PART TWO
07
Point-factor schemes are the most common forms of analytical job evaluation. They were used by 70 per cent of those with job evaluation schemes who responded to the 2017 e-reward job evaluation survey. This chapter explains the basic methodology and the main features of such schemes and describes how they should be designed. Examples are given at the end of the chapter.
Methodology
The methodology of point-factor schemes is to break down jobs into factors or key elements representing the demands made by the job on job holders. It is assumed that each of the factors will contribute to the value of the job and is an aspect of all the jobs to be evaluated but to different degrees. A point-factor scheme can be operated manually – a ‘paper’ scheme – or computers can be used to aid the evaluation process as described in Chapter 2 and further on in this chapter.
Each factor is divided into a hierarchy of levels. Definitions of these levels are produced to provide guidance on deciding the degree to which the factor applies in the job to be evaluated.
A maximum points score is allocated to each factor. The scores available may vary between different factors in accordance with beliefs about their relative significance. This is termed explicit weighting. If the number of levels varies between factors this means that they are implicitly weighted because the range of scores available will be greater in the factors with more levels.
The total score for a factor is divided between the levels to produce the numerical factor scale. Progression may be arithmetic, eg 50, 100, 150, 200 etc or geometric, eg 50, 100, 175, 275 etc. In the latter case, more scope is given to recognize the more senior jobs with higher scores.
The complete scheme consists of the factor and level definitions and the scoring system (the total score available for each factor and distributed to the factor levels). This comprises the ‘factor plan’.
Jobs are ‘scored’ (ie allocated points) under each factor heading on the basis of the level of the factor in the job. This is done by comparing the features of the job with regard to that factor with the factor level definitions to find out which definition provides the best fit. The separate factor scores are then added together to give a total score that indicates the relative value of each job and can be used to place the jobs in rank order.
An example of a point-factor scheme is given in Appendix B.
Job evaluation factors
Job evaluation factors are the characteristics or key elements of jobs that are used to analyse and evaluate them in an analytical job evaluation scheme. Although many of the job evaluation factors used in different organizations capture similar job elements (this is an area where there are some enduring truths), the task of identifying and agreeing factors can be challenging.
There are six main categories of job evaluation factors:
1. The combination of the skills, knowledge and expertise that the employee needs to do the job.
2. The thinking challenges of the job, for example planning, analysis, decision making and problem solving.
3. Interpersonal skills, including leadership, communication and relationship building.
4. The responsibilities that the job has for resources, eg human, financial or physical resources.
5. The kinds of impact that the role has, either on internal operational effectiveness or on the external customer or environment.
6. The environmental, emotional or physical demands that are made in a job, for example difficult working conditions, involvement in dealing with challenging behaviour, pressures on those with caring responsibilities, or operational dexterity.
Within these six areas there are many different ways in which jobs can be described. This will depend on the extent to which the organization wants to express jobs in terms of responsibility or the effects of the job on the organization, or in terms of the ‘inputs’ that a job requires, ie what combination of applied knowledge, skills or behaviours (competencies). For example, many organizations include a factor relating to communication skills in their scheme. However, one organization may define this as the interpersonal skills needed to build relationships; another might place emphasis on the level and type of internal or external contacts that the job is required to have; yet another might focus on core verbal and oral communication skills required at different levels.
Respondents to the 2017 e-reward survey listed 29 factors between them but the most frequently used ones were:
1. knowledge and skills;
2. decision making;
3. impact;
4. people management;
5. interpersonal skills.
The factor plan
The factor plan is the key job evaluation document. It defines each of the selected factors and the levels within them and guides evaluators in making decisions about the levels of demand and responsibility present in a job. The basic number of levels (often four, five, six or seven) has to reflect the range of responsibilities and demands in the jobs covered by the scheme. An example of the definition of a factor with five levels is given in Table 7.1.
TABLE 7.1 Example of factor level definitions
|
Judgement and decision making: the requirement to exercise judgement in making decisions and solving problems, including the degree to which the work involves choice of action or creativity. |
|
|
1 |
The work is well defined and relatively few new situations are encountered. The causes of problems are readily identifiable and can be dealt with easily. |
|
2 |
Evaluation of information is required to deal with occasional new problems and situations and to decide on a course of action from known alternatives. Occasionally required to participate in the modification of existing procedures and practices. |
|
3 |
Exercises discriminating judgement in dealing with relatively new or unusual problems where a wide range of information has to be considered and the courses of action are not obvious. May fairly often be involved in devising new solutions. |
|
4 |
Frequently exercises independent judgement when faced with unusual problems and situations where no policy guidelines or precedents are available. May also frequently be responsible for devising new strategies and approaches, which require the use of imagination and ingenuity. |
|
5 |
Deals with widely differing problems calling for extreme clarity of thought in assessing conflicting information and balancing the risks associated with possible solutions. Additionally, one of the main requirements of the role may be to develop fundamentally new strategies and approaches. |
The total points score is allocated to each factor and distributed between each level as in the example shown in Table 7.2.
