ODOT Social Equity Index
2023 Update

1 Introduction

The Oregon Department of Transportation’s Strategic Action Plan highlights Social Equity as a core priority for the agency to focus its projects, policies, and programs toward. As a part of recognizing Social Equity as a core priority, the agency developed the Social Equity Index (SEI) to help agency staff and leadership understand where communities of concern are located throughout Oregon. Using U.S. Census data at the block group level, the SEI aims to be a decision support tool that helps to target agency resources in a way that reduces social disparities related to transportation resource allocation. This report documents the data and methods used latest SEI for use by ODOT staff. You can explore the SEI in the dynamic map near the end of this document, or go to ODOT’s TransGIS page here and search under the Planning and Climate Change Resilience category. To pull the data into a GIS application data you can use the service here.

2 Data and Methods

Data to inform the SEI comes from the U.S. Census Bureau, specifically the American Community Survey (ACS). Sociodemographic data primarily drives of the SEI and the current index also uses household vehicle ownership. These data are described in more detail below. In addition to documenting input data, this section documents the methods used to combine the inputs into a composite index including a description of how SEI categories are created. This documentation builds understanding of how the SEI was constructed so that users can describe what the SEI means and future updates can be more easily performed.

2.1 Census Data

This update of the SEI uses the American Community Survey 5-year sample, a data product of the U.S. Census, including the attributes described in the table below. These Census data elements are acquired from the Census using an R script using the Census’ API and is available on Github here. This script requires an API key that can be retrieved here for free but should otherwise work as long as all the necessary R libraries are installed. This update of the SEI uses the 5-year ACS representing years from 2017 to 2021 and uses the 2020 census geographies. In the previous 2010 census Oregon was represented by 2,634 block groups but with more population those block groups are split and new ones added so in the current Census Oregon is represented by 2,970 block groups. The current SEI utilizes the latter spatial geometries represented by 2020 Census geographies.

Table 2.1: Census Data Source Tables and Geographic Levels

Data Element	Table Name
Population	B01003
Population living at 200% poverty or below	C17002
Population age 20 to 64 w/ disability	B23024
Limited English proficiency	B16004
Population age 65 or older	B01001
Population 18 years old or younger	B01001
Population white	B03002
Households w/ 0 vehicles	B25044

For the purposes of this report the following Census data variables will be used and referred to as the “input variables”. These data are used directly from Census by simply dividing the value for the variable by the block group total population with the exception being the POC input variable, which are derived from Census data by subtracting the White population from the total population. A note for the Disability 20 to 64 population: this proportion is calculated by dividing by the population aged 20-64 and not the total population. Each of the input variables and their calculation methods are summarized in the table below.

Table 2.2: Input Values for Social Equity Index Development

Data Element	Description	Calculation Method
Poverty %	Percent of population living at 200% of poverty or below	Direct from Census
POC %	Percent of population that are Persons of Color (POC)	Total population minus white population
Limited English Proficiency %	Percent of population that speak English ‘not well’ or ‘not at all’	Direct from Census
Disability (20-64)	Percent of population age 20-64 that have a disability	Direct from Census
Age Over 64	Percent of the population that are 64 years or older	Direct from Census
Youth Population	Percent of population that are 18 years old or younger	Direct from Census
Vulnerable Population %	Percent of population that is 64 years or older, under 18, or age 20 to 64 and have a disability	Added 65 & older, 18 and younger and 20 to 64 with disability
Zero Vehicle Households %	Percent of households that do not own a vehicle	Direct from Census

2.2 Index Methods

This section aims to describe the approach used to create the SEI simply, but any indexing method requires some complexity. Different approaches were tested using various ACS data elements with the final index method and data inputs determined based on a combination of expert review and analysis of the transportation outcomes associated with the final index categories. The method uses quintiles to bucket each input value and then add those quintiles into a composite score which is then categorized into Low to High categories using quantiles. Quintiles are statistical measures used to divide a dataset into five equal parts (quintiles) with each part representing 20% of a given population value, so the first quintile represents the lowest fifth of the data (1% to 20%); the second quintile represents the 2nd fifth (21% to 40%) and so on up to 100%. Each of the input values (Poverty, POC, etc.) have their quintiles computed and used to inform breaks of the data where the first quintile is then assigned a 1, the second quintile a 2, and so on for each of the next 3 quintiles with the highest quintile being assigned a 5. Quintiles for each of the input values are then added resulting in a score of 5 to 25 for an index with 5 variables as input. A block group with a score of 5 would mean that that block group has input variables that all fall in the bottom 20% of observed values for all the input variables. A score of 25 would mean that that block group ranks in the top 20% for all input variables used in the index.

