LOCAL GOVERNMENT SERVICES JOB EVALUATION SCHEME:

TECHNICAL NOTE NO. 4: OPTIONS FOR DEALING WITH NON-BENCHMARK JOBS

1. Introduction

Local authorities using the Local Government NJC JES have been recommended to select and evaluate a benchmark sample of jobs, in order to develop local conventions and test their evaluation procedures (for advice on selecting benchmark jobs, see Technical Note No 3).

The question then arises as to how to deal with the remaining jobs. Do they, for example, all need to be individually evaluated? Or can some form of further sampling take place? Can non-benchmark jobs be compared with and `slotted' against benchmark jobs?

In practice, only a small district council could feasibly ask all its employees to complete Job Description Questionnaires (JDQs) or to participate in job interviews with Gauge facilitators. For most local authorities, some form of simplified system for dealing with non-benchmark jobs is inevitable if the exercise is to be completed within a reasonable timescale and costs.

2. The Legal Position

The legal position is that a job is not covered by a job evaluation scheme, for the purposes of dealing with an equal pay claim, unless it has been through an analytical process (that is, a factor based JE scheme). The relevant case is Bromley v Quick in the Court of Appeal [1988 IRLR 249 CA].

In this case, the employer undertook a job evaluation exercise in which a benchmark sample of jobs was fully analysed and evaluated. The remaining jobs, including the applicant and some of the comparator jobs, were `slotted' into the resulting rank order on a whole job basis. The Court of Appeal said that some of the jobs in question were not covered by the `job evaluation study' defence provisions of the Equal Pay Act, because they had not been analysed and evaluated under the scheme.

This suggests that `slotting' of non-benchmark jobs against benchmark jobs in the traditional sense of comparing the two on a whole job basis could leave an authority open to equal pay challenge.

However, the Decision does not necessarily imply that every individual job must be evaluated. In Bromley v Quick, Woolf LJ stated that there would be no objection to using benchmark jobs provided there was no material difference between the benchmark jobs and other jobs.

What might constitute a `material difference' is not defined in the Decision, but it seems to leave open a number of possible ways of handing non-benchmark jobs satisfactorily. These are considered below.

3. Dealing with Multiple Jobholder Situations

If, after consistency checking or moderation, all the samples evaluate similarly and fall within the same proposed grade range or within a narrow range of JE scores, then one evaluation (agreed to be the most typical) should be applied to the whole group. If not, then decisions must be taken as to how to allocate the remaining jobs in the group to the varying sample evaluations. This may mean undertaking further sample evaluations to obtain a sufficiently reliable framework to allocate all other remaining jobs in the group.

When using Gauge, the sample Job Overview(s) can be circulated to other members of the group to check for any significant differences (see Appendix 1).

4. Use of Generic Job Descriptions

A number of local authorities have proposed using generic job descriptions for groups of employees, in order to reduce the number of jobs to be separately evaluated, and, in some cases, to increase harmonisation for jobs and grades across departments/ services.

This practice complies with the requirement of the Bromley v Quick Decision, ONLY as long as the generic job description does not conceal significant (material) differences between those covered by it, which would affect the evaluation of different versions of the job, especially if these would result in a grade difference.

If generic job descriptions are to be used as the basis for job evaluation, then the following should apply:

5. A Template Approach

In a number of authorities, the generic job description approach has been developed to deal with large numbers of varying jobs, for example, support jobs in schools. A sample of typical jobs have been identified and evaluated. From the job information and evaluation records, templates have been developed, against which all other jobs can be compared. Only those that do not match any of the templates need to be individually evaluated.

This approach could be adopted for groups of, for example, clerical and secretarial jobs, which vary from department to department. As long as the templates include factor related information, fully reflect the job content and the job information used for comparing other jobs with the templates is reliable, then the legal requirements seem to be met.

6. Options for Dealing with the Remaining Non-Benchmark

Jobs

The discussion so far has focussed on dealing with groups of identical or similar jobs. However, even when these have been allocated to grades or job clusters (see comment above) by reference to evaluated benchmark and/or generic jobs, most local authorities will still have large numbers of unallocated non-benchmark jobs.

The options for dealing with these include:

On the other hand, this may result in a need to carry out hundreds of non-benchmark evaluations, with associated risks of inconsistencies creeping into the process. In a large authority, the work involved may not be feasible within a reasonable timescale and cost.

It too should satisfy the requirements of the Bromley v Quick decision, but may still leave many jobs to evaluate.