TABLE 7.2 Example of an unweighted factor plan with arithmetical score progression
|
Levels |
|||||
|
Factor |
1 |
2 |
3 |
4 |
5 |
|
Knowledge and skills (general) |
20 |
70 |
70 |
80 |
100 |
|
Interpersonal skills |
20 |
70 |
70 |
80 |
100 |
|
Judgement and decision-making |
20 |
70 |
70 |
80 |
100 |
|
Complexity |
20 |
70 |
70 |
80 |
100 |
|
Responsibility for resources |
20 |
70 |
70 |
80 |
100 |
This is described as an unweighted factor plan. The scoring system is an example of even (arithmetical) progression. Each step increases by the same number of points. But weighting and scoring systems can vary, as described below.
Factor weighting
Weighting is the process of attaching more importance to some factors than others. An unweighted factor plan is one in which the factors are treated as being equally important and the maximum number of points is the same for each of them. The 2017 e-reward survey of job evaluation found that 39 per cent of the respondents had unweighted schemes. Where weighting is used there are two types:
1. Explicit weighting, in which the maximum points available for what are regarded as more important factors increase.
2. Implicit weighting, in which some factors have more levels than others but the same scoring progression per level exists as in the other factors. The factors with more levels would have more points available to them and would have therefore been implicitly weighted.
Explicit weighting is the most popular approach. It was adopted by 77 per cent of the respondents to the 2017 e-reward survey. Implicit weighting was used by only 17 per cent of respondents.
The number and choice of factors may also implicitly weight a scheme. If two factors that would normally be treated as being of equal importance are compressed into one then implicit weighting of that combined factor has taken place, ie it is undervalued in terms of its significance.
Scoring progression
The scoring progression within each factor can take place in two ways. The ‘arithmetic’ or linear approach assumes that there are consistent step differences between factor levels, as illustrated in Table 7.2. There may be a single score for each level as shown above or a range of scores to give a measure of choice. But most schemes have only a single score to recognize that scoring judgements are insufficiently accurate to be expressed along a scale within a factor. Judging which factor is appropriate is hard enough.
Alternatively, geometric scoring assumes that there are progressively larger score differences at each successive level in the hierarchy to reflect progressive increases in responsibility. This is illustrated in the example of a weighted factor plan in Table 7.3.
TABLE 7.3 Example of a weighted factor plan with geometric score progression
|
Levels |
|||||
|
Factor |
1 |
2 |
3 |
4 |
5 |
|
1 Expertise |
20 |
50 |
90 |
170 |
200 |
|
2 Responsibility |
15 |
35 |
75 |
105 |
170 |
|
3 Problem solving |
15 |
35 |
75 |
105 |
170 |
|
4 Relationships |
15 |
35 |
75 |
105 |
170 |
|
5 Impact |
20 |
50 |
90 |
170 |
200 |
The rank order produced by either of these scoring methods is unlikely to differ much, but it can be argued that the greater ‘distance’ between level scores in geometric progression reflects an increasing gap between job requirements at the top end of the hierarchy. It can also indicate clearer breaks between grades when it comes to designing the structure.
Designing a point-factor job evaluation scheme
The following steps are required to design a point-factor scheme. Methods of introducing such a scheme were described in Chapter 5. It is best to set up a project team consisting of managers and staff representatives to carry out the design with guidance from HR and, if possible, an external consultant. It is advisable to brief them thoroughly on the principles of job evaluation and what the project team will be expected to do.
Step 1: Select and define factors
The guidelines for selecting factors are:
1. The number of factors should not exceed 12 or so otherwise the scheme would become too complex. Many schemes have no more than five or six factors, which can be sufficient.
2. The factors must be capable of identifying the relevant and significant features of jobs in order to support the assessment of the similarities or differences between them.
3. The factor definitions should be clear, relevant and understandable and written in a way that is meaningful to those who will use the scheme.
4. The factors should reflect the values of the organization.
5. They should apply equally well to different types of work including specialists and generalists, lower-level and higher-level jobs.
6. The whole range of jobs to be evaluated at all levels should be covered without favouring any particular job or occupation.
7. The scheme should fairly measure features of female-dominated jobs as well as male-dominated jobs.
8. The choice should not lead to discrimination on the grounds of gender, race, disability, religion or age, or against anyone in one of the LGBT+ categories. Experience should not be included as a factor because it could be discriminatory either on the grounds of gender or age. The same principle applies to education or qualifications as stand-alone factors.
