The point factor method of job evaluation, with a worked example

The method in five steps

1. Choose the factors and subfactors. Article 4(4) of Directive (EU) 2023/970 requires the criteria to include skills, effort, responsibility and working conditions. Each is split into subfactors: the EU guidelines use fourteen, and the ILO guide says most methods use between 10 and 16.
2. Describe the levels. Each subfactor gets levels, from 0 (not required) up to 5, with knowledge running to 8 to match the European qualifications framework. Each level says what the job requires at that point.
3. Weight them. The weights say how much each subfactor counts, and add up to 100%. The EU default puts 40% on skills, 35% on responsibility, 15% on effort and 10% on working conditions, out of 1,200 points.
4. Score each job. For every subfactor, pick the level that matches what the job requires. Its points are the subfactor’s points divided by its top level, times the level, and the job’s total is the sum.
5. Group the totals into grades. Jobs whose totals fall in the same range are treated as work of equal value. The guidelines’ workbook uses ten ranges of 120 points.

A worked example: two roles in the same grade

The two roles below come from the tool’s synthetic sample company. Customer service is held mostly by women (26 of 34) and driving mostly by men (24 of 26). The levels are the sample’s; you can check every figure by hand.

The jobs are nothing alike: customer service scores on communication, emotional effort and customer data, driving on vehicle skills, equipment and road conditions. Both land in grade 5, 481 to 600 points, so they are work of equal value and one category of workers. If drivers are paid more, the difference needs a reason that has nothing to do with sex.

Scoring the person instead of the job is the commonest slip: rate the level the job requires, not the qualifications or the performance of whoever holds it (EU guidelines, Tool 5, step 4).

Weights: the EU default and the ILO guidance

There is no single correct weighting. Both sources below present theirs as examples, and both say the weights must suit the organisation and be justified without reference to who does the work.

The guidelines weight skills and responsibility highest because they are fundamental to most jobs, and say their weights were set so that elements frequent in jobs held mainly by women, such as communication, are properly recognised. To change them, they ask for a strong, objective, gender-neutral reason, testing on women-dominated and men-dominated jobs, and a written record of the change.

The ILO guide adds a test: once jobs are scored, call a subfactor “female” where women-dominated jobs score higher on it and “male” where men-dominated jobs do, then compare the average weights of the two groups. In its example the male subfactors averaged 12.4% and the female ones 7.5%, a gap the guide says calls for the weighting to be re-examined. The tool runs this test on every evaluation.

From levels to points

The guidelines’ workbook spreads each subfactor’s points evenly over its levels, so every step up adds the same points and level 0 adds none. The ILO guide recommends this arithmetic progression and advises against geometric ones, which widen the gaps at the top of the scale between entry-level jobs, often held by women, and senior jobs, often held by men.

The same applies to grades. The ILO guide asks for intervals of equal width in points, not percentages, with no overlap, and warns against placing women’s jobs systematically just under a grade boundary. Too narrow a range creates pay differences that are hard to justify; too wide a range empties “equal value” of meaning.

Where bias gets in

• Missing demands. Emotional effort, active listening, coordinating schedules and responsibility for people’s well-being are often left out or folded into a single line (EU guidelines, Tool 0).
• Double counting. The same demand scored twice, such as handling heavy equipment under both skills and effort, overvalues the jobs that have it (ILO guide, chapter 4).
• Halo effect. A job that scores high on the first factors is pushed up on the rest. Score each subfactor on its own and do not steer towards a total you expect (Tool 5, steps 4 and 5); the tool can hide the running total while you score.
• Weighting by hierarchy. Giving high weights to what senior jobs do, or low weights to what service jobs face, such as verbal aggression, reproduces the pay structure you are testing (ILO guide, chapter 7).

The tool checks for concentrated weights, subfactors that give every role the same level, women-dominated roles a grade below comparable men-dominated roles, and the ILO weighting test.

Other methods, briefly

Ranking orders whole jobs from highest to lowest; classification sorts them into predefined classes; benchmarking compares new jobs with existing ones. These non-analytical methods rest on overall judgement, which the guidelines say can favour jobs dominated by men and overlook tasks typical of jobs held by women. The guidelines offer a simplified factor comparison for micro-organisations and a pair comparison for small and medium-sized ones, and the point factor method for organisations of 250 or more workers and medium-sized ones with diverse jobs.