-Above level testing can help reduce regression towards the mean for gifted students because students are less likely to initially obtain an extremely high score.
2) Achievement level
-When gifted students score in the top percentiles of a standardized test it is difficult to translate the score into an accurate achievement level.
-For example, your upper grades students take the End-of-Grade tests each year. Currently, a student scoring the top percentiles earns a 5 on an EOG test.
-The problem is that the student who earns a level 5 on the fifth grade science EOG test may also have mastery of science content in upper grades but cannot demonstrate that mastery on a grade level assessment.
-Using above level testing provides a more accurate depiction of the level of mastery of a student because it raises the test ceiling.
3) Validity
-Test scores of gifted students are more prone error than the average, grade level student. Using above level testing helps reduce this error because it allows gifted students to answer a higher number of questions that the students find difficult.
-When designing a ClassScape (or other) assessment the general rule of thumb is 20% easy questions, 60 % medium questions and 20% difficult questions in our district.
-The problem is that a gifted students may find all of the easy, medium, and even some of the difficult questions easier than their designation. A gifted student may find all of the medium questions easy.
-The easy-medium-difficult question breakdown of 20-60-20 may actually be an 80-10-10 or 90-5-5 for a gifted student.
-This is a problem because there are fewer questions that help determine what a gifted student actually knows.
-With above level testing the question difficulty breakdowns are closer to the 20-60-20 breakdown, reducing error, and improving the validity of results.
Above level testing is when students are given a test that is meant for an older group of students. For example, many talent search programs require elementary and middle school students to take the SAT or ACT. The SAT and ACT are intended for high school students.
Benefits
There are many reasons that using above-level testing is an appropriate way to measure the individual progress of gifted students.
1) Regression towards the mean
-There is a phenomenon in statistics called regression towards the mean. This means that if a student has an extremely high (think 98th-99th percentile) score the first time he/she takes an assessment, then it is likely that the next time the same assessment is taken the score will drop (move closer to the mean). Gifted students are at a higher risk for growth results to be skewed by regression to the mean because they are more likely to obtain the high scores.
-Above level testing can help reduce regression towards the mean for gifted students because students are less likely to initially obtain an extremely high score.
2) Achievement level
-When gifted students score in the top percentiles of a standardized test it is difficult to translate the score into an accurate achievement level.
-For example, your upper grades students take the End-of-Grade tests each year. Currently, a student scoring the top percentiles earns a 5 on an EOG test.
-The problem is that the student who earns a level 5 on the fifth grade science EOG test may also have mastery of science content in upper grades but cannot demonstrate that mastery on a grade level assessment.
-Using above level testing provides a more accurate depiction of the level of mastery of a student because it raises the test ceiling.
3) Validity
-Test scores of gifted students are more prone error than the average, grade level student. Using above level testing helps reduce this error because it allows gifted students to answer a higher number of questions that the students find difficult.
-When designing a ClassScape (or other) assessment the general rule of thumb is 20% easy questions, 60 % medium questions and 20% difficult questions in our district.
-The problem is that a gifted students may find all of the easy, medium, and even some of the difficult questions easier than their designation. A gifted student may find all of the medium questions easy.
-The easy-medium-difficult question breakdown of 20-60-20 may actually be an 80-10-10 or 90-5-5 for a gifted student.
-This is a problem because there are fewer questions that help determine what a gifted student actually knows.
-With above level testing the question difficulty breakdowns are closer to the 20-60-20 breakdown, reducing error, and improving the validity of results.