9. Job features frequently found in jobs carried out mainly by one gender should not be omitted, for example manual dexterity, interpersonal skills and ‘caring’ responsibilities. If such features are included, it is important that the scheme captures the range of skills across all jobs, including those that might be dominated by another gender.
10. Double counting should be avoided, ie each factor must be independent of every other factor – the more factors (or sub-factors) in the plan, the higher the probability that double counting will take place.
11. Elision or compression of more than one significant job feature under a single factor heading should be avoided. If important factors are compressed with others it means that they could be undervalued.
12. The factors should be acceptable to those who will be covered by the scheme.
The following sources of information and approaches to support the selection of factors are available.
Reviewing internal strategy/business documents
Looking through existing written materials such as organization or human resources strategy documents can give an insight into the current values and language.
Reviewing people related frameworks or processes
In the past, job evaluation criteria were not necessarily linked to other human resources processes or frameworks. However, many organizations have now accepted the need to adopt a more coherent approach by applying the organization’s values and language across related processes. Reviewing existing job descriptions may be a place to start.
The most obvious potential link is with an organization’s competency framework, as many of the concepts reflected in competencies are similar to job evaluation criteria, albeit expressed in behavioural or skills-based language. The extent to which a link can be made with an existing competency framework will depend on how it has been defined. It will be easier to do where the competencies are skills related to job requirements, eg business acumen (job-centric) rather than person-oriented, eg tenacity (person-centric). The desirability of achieving a degree of linkage was a finding from the e-reward survey, and was one of the main reasons for companies favouring a tailor-made scheme.
Interviews with line managers and other stakeholders
Discussions with key managers can help to get an early perspective on the senior management priorities for the scheme. This group is most likely to have a view about the future demands on the organization and what work will be valued. Early senior manager involvement can also help to dispel myths and misconceptions about job evaluation, and will support the overall communications process – particularly if the managers concerned are those who will later be responsible for approving the scheme.
Focus groups
Structured meetings with employees can be a good way of understanding what aspects of jobs are currently valued and what people think are the most important. Such meetings can also contribute to achieving better involvement and communications. Because employees may be unfamiliar with job evaluation concepts, the agenda will normally need to cover an overview of what job evaluation is, the rationale for introducing job evaluation, what the factors are, and what makes a ‘good’ factor. Views can be explored on possible factors. Focus groups can also be used to obtain the views of employees about how the scheme should be communicated.
Focus groups can be particularly useful for organizations with geographically or functionally diverse constituencies or for developing sector-wide schemes. In developing the UK further education scheme focus groups were run in about a dozen colleges around the country. They were selected to represent different types of institution as well as geographic diversity. The focus groups generated a long list of possible factor headings, which showed a high degree of consistency across the institutions. This input was clustered into 10 main groups, which became the factors.
Consideration should also be given to whether input from other stakeholders is helpful. Where a scheme is to be jointly agreed with trade unions their representatives should be involved at an early stage. Staff groups can also be consulted. A voluntary organization may want to include volunteers in focus groups, or to obtain the views of key trustees.
Project team input
A project team set up to develop the job evaluation scheme can explore possible factors in a number of ways, for example:
· Open discussion – drawing on the inputs that are available to the team from other sources.
· Selecting a number of jobs/roles and exploring the differences between them – what makes one ‘bigger’ or ‘smaller’ than another. This can be done informally or through a process such as whole-job ranking or paired comparison.
· Using an existing database or list of common factor headings, posting these up on a flipchart and drawing out and clustering the job dimensions that seem most relevant to the organization; if a consultant is being used, this exercise is likely to use headings from their factor database.
Step 2: Define factor levels
A decision has to be made on the number of levels required (eg four, five, six or seven). This has to reflect the range of responsibilities and demands in the jobs covered by the scheme.
The starting point can be an analysis of what would characterize the highest or lowest level for each factor and how these should be described. For example, the highest level in a judgement and decision-making factor could be defined as: ‘Deals with widely differing problems calling for extreme clarity of thought in assessing conflicting information and balancing the risks associated with possible solutions. Additionally, one of the main requirements of the role may be to develop fundamentally new strategies and approaches.’ The lowest level could be defined as: ‘The work is well defined and relatively few new situations are encountered. The causes of problems are readily identifiable and can be dealt with easily.’
It might then be decided that there should be three levels between the highest and lowest level on the basis that this truly reflects the graduation in responsibilities or demands. The outcome would then be the definition of the factor and each of the five levels as illustrated in Table 7.1. This process is repeated for each factor.
The following guidelines should be used in defining levels:
1. Each level should be defined as clearly as possible to help evaluators make ‘best fit’ decisions.
2. The levels should cover the whole range of demands in this factor that are likely to arise in the jobs with which the evaluation scheme is concerned.
