Home | Contact

Norm Groups and Scale Scores

     

When viewed on its own, any given test score is meaningless. For example, if we know that a person got 49 questions right out of 50 on a reasoning test, this only tells us that the person did well on that particular test. It does not tell us whether such a performance is good, bad or average. If the test were extremely easy, obtaining a score of 49/50 might not signify much. However, if the test were extremely hard, a score of 49/50 might be considered to be very impressive. Hence, for scale scores to be meaningfully interpreted they have to be compared to a particular norm reference group (i.e. to the scores obtained on that test by the general population, by graduates, by apprentices, etc.). This is done by standardising raw (scale) scores against a particular norm group, so the standardised score indicates how the person performed on the test relative to the particular norm group against which the scale was standardised. There are a variety of different scales on which scores can be standardised. These include z-scores, T-scores, IQ scores, stanines, etc. All of these different standardised scales relate the score any one person obtained on the test to the range of scores obtained on that test by the specific norm group against which the scale was standardised. The table below indicates how these different scale scores are related to each other.

 

z-score

T-score

IQ

Stanine

Percentile

-3

20

55

 

 

-2

30

70

1

2

-1

40

85

3

16

0

50

100

5

50

+1

60

115

7

84

+2

70

130

9

98

+3

80

145

 

 

 
 

 

 
 

 
 

 

   

Copyright Psychometrics Ltd 2006 | Privacy policy
Disclaimer | email: info@practicetests.co.uk