Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

Interpretation of Test Scores

VIEWS: 70 PAGES: 2

									                                               Interpretation of Test Scores
Many tests express scores in several forms

Raw Scores
Raw scores are the total number of items scored correct on a test. Different tests contain varying numbers of items, so comparisons
between scores on several tests cannot be made using raw scores. They are useful for measuring progress if the same test is repeated at
a later date.

Standard Scores
Standard scores provide the clearest indication of a student’s performance on a particular test. Two ranges are commonly used, one with
10 as the average and one with 100 as the average. The ranges are:

                     Scaled scores with average = 10          Standard scores with average = 100
                     17 – 20   very high                      130 and above     very high
                     15 – 16   high                           120 – 129         high
                     13 – 14   above average                  110 – 119         above average
                     8 – 12    average                        90 – 109          average
                     6–7       below average                  80 – 89           below average
                     4–5       low                            70 – 79           low
                     1–3       very low                       69 and below      very low


Percentiles
A percentile rank indicates how a child’s score compares with that of other children of the same age. For example, a percentile rank of 60
indicates that 40% of children of the same age would be better on that particular test and 59% of children would do worse.

Stanine
Stanine scores are useful in comparing a student's performance across different content areas using units from 1 to 9. Typically, stanine
scores are interpreted as above average (9, 8, 7), average (6, 5, 4), and below average (3, 2, 1).Stanine scoring is usually easier to
understand than other scoring models. Though they are good at signifying broad differences in performance, stanines should be used
cautiously when making any finer distinctions about performance.



Created by Debra McKeown
The normal curve shows how these scores relate to each other. The top of the curve indicates where the largest number of people would
score (i.e. the average Standard Score is 100, the average percentile is 50%).




Created by Debra McKeown

								
To top