This guide explains three things that are connected in the pay equity analysis: what the adjusted pay gap is and how it is calculated, what pay factors are and how they affect the analysis, and how your pay philosophy governs which factors are used. The guide gives you the background you need to interpret results in Sysarb and make well-informed decisions.

The most common pay gap that most people are familiar with is the unadjusted pay gap. It measures the salary difference between men and women expressed as a percentage of men’s salary.

The adjusted pay gap similarly to the unadjusted pay gap shows the salary difference between men and women but accounts for gender-neutral factors that can legitimately influence pay such as grade, experience and performance. The part of the pay gap that can be explained by the selected factors are shown as explained. The remaining part after explanations is the adjusted pay gap.

💡 Tip: An adjusted pay gap does not automatically mean something is wrong. It is the part that cannot be justified by the pay factors you have chosen – and which the organisation needs to investigate and take a position on.

⚠️ Important: The adjusted pay gap reflects your pay philosophy – not an external evaluation. The model does not determine what is fair. Your organisation does.

Sysarb uses regression analysis to understand how salary relates to your selected pay factors across the whole organisation. This same analysis is used consistently to calculate:

An important distinction is that the relationships between pay factors and salary are learned from the entire audit, not from individual teams or roles. This means:

Sysarb calculates the adjusted pay gap using a log-linear regression analysis. This is the standard approach in pay equity analysis and is widely used by both researchers and practitioners.

The adjusted pay gap can be calculated with two different methods:

- The decomposition method

- The direct method

The decomposition method (Blinder–Oaxaca) is the primary method. It does not only show how much of the gap is explained in total – it also breaks down how much each individual pay factor contributes to the explanation.

The direct method is used automatically when the decomposition method becomes unstable – for example if the dataset is limited or if pay factors overlap too much. It is more stable but only shows the combined explanatory value, not each factor's individual contribution.

💡 Tip: If Sysarb switches to the direct method, it may be a sign that you have too many factors or too few employees per combination of factors. Try reducing the number of factors.

It may seem unexpected, but a pay factor can sometimes increase the adjusted pay gap instead of reducing it. This depends on two things: how the factor is linked to pay in your organisation, and how men and women differ in that particular factor.

💡 Tip: This is expected behaviour and does not mean the model is wrong. It means that the factor, based on your data, does not explain the pay difference in the way one might have expected. Investigate why and consider whether it reflects your intention.

The adjusted pay gap and expected salaries are two perspectives of the same model.

⚠️ Important: The expected salary is a starting point for analysis, not a pay recommendation. HR context and business judgement are always necessary.

💡 Tip: Expected salary in equal work analysis is a Labs feature – it needs to be activated before it appears in the pay equity analysis. If you have not activated it yet, you will find the setting under Labs. Read more about how to do this here:

Pay factors are divided into two types. Understanding the difference helps you choose factors that produce a stable and interpretable model.

Have a natural order where each step represents a meaningful change. Generally stable and rarely cause model problems on their own.

Examples: Age, Length of service, Level (grade)

Divide employees into groups without a natural numerical order. Each category is treated as its own group and requires more data.

Examples: Job family, City, Agreement

⚠️ Important: Categorical factors are the most common cause of model instability. Use them thoughtfully and avoid combining too many categorical factors in the same model.

Here you can see all pay factors available to choose in the pay philosophy. For a factor to count as a pay factor, it must explain pay in your organisation and at the same time be gender-neutral.

A role's complexity and responsibility based on job valuation. Use when job valuation is a central part of how you structure pay. The strongest and most defensible explanation for pay differences in most organisations.

An alternative way to capture the role's complexity, for organisations that conduct their job valuation outside Sysarb. Use either Level or Position level – not both.

Whether the role involves formal managerial or personnel responsibility. Use when managerial responsibility actively affects pay and managerial and specialist tracks run in parallel.

Individual performance outcomes used in pay-setting. Use only if performance is measured consistently and has a clear link to pay. Review that ratings are applied equally regardless of gender.

Age as a proxy for experience. Use when experience is an accepted part of your pay philosophy and specific experience data is unavailable. Avoid combining with Length of service if they are strongly correlated.

Time the employee has been employed in the organisation. Common in the public sector and collective agreement-based models. Ensure that the factor represents intent, not inertia.

Time the employee has held their current role. Easier to justify than Length of service as it is directly linked to role-specific experience. Often overlaps with Length of service.

A broad functional grouping of roles. Use when market pay differs clearly between different functions. Requires a sufficient number of employees per family for a stable model.

More specific role groupings within a job family. Use when sub-families reflect stable market differences and the groups are sufficiently large.

A role's overall level of responsibility and impact. Use when career bands are clearly defined and actively guide pay decisions. Supports transparent pay dialogues when applied consistently.

Where the work is performed. Use when you deliberately differentiate pay based on geographical location, for example higher pay in major cities. Important in distributed organisations.

Type of employment agreement. Use when different agreement types are linked to different pay structures. Can reveal structural risks if certain groups are overrepresented in specific contracts.

Whether the employment is full-time or part-time. Use with great caution – salaries in Sysarb are already converted to full-time equivalents. Use only if full-time/part-time is clearly supported by your pay philosophy.

Organisation-specific factors, for example certifications, shift type or critical competencies. Use when the factor is clearly defined, applied consistently and actually affects pay.

💡 Tip: Want to see where you configure which factors to include in the analysis? You do this under Pay philosophy in the pay equity analysis:

It may be tempting to choose many pay factors to explain away pay differences. In practice, this often makes the model weaker and harder to communicate. Too many factors – especially categorical ones – lead to groups that are too small and can cause:

💡 Tip: Choose 2–4 core factors that clearly reflect your pay philosophy. Be especially careful about combining multiple categorical factors. If the model fails – start by reducing the number of factors.

💡 Tip: An adjusted R² above 50% generally indicates a reliable model. The goal is not to eliminate the adjusted pay gap entirely – but to make the remaining gap meaningful and actionable.
​
​
​