The quintile scores are added using the below formula:

\[ \begin{aligned} SEI_{i} = Poverty Prop_{i} + POC Prop_{i} + Limited English Prop_{i} + Vulnerable Population Prop_{i} + Zero Vehicle Hh Prop_{i} \end{aligned} \]

*where:

\[ \begin{aligned} Vulnerable Population Prop_{i} = Youth Prop_{i} + Disability Prop_{i} + Population65Older Prop_{i} \end{aligned} \]

To demonstrate how these input values are represented by quintiles the following figure is presented. The values used in the index and represented in the figure below are proportions of the total population that are composed of the specific sub-population, e,g, percentage of the population living at or below 200% of the poverty federal poverty line.

Figure 2.1: Distribution of Input Census Variable by Quintile

The final composite index scores are then categorized using quartiles. The distribution of these composite scores and the index category are displayed below and show the minimum and maximum score ranges for each block group’s composite score (each point represents a block group’s composite score). For the Low category the composite index scores range from 9 to 11, with the High category ranging from 18 to 25. Note that the chart jitters each observation making some points appear to slightly exceed these reported values.

Figure 2.2: Distribution of Final Composite Scores by SEI Category

3 Mapping the Social Equity Index

This section applies the data and methods described above to a map of Oregon’s Census block groups resulting in a map of the SEI. This geospatial data can then be used by ODOT staff and leaders as a decision support tool. Of the 2,970 block groups in Oregon, 14 include no population because they either are water or land with no population so are removed from the geospatial data.

Figure 3.1: ODOT Social Equity Index (2017 - 2021)

4 SEI and Transportation Outcomes

To further demonstrate how the SEI classifies Census block groups and how other data elements relate to the final SEI categories the table below summarizes average values for various measures. These measures include three broad categories including sociodemographics, travel and the built environment, and crash injury along three categories including all modes, pedestrian and bicycle injuries. Following the table are sections describing how the data for the table were developed.

4.1 Population and Sociodemographic Outcomes

From the population and sociodemographic measures its clearer to see how the SEI uses the input variables to categorize the block groups. For example, the average poverty rate in block groups classified as High SEI have on average 45% of the population living in poverty whereas Oregon as a whole the rate is only 29% while the block groups categorized as Low have on average just 15% of the population living in poverty. This gradient is observable with other sociodemographic factors like the percentage of population that are people of color, disabled, have limited English proficiency or live in crowded households. There are roughly 870,00 Oregonians living in 576 block groups categorized as High SEI, which is about 21% of the total population.

4.2 Travel and Built Environment Outcomes

Measures summarized in this section aim to reflect differences in traffic exposure and travel options and preferences and show differences across the SEI categories. Traffic proximity and volume is highest in High SEI block groups, as is arterial VMT density and high-speed arterial density. These measures highlight the disparity in exposure to vehicle traffic either as a motorist or nonmotorists. The traffic proximity measure also aims to highlight the likely disparity in exposure to air toxics across the SEI. Transit stop density is higher in High SEI areas indicating higher transit usage and likely higher pedestrian traffic as people access transit. Lastly, the percentage of workers who commute to work by walking, biking and transit is higher in High SEI block groups compares with the state average with 11% and 9% of the work force commuting by those modes respectively.

4.3 Crash Outcomes

Research has established that in the United States disparities exist in crash outcomes and measures such as income, educational attainment and race. ODOT’s Research Unit documented the disparities in Oregon for pedestrian injury by race using Fatal Analysis Reporting Data. The research further analyzed factors at the Census tract level affecting the likelihood of a pedestrian injury. This research concluded that race and income along with built environmental factors and exposure measures such as traffic density, bus stops density, and the percentage of the workforce commuting by transit were all correlated with pedestrian injuries. Less is known about how sociodemographic factors affect traffic injury outcomes in general in Oregon. The All Mode Traffic Injury rate based on injuries per VMT and population are presented in Table 4.1 below. Though no disparity exists between the SEI categories for All Mode fatal and serious injury a disparity is present for the measure of all injury.