3. The link between the content of level definitions should be related specifically to the definition of the factor concerned and should not overlap with other factors.
4. There should be uniform progression in the definitions level by level from the lowest to the highest level. There should be no gaps or undefined intermediate levels which might lead to evaluators finding it difficult to be confident about the allocation of a level of demand. Some schemes have undefined interim levels. This is only likely to work where there are clear steps in demand between the defined levels and where there are protocols on how to use the undefined levels.
5. The factor levels should represent clear and recognisable steps in responsibility and the demands made on jobholders.
6. The levels should be defined in absolute, not relative terms. They should not rely upon a succession of undefined comparatives, eg small, medium, large. So far as possible any dimensions should be defined.
7. Each level definition should stand on its own. Level definitions should not be defined by reference to a lower or higher level, ie it is insufficient to define a level in words to the effect that it is a higher (or lower) version of an adjacent level.
There is a limit to the number of words that can be included in a factor level definition (too many confuse rather than clarify) and the words themselves cannot always convey accurate meanings. For example, a Level 3 definition for a planning and organization factor may refer to ‘planning and organization of a number of complex activities’. But this does not tell the evaluator how many activities justify a Level 3 rating or what constitutes a ‘complex’ activity. What has to happen is that, as evaluators gain experience in interpreting the level definitions, they record conventions or protocols that help to achieve consistent evaluations by providing guidance on what these or similar words mean, preferably with quantitative examples illustrated by reference to benchmark jobs that have already been evaluated. One of the main purposes of testing in a design program is to identify where conventions are necessary and ensure that they are made available to evaluators. But this process of developing conventions continues when the benchmark jobs are evaluated, which means that in a large job, project evaluators may have to revisit the conclusions reached in earlier stages of the programme and re-evaluate in the light of the conventions.
Step 3: Decide scoring system
The aim is to design a point-factor scheme that will operate fairly and consistently to produce a rank order of jobs, based on the total points score for each job. Each level in the factor plan has to be allocated a points value so that there is a scoring progression from the lowest to the highest level. There are no rules on what the maximum points should be as long as it is possible to discriminate between levels. The maximum for a factor is typically between 100 and 200.
A decision needs to be made on how to set the scoring progression within each factor. As explained earlier in this chapter, the choice is between two methods: (1) ‘arithmetic’ or (2) linear.
It is best to allocate a single finite score for each level because giving a choice from a range of scores can complicate evaluations. If it is decided that there should be some choice, this should be as simple as possible, for example low, standard and high. If this approach is used, protocols should be developed to explain what each step change means and it should be subject to thorough testing to ensure consistency.
Step 4: Decide on the factor weighting
A decision has to be made on whether the factors should be weighted or not and if so, whether weighting should be explicit or implicit. It is often assumed that weighting is desirable because it is necessary to recognize that the value of a job is likely to be affected more by some factors than others. But experience has shown that the initial choice of factors is more significant than weighting in determining the relative score of jobs. And decisions on weighting can be arbitrary and difficult to justify.
Approach to weighting
There should be a rationale for the weighting system, eg relative value of factors to the organization, agreed relative values – it should not just give the answers people want. Explicit weighting accords with the belief that some factors are more important than others. Implicit weighting is more likely to take place when there is a large number of factors – 10 or more – and the impact of explicitly weighting any factors is less (unless the weighting is so disproportionate that the non-weighted factors become immaterial).
Rather than rush into a weighting decision because it seems the right thing to do, consideration should be given to initially try out an unweighted scheme and test it as described in Step 7. But if it is felt strongly by the team that weighting is desirable then the design process would continue as indicated below. However, unless the scheme is to be weighted in favour of a factor that applies strongly to a specific area of work, the re-modelling of weightings is likely to yield fairly small changes in rank order as all jobs will be affected to the same degree by the re-weighting. Whatever approach to weighting is adopted, the results should be tested and if these are unacceptable, further testing can be undertaken using alternative forms of weighting.
Making weighting decisions – the judgemental method
Explicit weighting decisions are often judgemental although they can be based on agreed guiding principles (eg no factor will have a weighting of less than 5 per cent or more than 30 per cent). Decisions can be made by a job evaluation project team although discussions tend to be based on opinions and feelings. For example, it is often held that knowledge and skills or expertise are more important than anything else and must therefore be weighted. A typical method is to ask each member of the team to distribute 100 points amongst the factors. The distributions are then reproduced on a flip chart by the facilitator, who does not reveal their source. The team next discusses the distribution of suggested weightings and comes to a conclusion. Averaging the weightings should be avoided if possible. It is far better to reach a decision through consensus after a debate. This process is totally unscientific. It is essential, therefore, to make the initial weightings provisional and test them using the full weighted factor plan.
