Download Arc Guide to Special Education Testing Terms

Arc Guide to Special Education Testing Terms
_______________________________________________________
Terms used in special education testing can be confusing. Test scores may be reported in a
variety of ways including age equivalents, grade equivalents, raw scores, scale scores, subtest
scores, and standard scores. This Arc Guide explains commonly used terms.
Statistics
• Way to measure things including change and lack of change
• Describes relationships between things using numbers
• Using numbers to measure progress or lack of progress
• Mathematics that involves the collection, analysis, interpretation, and presentation of
numerical data
• Summarizes data
• May involved mean, median, mode, and standard deviations
Average Score/Mean Score
• Number or amount that is typical of a group
• Midpoint between extremes
• Number or score that is obtained by adding up all the numbers and dividing by the amount
of numbers. For example, 15 + 5 + 10 = 30, divided by 3, average is 10
Age Equivalent Score (AE)
• Provides age at which most individuals perform a skill or task
• Compares student’s performance to that age
• Almost always used for tests of intellectual ability
For example, if a 5 year old student gets a raw score of 20 on a test, and that is the
average score for 5 year olds, the AE would be 5.
Bell Curve
• A theoretical curve used to diagram scores
• Continuous frequency distribution that defines the probability of something, graphically
represented as a bell, with equal numbers on each side and approaching zero at the end
• Most scores would be at the top of the bell with a curve narrowing to zero on each side
• Allows for objective comparison of a student’s percentile rank against a group
• Allows for comparison of a student’s progression or regression against a group
• Scaling of scores that reflects the shape of a bell
Cluster Score
• Based on a grouping of scores
• Must be viewed with caution when there is a difference between subtest scores
Composite Score/Index Score
_____________________________________________________________________________________
For further information or advocacy services, contact The Arc Greater Twin Cities at 952-920-0855
or visit www.arcgreatertwincities.org. Thank you!
This document is not legal advice, and should not be construed as such. Thus, no information herein
should replace the sound advice of an attorney.
Arc Guide to Special Education Testing Terms
_______________________________________________________
•
•
An overall score comprised of several subtests
The score determined by summing two or more scores
Confidence Interval
• Also known as margin of error
• The range in which a student’s scores will fall within a certain percentage most of the
time
• Most confidence intervals are set at 95%
Confidence Level
• Describes how sure (probability) one is that the score is between a certain interval or
range of numbers
• Intervals are most often referred to as 95% or 99%
• Describes the uncertainty associated with a sampling method
• A measure of how much it can be thought that the result is correct
Criterion Reference
• Determines whether or not a student meets a certain criteria or content (something the
student does/does not do, and/or how much they do), has no reference to a norm group
• Measures student performance against a fixed, predetermined standard
• Description of what a student should know or do at a specific state of development
• Goal is to obtain a description of the knowledge and/or skill of the student
Grade Equivalent Score (GE)
• Compares the student’s performance on grade-level activities against the average
performance of student’s at other grade levels
• Reported in terms of grade level and month
For example, a GE of 4.5 in math means the students solves math problems similar to
the average 4th grader
• Scores present an overall approximation of grade level achievement
• A single GE score does not mean that the student is working at the grade level in all areas
• States an approximation, not a truly reliable way of documenting ability
Norm/Mean/Average
• Average level of achievement
• 50 percentile; divides the group into 2 groups
• Characteristic of most of the population (which may vary depending on a variety of factors
including age, size, sex, ethnicity, culture, etc.)
• Scores often compared to others of the same age (age normed) or grade (grade normed)
Norm Referenced
_____________________________________________________________________________________
For further information or advocacy services, contact The Arc Greater Twin Cities at 952-920-0855
or visit www.arcgreatertwincities.org. Thank you!
This document is not legal advice, and should not be construed as such. Thus, no information herein
should replace the sound advice of an attorney.
