Guide to the PBOT Equity Matrix

Guide
Screen capture of the PBOT Equity Matrix, showing a map of Portland with each census tract a different shade of purple depending on ranking in the index.
The Portland Bureau of Transportation (PBOT) uses a simple ranking index called an Equity Matrix to help make decisions on projects and programs. This map uses data on race, ethnicity, and income to apply a score to census tracts. It also uses data on limited English proficiency for more context.
On this page

About the Equity Matrix

PBOT launched its Equity Matrix in 2017. It was developed based on national equity best practices and guidelines from Portland’s Office of Equity and Human Rights (OEHR). Thus, it measures race, ethnicity, and income and uses these measurements to rank census tracts based on these factors. Limited English proficiency (LEP) data is used to provide additional context, highlighting census tracts where the percentage of households with limited English proficiency is above the city median. Read more on OEHR’s Language Access page.

The Equity Matrix gives PBOT staff a critical tool to use. Furthermore, it helps us embed the work of transforming systems that have excluded and oppressed people. It helps us make these systems more equitable. We work to ensure our staff know how to use it and apply it consistently across our various work groups.

Although this map was developed for use by staff, PBOT has made the map public. This lets us be transparent with our budget committee, various advisory groups, as well as the general public about the factors we use to determine when and where to allocate resources.


How to use the Equity Matrix

PBOT’s Equity Matrix assigns a score (maximum of 10) to every census tract using the demographic variables of race, ethnicity, and income.

Race and ethnicity are combined for a rating on the index from 1 to 5. For the race and ethnicity score, the higher the percentage of residents who self-identified in the U.S. Census as people of color or Latinx (of any race) in a census tract, the higher the score. People of color are defined here as any person who self-identifies as American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or other Pacific Islander, two or more races, other, or Latinx (of any race).

Income is also ranked from 1 to 5. The lower the median income in a census tract, the higher the score.

Why do we only score based on race, ethnicity, and income? Data is much more inconsistent and unreliable for limited English proficiency (LEP). PBOT illustrates LEP in the Equity Matrix on its own chart as well as overlaid with the scoring method outlined above. In the Equity Matrix, we have placed a bold edge around the census tracts where we find LEP higher than the citywide average of 3.8%. This helps tell part of the story and lets us see where these numbers overlap with our race and income scoring.

Explore the Equity Matrix on your own at the link below:

PBOT Equity Matrix

Use the plus (+) and minus (-) buttons to zoom in and out. Use the magnifying glass search icon (🔍) to look up addresses or street names. Click on any census tract to see census data and scoring for that tract.

For technical assistance with the Equity Matrix, email pbotgissupport@portlandoregon.gov.


Data sources and methodology

Data for the Equity Matrix is taken from the U.S. Census, specifically the 2014-2018 American Community Survey (ACS) 5-year estimates.

It’s important to note that the 2010 Census did not collect data on disability, and the sample sizes for ACS data are too small for us to map disability effectively at a census tract or neighborhood level. For this reason, we chose variables with the highest correlation to disability to ensure we were capturing this demographic as well.

PBOT uses the Jenks “natural breaks” classification method for scoring in the Equity Matrix. Natural breaks is a statistical clustering method that groups the most similar values in a set together by minimizing the variance within a class and maximizing the variance between classes. Resulting groups don’t always have the same number of members. However, members of these groups are more similar to one another and more different from the members of the other classes than would be achieved with a simple quintile method.

The race and ethnicity index is scored on an ascending scale where the census tracts with the lowest percentage population of people of color or Latinx of any race receive a score of 1 and those with the highest percentage of those populations receive a 5. The income index is scored on a descending scale where the census tracts with the highest median household income receive a score of 1 and those with the lowest median income receive a 5. Using this method, with scores on race/ethnicity and income separate, the lowest possible combined score is 2 points (race and ethnicity index = 1, income index = 1) while the maximum is 10 (race and ethnicity index = 5; income index = 5).

For technical assistance with the Equity Matrix, email pbotgissupport@portlandoregon.gov.


Additional data considered

PBOT’s Equity Matrix is one metric staff use to evaluate projects. Staff must also keep in mind the goals laid out in Portland’s Racial Equity Plan, Climate Action Plan, the 2035 Comprehensive Plan, the Transportation System Plan, and Vision Zero Action Plan.

For this reason, PBOT may score projects using additional data and factors. This includes, but is not limited to:

  • Additional demographic indicators specific to a project or project area
  • Safety factors and data such as crashes, crime, existence (or lack of) sidewalks and other infrastructure
  • Accessibility as it relates to pedestrian and biking networks as well as transit
  • Environmental impacts
  • Health impacts
  • The level of community support
  • Community benefits agreements
  • Cost-effectiveness

PBOT staff may use the Equity Matrix as a standalone tool or in conjunction with other variables such as those listed above. The process of ranking projects within the Transportation System Plan is a good example of using an equity index in combination with other weighted factors.


Why lead with race and ethnicity?

The City of Portland, as well as the Government Alliance on Race and Equity (GARE), lead with the metrics of race and ethnicity because of the very inequities created and perpetuated by institutions such as government. These inequities, across all indicators for success, are deep and pervasive. Focusing our work on racial and ethnic equity allows us to introduce a framework, tools, and resources we can then apply to other forms of systemic exclusion and institutional oppression. This may include discrimination, systemic exclusion, or institutional oppression based on gender, ability, age, or sexual orientation.

At PBOT, we strive for:

  • Specificity. One-size-fits-all strategies rarely succeed. Strategies to achieve racial and ethnic equity differ from those to achieve equity in other areas.
  • Clarity. A racial and ethnic equity framework must be clear about the differences between individual, institutional, and structural racism, as well current and historical inequities.
  • Unity. Race and ethnicity can be an issue keeping other systematically excluded and institutionally oppressed people from effectively coming together. An approach that recognizes the intersectional ways such exclusion and oppression takes place will help achieve greater unity across communities.

Prior equity measures at PBOT

In the past PBOT has used different equity measurements for different projects and programs. These were all quite different and resulted in different outcomes:

  • Vision Zero. PBOT’s Vision Zero program uses an equity index designed by TriMet which combines 10 different demographic variables into one conglomerate index. This index has been embedded into the current Vision Zero Action Plan and will remain the matrix used by Vision Zero.
  • LED Streetlight Conversion. This PBOT project hired a consultant, the Coalition for a Livable Future, to design a simple equity ranking index that used four demographic variables: race, income, and both elderly and youth populations. This matrix was used for this project to help with mapping and the timeline of the streetlight conversion. With that work complete, this index is no longer used.
  • Transportation System Plan (TSP). Planners working on the TSP designed an equity index for the plan that looked at the same four demographic variables the LED Streetlight Conversion project did: race, income, and both elderly and youth populations. This index was more robust, however, and was vetted through the Office of Equity and Human Rights to ensure it could be used on the TSP.
  • Safe Routes to School. This PBOT program was the first at the bureau to use the three demographic variables of race and ethnicity, income, and limited English proficiency (LEP) which has become best practice at the city. The index Safe Routes used was designed in-house and became the basis for PBOT’s current Equity Matrix which the Safe Routes team uses currently.