Language Norms – Extract from KAI Manual.

This information is deemed of general interest to many researchers and to those currently considering certification.

It has another use and that is to warn those reading research reports which ones can be relied upon. We have discovered that some research students have broken copyright and made unauthorised versions of the instrument: without training, using the approved format and instructions, as well as using item variants taken from published articles (e.g., Kirton 1976, which are copyright to journal and author but are not in the fully validated form). Besides breaking the law they are clearly unaware, as were their tutors, that these unvalidated variations of item, format and procedure produce variant (unreliable) results.

Below are the data for the validated language translations.

English Mean (s.d.) range reliability number
UK: Kirton, 1976, 2003 95.5 (17.5) 45-145 .88  (.88*) 808
US: Goldsmith 1985 94.8 (15.9) 44-147 .86 * 214
Tullett & Kirton 1995 94.2 (15.9) 43-147 .89 264
Prato Previde 1984, 1991 94.1 (17.7) 46-145 .87 835
Tullett & Kirton 1995 95.6 (17.3) 46-146 .87 449
Kubes 1989; 05.1 (15.8) 51-149) .84 353
Kirton & Kubes 1992

* Reliability coefficients are Cronbach alpha except those marked * KR20
** French includes Belgian and Canadian; Dutch includes Flemish samples

Female averages are between 6 and 9 points more adaptive.
There are no consistent significant differences in age, status or education.
There are predictable differences between occupation means.

Studies using validated measures by trained KAI practitioners are listed in Publication List. Please access these if there are doubts about procedure or results.
The references above are also in that list, including the latest book: Kirton:
Adaption-Innovation: in the context of diversity and change, Routledge, 2003