In response to: Ransom E. and the Job Evaluation Group (2007) JAN Forum: In response to Kahya E. (2006) A revision of a job evaluation system. Journal of Advanced Nursing 56(3), 314–324.


I am glad that a member of Job Evaluation Group (JEG) has offered comments on my revision and testing of the United Kingdom (UK) job evaluation system in four Turkish hospitals because this provides the opportunity for further discussion and debate. The corrections of minor points of inaccuracy, including the time period of development, are noted. However, the more important assertions are not underpinned strongly by empirical evidence and could only be resolved by testing the revision reported in my Journal of Advanced Nursing (JAN) paper (Kahya 2006a) in the context of the National Health Service (NHS). The results of its testing in Turkish hospitals do suggest that the revised plan produces job scores more fairly than the NHS version of the Job Evaluation System (JES).

Knowledge, education and experience factor

There are four tests for the internal and external consistency of a developed plan (Kahya 2006b, Das & Garcia-Diaz 2001). One is to analyze the degree of field superposition of factors. The correlation coefficient among any two factors shows the degree of field superposition of a factor. If the correlation coefficient is larger than a pre-specified value (e.g., r > 0·85) it is suggested that two factors, to a significant extent, are measuring the same field or requirement, which is defined as multicollinearity. In this case, either may be removed from the plan. In a study (Kahya 2006b), as expected, Pearson correlation coefficients (at the 0·01 significant level) have been found r = 0·70 for ‘Knowledge’ and ‘Education’, r = 0·76 for ‘Knowledge’ and ‘Experience’, and 0·70 for ‘Education’ and ‘Experience’. In some studies conducted on white-collar job evaluation plans, similar results have been reported. These results imply that only one factor may be defined to measure these areas. On the other hand, these factors do not overlap completely: for example, while the education level required for a job increases, the other knowledge area may not change. Therefore, it is possible that the knowledge required to perform a job may include a combination of higher or lower knowledge areas mentioned at different levels, e.g. level 5 for practical knowledge, level 4 for theoretical knowledge, and level 6 for practical experience. Such correlation coefficients as r > 0·85 indicate that job evaluation plans where the ‘Knowledge’ factor measures all the forms of theoretical, technical, and practical knowledge do unfairly measure almost 15% of jobs in an organization. Nevertheless, in many job evaluation systems (e.g. HEY plan), only one factor is preferred in order not to increase the number of factors and make it easier to evaluate jobs.

Responsibility for human resources (HR) factor

I agree that the nature of the demands is more important than the number of staff being supervised. However, that should not mean that the number of staff should not be taken into account for fair job evaluation. The NHS JES plan assumes that there is no difference in managing 8 or 24 nurses in a clinical setting.

Working conditions factor

In job evaluation systems, although not in the NHS JES, environmental conditions and hazards are two separate factors in working conditions category. In several studies, as expected, Pearson correlation coefficients among the factors have ranged from r = 0·80 to r = 0·85. It means that they are not completely distinct job demands. The sample is an extreme. Although the level to ‘Hazards’ factor for an electrical maintenance job is the highest, the level to ‘environmental conditions’ factor can be one or much level lower than the highest.

The study of clinical nursing jobs

It is claimed that the evaluations in my study were based on false assumptions, and also unacceptable on grounds of equality. Each issue in the evaluation was clarified in my JAN paper . To test the claim that the evaluations in my study are unacceptable in relation to equality, a study would need to be undertaken comparing clinical nursing jobs under the NHS JES and under the revised plan in a NHS setting, and then it could be shown whether or not the two plans produce similar results.

It is believed that the majority of nursing jobs in clinical settings has been matched to a national profile. Were all the nursing jobs matched to a national job in ‘Nursing Services’ or did panel members look for a national job in another category to match a nursing job? It is recommended that all the national jobs required for matching a nursing job should take place in the ‘Nursing Services’ category.

I agree that AfC JAQ is a excellent job analysis questionnaire. In my study, AfC JAQ was not used for gathering current job information. As stated, a more specific factor-based questionnaire (including questions under job evaluation factor headings) was designed, and used for the job analyses.


It is clear that the UK’s NHS Job Evaluation Scheme is arguably the most comprehensive and robust plan developed in recent years. It is hoped that issues raised by the attempt to develop the system in Turkey will be considered by JEG in future review of the AfCJES.