Implicit weighting decisions can be made when analysing the number of levels to be made available for each factor. A restricted range of levels might be adopted for one or more factors when the scope of the factor or factors is considered to be more limited than others, and the factors will thus have been implicitly weighted. But this is a matter of judgement, which can easily be subjective.
Making explicit weighting decisions – the statistical method
The statistical method of multiple regression analysis has been used in an attempt to make explicit weighting decisions more scientific. Multiple regression analysis measures the relationship between a number of variables to indicate how they affect one another. When applied to weighting, the aim is to find the weighting combination that will most closely replicate a predetermined rank order of jobs (the reference ranking). This ranking might be based on comparing the market rates of a number of benchmark jobs, paired comparisons of the benchmarks, or what is considered subjectively to be a fair ranking of the benchmark jobs (‘felt-fair ranking’, which is in accordance with the somewhat circular felt-fair principle originated by Elliott Jaques to the effect that if someone feels a payment system is fair then it is fair)
However, unless the reference ranking has itself been developed using an objective process, the regression analysis may simply reproduce the existing hierarchy, thus reinforcing inequalities or preconceptions. The use of market data is particularly suspect in this respect, as pay discrimination in favour of male-dominated roles may already be embedded in the market.
This step will complete the design of the draft factor plan.
Step 5: Select and analyse benchmark jobs
A sample of benchmark jobs should be identified. These will be used to test the factor plan and retained as ones for which external market rate comparisons can be made and as examples of jobs in each grade to assist in evaluation. They can also be used for matching on a job-to-job basis (see Chapter 8).
Benchmark jobs should represent each level in the organization and be chosen from each of the main functions. The proportion depends on the size and complexity of the organization; in a large organization it may be up to 10–15 per cent of the jobs if the organization is fairly simple, but it could be much higher in some complex organizations (30 per cent or more). The benchmark jobs are then analysed in terms of the factors.
Step 6: Evaluate benchmark jobs
The project team uses the factor plan to evaluate the benchmark jobs. Initial training sessions are helpful; the team practises making evaluations with the plan using made-up job profiles. One of the following three methods of evaluation can be used.
1. Factor by factor
The team takes each factor in turn and evaluates all jobs in respect of that factor, ie whole jobs are not evaluated in turn. This is the best method but it takes time. A factor-by-factor approach rather than a job-by-job approach means that panel members are less likely to make decisions on total scores based on a priori judgements about the relative value of the whole jobs, which they might find hard to change. Instead they are looking at responses for that factor in relation to responses for other jobs. They are more likely to focus on ‘read-across’ analytical judgements about the level of particular factors and it will be easier for them to refer for guidance to previous evaluations of the same factor in other benchmark jobs.
When there are variations in factor evaluations, individual panel members are asked to give reasons for their conclusions. But the facilitator or chair has to be careful not to allow them to be pressurized to change their views. If team members have been properly briefed and if there is a carefully designed, tested and understood factor plan and good information about the job, the extent to which evaluations vary is usually fairly limited, which enables consensus to be achieved more easily.
2. Whole jobs disclosed
Each panel member evaluates whole jobs factor by factor and then informs the panel of their conclusions. The facilitator records their separate views and then initiates a discussion with the objective of achieving consensus on the rating for each factor and therefore the overall score. This can be time-consuming because panel members may be influenced by pre-formed judgements and, having made up their minds, find it difficult to shift their ground. Again, time is saved if individual or pairs of panel members evaluate the jobs prior to the panel meeting, and the responses are collated on a spreadsheet. It also helps if the collated responses can be reviewed by panel members prior to the group meeting, as it will give time for the panel members to review their own responses in the light of any areas of difference.
3. Whole jobs undisclosed
This is a variation of the second approach where the whole panel is involved in evaluating each job. Each panel member evaluates whole jobs factor by factor but does not communicate their views formally to other members of the panel. Instead, the panel as a whole discusses and agrees the evaluation of each factor in turn to produce a total job evaluation score. This speeds up the evaluation and consensus can be easier to achieve because panel members are not put in the position of having to defend their prior judgements against all comers. But there is the danger of the weaker members of the panel allowing themselves to be swayed by strongly expressed majority views. This approach therefore only works with good facilitation that ensures that the discussion is not dominated by one or two powerful members and that all members have their say. Where the complete panel is evaluating jobs, some organizations start with the whole job disclosed method and when panel members (and the facilitator) become more experienced move on to the whole job undisclosed method.
Step 7: Test draft factor plan
No one ever gets a factor plan right first time. Even if a project team has already worked through two or three drafts of the plan, they are likely to come up with some changes or queries about relativities between statements, weighting and scoring systems.