Table 4.1: SEI and Transportation Outcomes

Measure	Low	Low/Medium	Medium/High	High	Statewide
Population & Sociodemographics
Poverty % (200% & less)	15.2%	26.5%	35.7%	44.7%	28.7%
BIPOC %	14.6%	19.8%	27.8%	41.6%	24.3%
Disabled Pop. 20-64	8.5%	12.8%	14.3%	16.9%	12.6%
Limited English Prof. %	0.3%	0.8%	2.3%	7.0%	2.2%
Age Over 64 %	19.1%	20.5%	19.1%	16.5%	19.0%
Crowded Housing %	1.5%	2.5%	3.8%	6.1%	3.2%
Youth Population %	14.1%	13.9%	13.6%	16.9%	14.5%
Vulnerable Population %	38.3%	41.9%	41.2%	43.4%	40.9%
Population	1226341	1156302	954603	869931	4207177
Urban Block Group Count	571	560	508	490	2129
Rural Block Group Count	318	270	153	86	827
Block Group Count	889	830	661	576	2956
Travel & Built Environment
Traffic Proximity and Volume	106.35	157.66	209.17	269.60	175.57
Arterial VMT Density (per sqmi.)	18.84	25.70	32.74	46.67	29.29
High Speed Arterial Density (per sqmi.)	0.23	0.33	0.39	0.63	0.37
Transit Stop Density	12.03	14.80	18.91	20.77	16.05
Zero Vehicle Hhs %	1.6%	5.3%	9.7%	13.6%	6.8%
Workers Commuting by Bike/Walk/Transit %	6.4%	8.4%	10.4%	10.5%	8.7%
Workers Commuting by Bike/Walk/Transit	37914	46990	47495	39911	172310
All Mode Traffic Injury
All Mode Fatal & Serious Injury	3141	3153	2373	2189	10856
All Mode Injuries	38409	40686	38345	43202	160642
All Mode Fatal & Serious Rate (Pop.)	366.2	347.9	332.6	296.9	340.0
All Mode Injury Rate (Pop.)	4392.1	4323.3	4842.5	5650.6	4718.7
All Mode Fatal & Serious Rate (VMT)	4.1	4.2	4.1	4.0	4.1
All Mode Injury Rate (VMT)	51.2	57.8	67.5	79.6	62.3
Pedestrian Traffic Injury
Ped Fatal & Serious	159	208	227	314	908
Ped All Injury	621	939	1092	1435	4087
Ped Fatal & Serious Rate (Pop.)	16.7	23.6	27.7	41.3	25.9
Ped All Injury Rate	67.1	110.4	134.1	193.3	118.8
Ped Fatal & Serious Rate (VMT)	0.2	0.3	0.4	0.6	0.4
Ped All Injury Rate (VMT)	1.1	1.9	2.3	2.9	2.0
Bicycle Traffic Injury
Bike Fatal & Serious	67	70	91	69	297
Bike All Injury	577	813	844	911	3145
Bike Fatal & Serious Rate (Pop.)	6.9	7.4	11.2	9.2	8.4
Bike All Injury Rate (Pop.)	84.5	88.9	102.2	125.9	97.8
Bike Fatal & Serious Rate (VMT)	0.1	0.1	0.1	0.1	0.1
Bike All Injury Rate (VMT)	1.2	1.4	1.8	1.9	1.5

5 Supplemental Description

This section describes the sources and process used to develop the supplemental data featured in Table 4.1 above.

5.1 Crash Data

One of the uses of these Census data has been to analyze the relationship of sociodemographic information and traffic injuries. In order to understand how the SEI relates to the frequency and population-based rate of traffic injury ODOT Crash data will be used. Since the Census data used represents a 5-year sample of the population this report uses 5 years of crash records data will also be used in each analysis period. These data are derived from citizen filed crash reports and police reports which are then processed and augmented by ODOT’s Crash Analysis and Reporting Unit (CARS) in order to produce the state’s official traffic injury data.

Crash data are joined to the block group using spatial joining function in R which relies on highly precise spatial location of the crash which then places the crash point in just one block group. Many crashes occur on roadways that coincide with block group boundaries as many Census geographies use roadways to inform boundaries so this method of spatial joining is imperfect. However, ODOT’s Research Unit documented how this issue is unlikely to significantly impact analysis using block group level data, especially when grouped into an index due to spatial autocorrelation of Census data. Spatial autocorrelation, in this context, relates to the phenomenon of sociodemographic data in which values in one block group are correlated or similar to the values in a neighboring block group. The implications of this phenomenon for our analysis is minimal however for any statistical analysis would need to account for this using various techniques like geographically weighted regression or regression with mixed effects. The latter approach was employed in Roll and McNeil (2020) where multilevel modeling technique was used.

5.2 Traffic Expsoure Data Highway Performance Monitoring System Data

In the Travel and Built Environment Outcomes section above measures of traffic exposure including Traffic Proximity and Volume, Arterials VMT Density, and High Speed Arterials Density are featured and are used to represent exposure to motorized traffic. Exposure to vehicle traffic increases the risk of traffic injury and air pollution related health effects. Table 4.1 above includes summary information about traffic exposure including a measure from the Environmental Protection Agency’s Environmental Justice Index (EJI) and measures developed using the Highway Performance Monitoring System (HPMS). The traffic proximity measure from EJScreen is derived from HPMS but is preprocessed to the block group level while the other exposure measures were creating by mapping the HPM segments to block groups and counting only arterial vehicle travel and roadways with 35 mph posted speed.

5.3 Transit Stops

In order to understand how various indexing methods developed and tested in this report relate to transit usage data the location of transit stops are utilized as a proxy. ODOT maintains a database of transit stop locations based on the GTFS standard and copies of the data are stored every two years. Transit stops are an imperfect measure of transit usage with ridership or service miles being a better and more direct measure but a statewide database does not exist, so transit stop count and transit stop count density are the best available data. For the 2017 to 2021 analysis period 2017, 2019 and 2021 data are utilized. Similar to the crash data, transit stop location is utilized to determine which block group they are assigned to using a spatial join. Since stop location is quite precise this spatial join should be clean and introduce little mis-association.