Arc Guide to Special Education Testing Terms
_______________________________________________________
•
•
•
Compares scores in comparison to the average student scores of the same age or grade
level
Ranks students in relation to others in the a statistically selected group
Tests should have been normed using similar test-taking group
Percentile Rank
• Divides the data into 100 equal parts, measures position from the bottom
• Compares test performance to a sample of similarly-aged students
• Compares a percentage of scores in the frequency distribution that is the same or lower
than the score (For example, if a student obtains a percentile rank of 70, the student
performed the same or better than 70% of the other students)
• Determines the relative standing or rank position within a population
Rank Order
• Arrangement according to rank (For example first to last)
• Relationship between items (For example high to low)
Raw Score
• An original student test score that has not been transformed for statistical analysis or
adjusted for position in relation to others’ scores
• Describes raw number of items answered or performed correctly
Reliability
• Describes precision
• A measure of consistency
• Degree to which a score should compare to previous (same) test results
Sample
• An individual or small group representative of a larger group
• Used to compare a small group against a larger group
Scale Score
• Conversion of the raw test score to a common scale that can be used for comparison
Standard Deviation (SD)
• Measures the amount of variation from the average
• In most educational and psychological tests, the mean is 100 and the standard deviation is
15. The mean is zero standard deviations.
One standard deviation above the mean (+1 SD) is always the 84th percentile
One standard deviation below the mean (-1 SD) is always the 16th percentile
Two standard deviations below the mean (-2 SD) is always the 2nd percentile
• Most subtests have a mean of 10 with a standard deviation of 3
_____________________________________________________________________________________
For further information or advocacy services, contact The Arc Greater Twin Cities at 952-920-0855
or visit www.arcgreatertwincities.org. Thank you!
This document is not legal advice, and should not be construed as such. Thus, no information herein
should replace the sound advice of an attorney.
Arc Guide to Special Education Testing Terms
_______________________________________________________
•
•
•
•
•
•
Two thirds of children will score between +1 SD and -1 SD from the mean (see bell curve
definition above)
A number indicating the extent of variation from a group
Indicates how widely individuals in a group vary (The more the group varies, the higher
the standard deviation)
One standard deviation above the mean (+1SD) and one standard deviation below the
mean (-1SD) typically accounts for 68% of the people in the group
Two standard deviations above the mean (+2SD) and two standard deviations below the
mean (-2SD) typically accounts for 95% of the people in the group
Three standard deviations above the mean (+3SD) and three standard deviations below the
mean (-3SD) typically accounts for 98% of the people in the group
Standard Score (SS) and Percentiles
• Compares a grading method in a normal (typical) distribution
• With Standard Scores, the average score (which is also referred to as the mean score) is
100 with the standard deviation of 15.
Example: If a student obtains a score of 115, that score can be reflected as 1 SD above
the mean (100 + 15 = 115). That score is the 84th percentile. Likewise, if a student
obtains a SS of 70, that score is 2 standard deviations below the mean (100 – 30 = 70).
That score is the 2nd percentile.
• Most individuals will have a score between 85 –115.
Low average is between 80 and 89
Average is between 90 and 109
High average be between 110 and 119
Standardized Test
• Requires all test-takers to answer the same questions in the same way and scored in a
consistent manner to allow performance to be compared
• All the questions, format, instructions, scoring and reporting are the same for all students
taking the test
• Standardized tests and standardized procedures for creating, administering, and analyzing
the test, allow for comparisons between and among students
Subtest/Subtest Score
• Short tests within a large test
• Measures a specific ability/skill that is combined with other subtests
• Two or more related subtests that reflect different aspects of the same broad ability
Subtest Scatter
• A significant difference between scores on subtests of a test; when subtest scatter occurs,
the reasons for the scatter should be explored
_____________________________________________________________________________________
For further information or advocacy services, contact The Arc Greater Twin Cities at 952-920-0855
or visit www.arcgreatertwincities.org. Thank you!
This document is not legal advice, and should not be construed as such. Thus, no information herein
should replace the sound advice of an attorney.
Arc Guide to Special Education Testing Terms
_______________________________________________________
•
•
Suggests that the student has areas of significant strength and significant weakness
Measured by differences between the student’s highest and lowest subtest scores within a
test or measured by differences between subtest scores and the average scores of each
subtest
Validity
• Accuracy of test measure
• Describe accuracy
• Are different types of validity
Predictive – how well future performance is predicted
Content – how well subject is covered
Consequential – inferences that can be made from test results
• Tests should be both reliable and valid
_____________________________________________________________________________________
For further information or advocacy services, contact The Arc Greater Twin Cities at 952-920-0855
or visit www.arcgreatertwincities.org. Thank you!
This document is not legal advice, and should not be construed as such. Thus, no information herein
should replace the sound advice of an attorney.