The plan needs to be tested on the benchmark jobs. The fundamental aims of the test are to check on the extent to which the factor plan enables evaluators to produce sensible and acceptable results and to identify any improvements that need to be made to any of the plan’s features. The clarity of the plan and the ease with which it can be used is also tested. The test can also begin the process of establishing conventions, which will be used to interpret the meaning of factor-level definitions.
Each benchmark job is evaluated and scored by the project team and then ranked according to the total score.
The project team then considers the extent to which it is believed the rank order is valid in the sense that the evaluations correctly indicate relative job value. There is no ideal method of doing this – no single, simple test that will establish the validity of a factor plan. The fact that it is difficult to prove that the results are valid is a major weakness of point-factor evaluation.
The following methods are available:
1. Reference ranking – the team compares the ranking produced by the job evaluation with the rank order produced by a ranking exercise. The technique of paired comparison (see Chapter 2) may be used to guide the ranking process. The problem with this approach is that the ranking may do no more than confirm existing relativities, which may not be appropriate.
2. Hierarchy comparisons – the rank order produced by the test is compared with the existing organizational hierarchy and any obvious discrepancies are investigated. However, care must be taken not to assume that the existing hierarchy is the correct one.
3. External market test – the internal rank order is compared with that existing in comparable jobs elsewhere. But this may reflect pay differentials between job families, rather than internal measures of job worth. It may also replicate existing inequities between male and female jobs.
4. The ‘felt-fair test’ – the rank order produced by the test is compared with what the job evaluation panel ‘feels’ is the fair and therefore appropriate ranking. Discrepancies are then identified. A danger of this approach is that it is often based only on opinion and can simply reproduce existing prejudices and perpetuate existing anomalies. Inevitably, however, a ‘felt-fair’ approach is often adopted as the main method of testing the plan. It is therefore necessary to be particularly alert to the possibility of pre-judgements affecting opinions and to challenge any apparent move in that direction. One or more of the first three tests can be used for this purpose.
The more specific aims of the test are to establish the extent to which:
· the factors are appropriate;
· the factors cover all aspects of the jobs to be evaluated;
· the factors are non-discriminatory;
· the factors avoid double counting and are not compressed unduly;
· level definitions are worded clearly, graduated properly and cover the whole range of demands applicable to the jobs to be evaluated so that they provide good guidance on the allocation of factor levels to evaluators and thus enable consistent evaluations to be made;
· the scoring system is appropriate;
· weighting decisions are defensible.
Step 8: Confirm factor plan
Changes are made to the factor plan in the light of the test results. If there have been major changes another test will be required.
Possible conversion of a paper scheme to a computer-aided scheme
The steps set out above will produce a paper-based scheme, which is the most common approach. But converting the paper scheme to a computer-aided scheme can offer a number of advantages including greater consistency, speed and the elimination of much of the paperwork.
Computer-aided schemes use software provided by suppliers such as Pilat but if the job evaluation scheme is bespoke, it will typically be derived from the paper-based scheme. If a computer-aided scheme is adopted, the job evaluation project team or panel will not be required to spend time carrying out detailed paper evaluations but it is necessary to set up a review panel which can validate and agree the outcomes of the computer-aided process. A grading outcome is only as good as the job information that has been entered into the system and hard lessons have been learnt by organizations that have ended up with a ‘fully automated’ scheme but unacceptable outcomes.
Examples
Crown Prosecution Service
As a government agency, the Crown Prosecution Service (CPS) is obliged to use job evaluation, a policy initiated by the Treasury in 1992 to support job analysis and grading in the Civil Service. There are two different point-factor schemes: JEGS – Job Evaluation and Grading Support – is used for jobs up to and including grade 7 posts, and JESP – Job Evaluation for Senior Posts – is used for senior civil servants, ie all those on grade 5 posts (assistant director level) and above.
The CPS says that the most important reasons for using job evaluation are to ensure consistency of grades across the service and to establish the relative worth of the jobs. It also facilitates the accommodation of new or revised jobs into the grading structure and helps to avoid discrimination. JE therefore underpins all the organization’s pay and performance initiatives.
Process
There is a job evaluation protocol, agreed between the trade unions and the CPS, which sets out under what circumstances a job will be evaluated and the requirements from employees, managers, the departmental trade unions and HR staff to ensure that the evaluation is accurate, objective and fair. All jobs are evaluated independently by HR and union representatives, and are then passed to a quality assurance panel, also consisting of representatives of HR and the unions who will determine the final assessment. There are trained evaluators around the country, although HR controls the process and policy centrally.
Roles are benchmarked against the public-sector labour market, and market analysis shows that the CPS pays upper quartile rates. Where there is conflict between internal equity and the market the CPS will sometimes pay supplements.
For new jobs, HR consults with the trade unions and produces a job description, person specification and an ‘indicative’ evaluation. Once the job holder has been in post for up to a year, the job holder fills out a job analysis form and the job is re-evaluated. A job may sometimes be advertised as a temporary grade, so that applicants are aware that it may change.