Instead of using national benchmarks, remaining non-benchmark jobs are compared against an agreed selection of the local benchmark evaluations on the basis of information from a job description and/or short version of the Job Description Questionnaire. (A table illustrating how the results might be recorded is at Appendix 2. An explanation of how the results can be entered into the Gauge system is at Appendix 3).

This approach is likely to require a relatively large initial benchmark sample to provide sufficient suitable comparator evaluations, but should significantly reduce the overall workload compared to evaluating individually all non-benchmark jobs.

As long as the job comparison exercise is carried out on a factor-by-factor basis, this approach appears to meet the Bromley v Quick requirement for all distinct jobs to have been subject to factor analysis and evaluation.

In the case of the Health Service, the evaluated benchmarks are national ones. If this approach were adopted in local government, an authority would obviously use its own benchmark sample as the basis for profiles for matching purposes.

As with factor comparison, this type of approach would require a relatively large initial benchmark exercise to provide sufficient benchmark evaluations for matching purposes. But, again, it appears to meet the legal requirement for all distinct jobs to have been subject to factor analysis and evaluation.

7. Right of Appeal

The ultimate safeguard that any of the above approaches provide acceptable, fair and non-discriminatory outcomes is the right of an individual employee to appeal on the grounds that their version of the job is significantly different from the evaluated job against which they have been allocated to a grade.

APPENDIX 1

DEALING WITH NON-BENCHMARK JOBS USING GAUGE

1. General Principles

The general principles regarding the evaluation of non-benchmark jobs set out in Technical Note No. 4 apply to all users of the Local Government NJC JES. This note provides options for those authorities using the Gauge software.

2. Job Overview

Gauge users will know that one of the key features of the Gauge system is the Job Overview, which is produced at the end of each evaluation. This Job Overview reflects the ways in which the various questions were answered during the evaluation process and provides a summary of the demands and responsibilities of the job. It also provides the rationale for the job score as that, too, is determined by the way in which the questions were answered.

The wording in the Job Overview is neither job- nor task-specific and it is quite possible for the evaluations of two very similar, but not identical, jobs to result in identical Job Overviews. If this occurs, the scores for the two jobs will also be identical for every factor. Slightly different Job Overviews can also result in the same factor scores, provided that the differences are relatively small.

3. Dealing with large groups of similar jobs

If one or more jobs are thought to be virtually the same as one of the benchmark jobs that have already been evaluated using Gauge, it is quite possible that the original Job Overview will be valid for some or all of these other jobs. As a first step a copy of that Job Overview should be sent to the jobholder(s) concerned with the request that they discuss the content with their line manager and agree whether or not it accurately describes their job.

If both agree that it does describe the job satisfactorily, they need only sign and return the Job Overview and that job can then quite properly be given the same score as the benchmark job.

If it is agreed that the Job Overview is broadly correct but differences exist in one or two specific factors, the `incorrect' statements can be highlighted and the Job Overview be returned for re-evaluation (see below). The line manager should carry the responsibility for ensuring that `smaller' as well as `larger' differences are identified

If significant differences (those likely to justify 1 level or more difference on the relevant JES factor) exist in five or more factors, the original benchmark job is unlikely to be sufficiently similar to be used as the basis for the evaluation of the job concerned. Another benchmark job should be selected or the new job evaluated from scratch.

4. The `Copy' facility in Gauge

The Gauge database has the facility to allow an exact copy of any existing evaluation to be made and added to the database as a `new' evaluation. It will be distinguishable from the original evaluation by having the job title extended by `(COPY)' and the evaluation status will be recorded as `Copy'. The new evaluation has all the features of the original one and can be worked on in the normal way.

5. Re-evaluating the `similar job'

If a job is returned for the re-evaluation of one or two factors (as per 3 above) the first step is to create a `copy' of the relevant benchmark job. The new job title should then be modified in an appropriate way, removing the `(COPY)' extension. The jobholder and line manager should then be invited to re-analyse the points of difference that had been identified.

The `re-evaluate' facility should be used, but applied only to those factors for which differences were identified. Within each factor, the first question for which a different answer would be more appropriate should be the starting point for the re-evaluation, thus ensuring as much consistency between evaluations as possible.

The result of the re-evaluation should then be submitted to the same audit or moderation process as all other jobs and the panel should have the right to look at the whole evaluation, not just those aspects that had changed (this should help ensure that scores are not allowed to `grow' by the inclusion of `advantageous' differences but the exclusion of `disadvantageous' differences).

