Reference: Choosing and Creating Intersections

An intersection is a group that ideally consists of workers with various demographics and characteristics. An intersection is a field or a group of fields that represent the dimensions you want to analyze in the VIBE Index™.

We strongly recommend that you use a minimum and maximum of 2 fields to define the intersection.

Examples of fields used to define an intersection: Ethnicity, Gender, Disability Status, Veteran Status, Sexuality.

Example of an intersection: Gender and Ethnicity.

The values of the fields you use to define the intersection determine the intersection values. Each intersection value is an individual intersection (people group).

Example: The intersection Gender and Ethnicity has the intersection values Female - Asian, Female - URM, Female - White, Male - Asian, Male - URM, Male - White. The intersection value Female - Asian is an individual intersection (people group).

Choosing an intersection and creating an intersection (if necessary) are prerequisites to configuring the VIBE Index topic in People Analytics.

Getting Started

Review all details for choosing intersections to determine which fields and conditions meet the business and development needs of your organization. Afterward, review the details for the default intersection field and the field specifications for intersections. You can then determine if:

Using the default intersection field or an existing field in your tenant is a suitable choice for an intersection.
Creating an intersection is the best choice for your organization.

We strongly recommend that you use a minimum and maximum of 2 fields to define the intersection. Example: Gender and Ethnicity.

The best intersections are intersections that align with the general (more broad) business and development strategy of your organization. This enables you to connect insights that are surfaced in the VIBE Index topic in People Analytics with other internal findings that you can act on.

The VIBE Index is a great tool for diagnosing areas for targeted intervention. Therefore, the fields you choose for the intersection should consist of dimensions in which you can dig deeper into using other data sources (VIBE Central dashboard or other reporting sources).

The VIBE Index evaluates parity in several outcomes across intersections. The intersection you choose should have a reasonable expectation of parity across all outcomes (Attrition, Hires, Leadership, Promotions).

Examples of choices that meet this criteria:

Gender / Ethnicity - Organizations often look at gender and ethnicity to determine bias in outcomes such as hires and promotions. We expect that the gender or ethnicity of an employee shouldn't have an influence on the rate at which the employee is promoted or on the probability of the employee being represented in leadership, and so on.
Sexual Orientation / Disability Status / Veteran Status - Organizations often look at these dimensions to determine bias across different outcomes. Therefore, we expect that these employee attributes shouldn't have an influence on certain outcomes, such as employee promotions. Before choosing an intersection with 1 or more of these fields, determine if you have:
Accurate and relatively complete data among the chosen dimensions.
Sufficient headcount in each intersection (minimum 10 workers). Example: The intersection Disability Status and Veteran Status includes the intersection NotDisabled - IsVeteran, which consists of 10 workers.

Example of choices that don't meet this criteria: Age Group / Generation

Although organizations often look at age and generation as important variables in business and development work, they aren't dimensions that you would expect parity in for some of the outcomes included in the VIBE Index. Examples of outcomes where we don’t expect parity for age group and generation:

Hires - This outcome usually skews in volume towards early-in-career roles, which results in uneven representation by age group or generation. Therefore, we expect disparity across intersections for this outcome.
Leadership - This outcome is often overrepresented by older age groups. We expect that the percentage of leaders should vary by age group or generation. Therefore, we expect disparity across intersections for this outcome.
Promotions - For the same reason job levels and management levels skew by age group, promotion rates are often lower as you move higher in an organization (longer time to promotion). We expect that the percentage of promotions should vary by age group or generation, and therefore expect disparity across intersections for this outcome.

Example of dimensions that might or might not be a good choice for an intersection, depending on your organization: Region / Location

If your organization is mostly US-based with only certain types and levels of specific roles that are subordinate to other locations or regions, you might not expect there to be equal representation in leadership across regions, equal promotion rates, and so on.

If operations in your organization are currently more homogeneous across regions or are intended to be more homogeneous across regions, you might expect parity across leadership opportunities, hiring, and promotions.

Attributes

Unlike some personal attributes (mentioned in the section Expected Parity), job-related attributes almost always have an impact on other job-related outcomes and don’t make good choices for intersections.

Examples of job-related attributes:

Job Level or Management Level - We don’t recommend using these attributes for an intersection because we expect they are:
Directly related to the outcome Leadership.
Highly correlated with promotion rates.
Skewed along Hiring.
Performance Rating - We expect this attribute to have a negative impact on the outcome metric Hiring because it's often:
Highly correlated with promotions.
Highly correlated with differences in belonging and attrition.
Skewed by job level.
Tenure or Time in Role - We expect that tenure has strong correlations with promotions, attrition, and leadership representation. Therefore, we don’t recommend using these attributes for an intersection because they might negatively affect the outcome metric Hiring.

The intersection that you use for VIBE should include the right populations. If a field that is included in the intersection returns a blank or null value for a worker, Workday doesn't include the associated worker in the VIBE calculation.

Additionally, if the headcount for a given intersection falls below the minimum requirement of 10 workers for the respective month, we don't include the intersection in the VIBE calculation.

If the fields that you choose for the intersection consist of both correct and incorrect populations, you can use condition functions to exclude the incorrect populations.

The minimum headcount requirement for each intersection within a defined intersection is 10 workers. People Analytics evaluates active headcount at each recurrent data refresh.

Example: The intersection Gender and Ethnicity consists of the intersection Female - Asian, which has a minimum headcount of 10 workers for the latest month that People Analytics analyzes data.

The minimum headcount enables us to derive aggregate-level insights. The VIBE Index relies on the expectation of parity across groups, not individuals. By having a sufficient population in each intersection, there's a reasonable expectation that individual-level variances cancel out, and we can make comparisons on group-level outcomes where we would expect parity.

The minimum headcount enables the VIBE Index to be stable and actionable. Otherwise, there would be extreme fluctuations in results from time period to time period.

Example: If an organization has a headcount of 5 in an intersection and has a 10% annual attrition rate, the expected number of attrition events for that intersection in a 12-month period is 0.5. Because you can’t have 0.5 attrition events, an organization can be perfectly at parity by having 0 attrition events in one period and 1 attrition event in the next period. However, in the first case (0 attrition events), the intersection with a headcount of 5 would be 100% below parity for the corresponding time period. In the second case (1 attrition event), the same intersection would be 100% above parity for the corresponding time period. In either time period, interpreting the attrition outcome for this group would be misleading and not actionable.

Field Specifications for Intersections

The table lists the requirements for intersections and provides guidance on key configuration elements of intersections.

The default intersection for the VIBE Index is the Workday-owned, text field Intersection 1. To view the setup and details for the default field, access the

Business Object Details

report (secured to the

Custom Report Administration

and

Custom Report Management

domains in the System functional area) and search for the business object Worker.

The table lists the components of the default field Intersection 1.

You can’t edit the Workday-owned, default intersection field or the components within the default intersection field. If you want to use a similar setup for your intersection, review the field specifications for intersections and the details for creating intersections. You can then determine if you can use an existing field in your tenant for an intersection or if you need to create an intersection.

When creating an intersection field, you can:

Use an existing field or fields in your tenant that meet the specifications for intersections.
Create a new field that meets the specifications for intersections.

For examples on creating intersections, you can view the setup of the default intersection field and the example topic on creating an intersection.