An Introduction To Progressive Achievement Tests in Mathematics (Patmaths) Fourth Edition
An Introduction To Progressive Achievement Tests in Mathematics (Patmaths) Fourth Edition
Scale Scores
Scale scores enable results on tests of different
levels of difficulty to be compared. Scale scores
take into consideration both the level of difficulty
of the test items and the level of ability of
students and are used to measure progress.
For example, a test score of 18 on PATMaths
Test 1 is equal to a scale score of 102.5, whereas
the same test score on PATMaths Test 2 is equal
to a scale score of 110.3. In both cases the
standard error of measurement is 3.9.
Test Interpretation
Percentile Ranks
Percentile ranks provide a simple means of
indicating the rank order and position of the
student’s result in relation to the reference
sample.
For example, a student with a percentile rank of
45 has a score that is higher than the score
obtained by 45 per cent of the reference group
students. A student with a percentile rank of 96
has a score that is higher than the score obtained
by 96 per cent of the reference group.
Test Interpretation
Stanines
Stanines are derived from percentile ranks.
Percentile ranks are divided into 9 categories –
called stanines (short for ‘standard nine’), and the
categories 1 to 9 are used.
Stanines are particularly useful for grouping
students. It is recommended that only differences
of two or more stanines should be regarded as
indicating a real difference in performance.
Relationship between Percentile Ranks and Stanines
Stanine Descriptor
9 Very high
8 High
7 Above average
4, 5, 6 Average
3 Below average
2 Low
1 Very low
The normal distribution curve
Relationship between percentile ranks and stanines
High 8 89-96 7
Above average 7 77-89 12
6 60-77 17
Average
5 40-60 20
4 23-40 17
Low 2 4-11 7
Score Ranges
Results as scale scores. It is also important to pay attention
to the error margin, so that small differences in scale
scores are not given more importance than they deserve.
Error margins tend to be larger for very high and very low
scores.
For example, a student with a test score of 30 on
PATMaths Test 5 would have a corresponding scale score
of 142.2. The error margin for this score is 4.9. We can be
confident that the student’s true score would be within the
range 137.3 and 147.10. That is, 142.2 plus or minus 4.9.
What to do with the data?
Key pages in the
PATMaths Fourth Edition Teacher Manual
Information Page Number
Administration instructions Pages 9–18
Describing test content Pages 1– 6
Sample reports Page 23–28
Test items by level of difficulty by test Page 48
Test items by strand Page 52
Description of scale scores Pages 29–32
Item descriptors and curriculum links Pages 88–108
Definitions for percentiles and stanines Page 68–70
Mean scale scores by year level Page 62
Score conversion tables Pages 77–87
Score key Page 76
Scoring and reporting masters (also on USB or as a Pages 75–108
download)
Some suggestions