6. Additional `benchmark' jobs

Once approved, this new job and its Job Overview can be used an additional `benchmark' for other jobs.

In an authority with large numbers of similar jobs it might be appropriate to create and approve two or three slightly different Job Overviews as described above, allowing jobholders to select which one most closely applied to their own jobs. This would be particularly suitable if the points difference between the evaluations did not result in any difference pay.

APPENDIX 2

JOB EVALUATION FACTOR COMPARISON

JOB TITLE: _________________________________________________________________________ POST No:­­­ _______________

SERVICE UNIT:___________________________________________________________________­­­­__________________________________

PANEL MEMBERS: ___________________________________________ _____________________________________________

___________________________________________ _____________________________________________

TOTAL SCORE:

FACTOR COMMENTS:

Knowledge _____________________________________________________________________________

Mental Skills _____________________________________________________________________________

Interpersonal & Communication Skills _____________________________________________________________________________

Physical Skills _____________________________________________________________________________

Initiative & Independence ____________________________________________________________________________

Physical Demands _____________________________________________________________________________

Mental Demands _____________________________________________________________________________

Emotional Demands _____________________________________________________________________________

Responsibility for People _________________________________________________________________________ ___

Responsibility for Supervision _____________________________________________________________________________

Responsibility for Financial Resources _______________________________________­­­­­­­­­­­­­­­______________________________________

Responsibility for Physical Resources _____________________________________________________________________________

Working Conditions ____________________________________________________________________________________________

GENERAL COMMENTS

Signed:………….…………………………………………………………(Management) Date: ………………………………………..

…………………………………………………………………… (Trade Union) Date…………………………………………..

APPENDIX 3

ENTERING JOBS SCORES DIRECT ONTO THE GAUGE DATABASE

1. Use of Gauge

One of the main benefits of the Gauge system for evaluating jobs is the `transparency' of the evaluation and the clear audit trail that is created. Anyone with the right to do so can examine what was said about the job (via the Job Overview or the Question Trace) and thus form an objective view as to whether or not the level determined for each factor and the overall evaluation result or score is `fair'.

The individual factor levels alone do not provide the same robustness for moderation purposes. However, there may be occasions when these levels have already been established by other means and authorities want to enter the results onto the Gauge database without going through the full question/answer process.

2. Security of data

For security and `tamper-proof' reasons, it is only possible to enter (or change) factor levels recorded in the Gauge database by going through the `Evaluate' or `Re-evaluate' process, ie. by answering questions about the job.

To enable a score to be entered `directly', therefore, new questions and answers have to be added to the system, one for each factor.

3. Enabling `direct entry'

During an evaluation, the introductory `*99' question at the start of each factor describes the focus of the next set of questions. It has only one answer, `OK', and selecting this leads to the first question about the job demands in that factor

To enable a pre-agreed factor level to be entered without going through the normal question/answer process, a second answer has to be added to the *99 question for each factor:

`Enter factor level direct'

Selecting this answer would lead to a new follow-up question:

`Select the appropriate level for this factor from the list below.'

The answers would simply be a list from `level 1' to the highest level available for that factor. Selecting an answer would allocate that level to the job and move the evaluation forward to the next factor.

The statement in the Job Overview for that factor would read: `The level for this factor was entered directly'.

4. One factor or all factors?

Once the system changes described in 3 above have been made, the factor levels for a job could be entered into the Gauge database in one of three ways:

Both the Question Trace and the Job Overview will indicate which, if any, factor levels have been entered directly.

5. When to use `direct entry'

Direct entry of factor levels should ONLY be undertaken if those levels have already been determined and agreed through an appropriate process that has been approved jointly. An example of this might be the levels agreed for the Authority's benchmark jobs, assessed by a properly constituted and trained job evaluation panel using the paper version of the NJC's Single Status scheme.

Direct entry should NOT be used to change a level for a factor that has been established for a job through the normal question/answer process, as this would destroy the integrity of the system. This applies even where a Moderation, Audit or Review Panel decides that a factor level is inappropriate.

6. Obtaining the `Direct Entry' facility

Because of the potential for abuse described above, the `direct entry' facility will not be built into the NJC Gauge system as a standard feature.

Any Authority wishing to take advantage of this facility should first get local agreement to its use and then contact Pilat, who will discuss the necessary arrangements to have their system suitably modified.

APPENDIX 4

PROFILE MATCHING FORM

JOB TITLE: PROFILE:

JOB STATEMENT: (up to 4 key elements)

TECH NOTE 4 APRIL 2004

Page 16 of 16