Dip sample
The streamlining process involves taking a dip sample, so that instead of interviewing everyone in a particular role, a sample of, say, 2.5 or 5 per cent of the population are asked to complete a job analysis form, and these are evaluated by the panel to establish the appropriate grade, which could be higher or lower than the existing grade. The sample is selected by gender, age, ethnicity and disability, to reflect the profile of the whole group.
Communication
The CPS recognizes that it is important to ensure that each step of the JE process and design is understood by all and that there is no room for different interpretations of what the design is trying to achieve. Involving the trade unions and training a pool of union representatives in the evaluation process makes it easier to get buy-in to changes in grades. The protocol agreed with the unions sets out details of the JE process and its scope. In addition, there is an easy-to-read process manual for staff and managers, which sets out the circumstances under which they can ask for a job to be evaluated.
Maintenance
The panel meets as and when needed. In general, however, job evaluation is a weekly activity, with new or changed jobs coming through regularly. It takes around an hour to carry out an indicative evaluation and about half a day to do a full evaluation. The process is fully computerized, which contributes to a speedy turnaround.
Civil Service JEGS and JESP schemes
TABLE 7.4 Civil Service JEGS and JESP schemes
|
JEGS |
JESP |
|
|
Who does it cover? |
All staff up to and including grade 7 |
Senior Civil Service (grade 5 – associate directors and above) |
|
Factors |
• Knowledge and skills • Contacts and communications • Problem solving • Decision making • Autonomy • Management of resources • Impact |
• Managing people • Accountability • Judgement • Influencing • Professional competence (where appropriate) |
Rencol Tolerance Rings
Rencol Tolerance Rings is an SME based in the UK. It is the only UK manufacturer of tolerance rings – industrial fasteners – and is unable to recruit in the external labour market for staff with product knowledge, so employee retention is a priority. Manual workers make up about 70 per cent of the workforce. Rencol is a non-union company, but consults extensively and has had a works council since 2005. It introduced a point-factor job evaluation scheme for all staff in 2007, largely at the behest of employees, many of whom believed their roles were undervalued.
The scheme was designed by a joint working party led by ACAS. Once the scheme had been designed and every job evaluated, a new seven-grade pay structure was constructed, with grade boundaries established where there were obvious gaps between clusters of jobs. Rencol is happy with what it has achieved, although in retrospect it says it would have taken a more strategic approach to the whole process, from embarking on JE to adopting new pay and progression arrangements.
Rencol’s factors
· knowledge/experience (the most highly weighted factor)
· communication skills;
· handling information;
· physical skills;
· autonomy;
· responsibility for staff;
· responsibility for financial resources;
· responsibility for product and physical resources;
· working conditions.
The process
The original purpose of the scheme was to establish relativities between jobs that would be understood and perceived as fair by employees. And since it was employee feedback, expressed in opinion surveys, that prompted the adoption of job evaluation, the company believed that it was important that employees were involved. It asked ACAS to assist, and the ACAS facilitator helped to set up a joint working party to design a tailor-made point-factor scheme and take part in its implementation. The working party consisted of six people and the HR manager as an ex-officio member.
The company asked for volunteers and then picked people to represent different occupational areas and levels of responsibility within the business. It took around six weeks to train everyone involved and to evaluate some benchmark jobs, and another 12 weeks for all 50 jobs to be evaluated. Two panel members interviewed each job holder and completed a job analysis questionnaire. The job holder and their manager then reviewed the job analysis to ensure that nothing had been left out, and the factor points were added together to give a total for each job.
Although this had not been part of the original plan, it became obvious once all the jobs had been evaluated that the company needed to introduce a new grading structure. Before this, everyone had been on spot rates, which had developed without any rationale over the organization’s 25-year history. A seven-grade structure was put in place and staff were informed about their new pay band, though not their individual JE scores. Grades were defined in terms of a range of points scores and jobs are placed in a grade if their score is within the range, but there is no link between an individual’s pay and where they sit in the band.
Communication
The company holds monthly meetings to which all staff are invited, and these, together with a monthly newsletter, were used to keep everyone informed about the job evaluation process. Once the work was finished there was a full presentation at a monthly meeting and every employee received a letter telling them about their new grade. Employees reacted favourably and most felt that the JE exercise had come up with the right results.
Maintenance
The JE scheme is maintained on a rolling basis by the HR manager, who evaluates one job a week. He meets every job holder, or representative where several people do the same job, to review the job analysis. Where there have been any significant changes, the job is referred to the six-member panel, some of whom were involved in the original working party. New jobs are also referred to the panel. The full JE process is used in every case.
Equity and other problems
When a job is about to be filled or a new job created, it is benchmarked against the external market. Only after this is it evaluated, because the evaluation process is based on the experiences of the person doing the job. That person may not be fully aware of all the demands of the job so the situation can arise that the internal JE score and grade is below their market rate. Even though this is dealt with through market rate supplements, the company is still concerned by what it sees as a problem of internal equity. In addition, existing staff were assimilated to the nearest point on the new pay structure, which also means that person A may be earning less than person B who has more responsibility but is lower down the grade.
Rencol is aware that there is a temptation for managers to add responsibilities to a job to get the post holder some extra money.
Rencol’s dos and don’ts
TABLE 7.5 Rencol’s dos and don’ts
|
Dos |
Don’ts |
|
• Ensure that you use a balanced, cross-functional team • Communicate well with all employees, both collectively and individually • Review each analysis, at least on an annual basis |
• Give undue weight to one particular factor as this may distort the outcome • Publish individual scores • Assume that because nobody has said anything a job hasn’t changed |
Stockport Council
Stockport Council started to use job evaluation in 1990: Hay for managers and the Greater London Whitley Council scheme for white-collar employees. But in 1997 a national agreement was reached to implement Single Status in England and Wales, that is, the harmonization of the terms and conditions of former manual workers and APT&C employees. This required a single job evaluation system for white-collar and manual workers. Along with most other councils in the UK, Stockport moved cautiously towards Single Status because of the problem of having to equalize the earnings of male manual workers who have traditionally received bonuses and female workers who have not.
The Greater London Whitley Council JE scheme could only be used for white-collar employees, so, for the last year, the council has been working with the trade unions to implement the Greater London Provincial Council (GLPC) JE scheme, which can be used for all employees, although Stockport is retaining Hay for managers. Jobs on the borders of the two JE schemes are being looked at carefully to ensure that someone at the top of the GLPC scheme would not earn more if they were evaluated through Hay, and vice versa.
Process
When introducing the new scheme, Stockport sought to build on the existing job descriptions and evaluations and evaluated a further 80 benchmark jobs across all the non-managerial occupational groups. Panels consisted of equal numbers of union and management representatives, all of whom had been trained in union methodology. The manager of the role being evaluated attends the meeting as an expert witness, but leaves the meeting once they have answered questions about the job.
There has been a long history of job evaluation in the organization, and there is general acceptance that the process is fair. Despite this, it has in some cases been difficult to get agreement on job content between the employee and their manager. Managers have been urged to ‘evaluate for the future’, that is, to think about future challenges and incorporate them into job descriptions.
Communication
Managers have been kept informed about the progress of the new scheme through the HR bulletin, which is circulated every six to eight weeks. A specific communications exercise focused on the employees who currently receive a bonus, some of whom will be going into pay protection arrangements; they have all received a face-to-face briefing and individual letters explaining what is happening.
Advice
Stockport Council suggests the following ‘dos’ and ‘don’ts’.
‘Do’:
· consider how your proposed scheme will fit into your particular organization;
· consult and communicate widely and prepare the organization for major changes in pay relativities;
· ensure you have sufficient time and resources in place to do the job properly.
‘Don’t’:
· underestimate the amount of time and effort involved in job evaluation activity;
· let it be perceived as purely an HR initiative;
· evaluate for the past. Ensure your scheme takes on board service modernization.
Greater London Provincial Council job evaluation scheme factors
Supervision/management of people – 7 levels
Assesses the scope of managerial duties and the nature of the work which is supervised. Accounts for flexible working patterns, deputizing, the number of staff supervised and their geographical dispersal.
Creativity and innovation – 7 levels
Measures the extent to which the work requires innovative and imaginative responses to issues and in the resolution of problems, and the impact of guidelines.
Contacts and relationships – 8 levels
Examines the content and environment of contacts required as part of the job. Measures the range and outcome of contacts.
Decisions
Operates as two sub-factors:
· Discretion – 7 levels – identifies freedom to act and the controls in place.
· Consequences – 5 levels – measures the outcome of decisions by effect, range and timescale.
Resources – 5 levels
Assesses the personal and identifiable responsibility for resources.
Work environment
Work demands – 5 levels
Considers the relationship between work programmes, goals, deadlines and the subsequent management of priorities.
Physical demands – 7 levels
Identifies a range of postures and demands of a physical nature.
Working conditions – 7 levels
Examines the typical elements encountered with working inside and outside.
Work context – 7 levels
Examines the potential health and safety risks to employees carrying out their duties.
Knowledge and skill – 8 levels
Assesses the depth and breadth of knowledge and skills required. This factor is the most highly weighted.
References
e-reward (2017) Job Evaluation Survey, Stockport, e-reward
Jaques, E (1961) Equitable Payment, London, Heinemann