Methods for creating an index map for social equity
Introduction
All over the world, organizations and government agencies are recognizing the profound impact of policies that have systematically denied certain populations from opportunities and resources. The increased commitment to advance and correct social equity with regards to race, ethnicity, gender, class, disability status, sexual orientation, and other social and economic lines has resulted in a growing interest to optimize interventions informed by GIS.
Index maps are an accessble tool for such workflows. An index is a composite value of multiple indicators to provide a reader with an ordered measurement within a given sample size.
2018 Social Vulnerability Index (SVI) . Created by the Centers for Disease Control and Prevention (CDC) / Agency for Toxic Substances and Disease Registry (ATSDR) / Geospatial Research, Analysis, and Services Program (GRASP).
Combining and analyzing multiple social, environmental, and economic factors and summarizing them into a single measure for prioritization provides an understandable and attractive way for organizations and governments to propose changes in policies, practices, and procedures.
However, if an index is developed in a way that does not account for the lived experiences of the community, it runs the risk of inadequately or even harming efforts to advance raciala and social equity. Therefore, it is important to recognize that there is no one-size-fits-all when it comes to creating an index. The methodology selected must be intentional and decided based on the specific geography, context, and input from the impacted community.
This story explores several methods for developing an index and examples currently used in real world scenarios.
A list of ArcGIS Lessons that explore creating an index map is provided at the end of this resource.
Similarity Search
The ArcGIS Pro tool Similarity Search identifies features are most similar (or dissimilar) to one or more reference features based on feature attributes.
Similarity Search tool can be used to find other cities or areas within a city that are just like a particular city or neighborhood in terms of population, education, and proximity to specific recreational opportunities. For example, an after-school fitness program was extremely successful in one neighborhood. Promoters want to find other neighborhoods with similar characteristics as candidates for program expansion.
In an equity example, Similarity Search tool can be used to identify census tracts that at higher risk for cumulative social burdens. It might be areas of a city where there is higher than average rates of poverty, low high school graduation rates, and access to healthy food options. The tool output includes an ordered rank value for each census tract in an entire jurisdiction that indicates which census tracts have the most cumulative social burdens.
See How Similarity Search works to learn more.
Social Equity Analysis Solution
The Social Equity Analysis solution delivers a set of capabilities that help you understand community characteristics, analyze community conditions and actions, and generate an equity analysis index that can be used to educate internal and external stakeholders.
The Identify Community Characteristics and Create Equity Analysis Index tools in the solution use Similarity Search to compare the input features to the highest focus case based on the input variables. All input variables are equally weighted.
Attribute field(s) identified as Focus Factors (High to Low focus variables) are analyzed to see which have a higher value and those identified as Focus Factors (Low to High focus variables) are analyzed to see which have a lower value.
To prepare the data for the Similarity Search, first "High to Low" (high value means more focus/risk) input field values are normalized to be between 0 to 100. If the input fields are "Low to High" (low values mean more focus/risk), the same formula is used and then converted to an inverse score value.
A single sample polygon feature layer is then created with the highest 0 to 100 score (i.e. the highest focus/risk). This polygon is the reporting area which has the highest preponderance/concentration of focus variables.
The tools rank the reporting areas by seeing how similar they are to this highest focus area. The highest focus area gets a rank of "1". All other reporting areas are assigned a focus rank score in order of how similar they are to this highest focus area by comparing the attribute values of the focus variables to that of this highest focus area.
Considerations
Similarity Search is an accessible way to create an index rank for a set sample size. With the Social Equity Analysis solution, the steps to create an index using Similarity Search are even more streamlined.
However, this method of index creation is not able to account for the degree of difference for each attribute or is there a way to add a weighted value for specific indicators. Therefore, it is important for the GIS analyst to consider what is most appropriate for your given community, intervention, and the distribution of values for each indicator required for the index.
Standard Deviation
Another way to develop an equity index is to reclassify each attribute value for each feature by standard deviation and calculating the sum value of the reclassified values. Depending on your project target area and intervention, it may be valuable to develop an index using standard deviation groups if the degree or magnitude of variation for indicators is important to account for.
The standard deviation classification method shows you how much a feature's attribute value varies from the mean.
The mean and standard deviation are calculated automatically. Class breaks are created with equal value ranges that are a proportion of the standard deviation—usually at intervals of one, one-half, one-third, or one-fourth—using mean values and the standard deviations from the mean.
Reclassify Field tool
The Reclassify Field tool reclassifies values in a numerical or text field into classes based on bounds defined manually or using a reclassification method. When Class ranges are created using a number of standard deviations above and below the average, it can help you understand where values lie in a distribution. For example, you can reclassify rainfall by one standard deviation to identify areas with rainfall greater than two standard deviations from the mean.
To create an index, you will run each indicator you want to include in your index through the Reclassify Field tool.
In the Reclassify Field tool pane, for Reclassification Method, choose Standard Deviation. For Output Field Name, name the new field in a way that will help you easily identify the field for calculation later.
The tool will also automatically add “_CLASS” at the end of the fields with the generated class values.
When you have run the Reclassify Field tool for all your indicators, you will create a new field to represent the values for your index.
Use the Calculate Field tool and Arcade to build an expression to sum the reclassified fields.
Set the symbology for the layer to Graduated Colors for the Standard Deviation Index field. Ensure the color ramp values are set so the larger the value has a darker color.
Additional Standard Deviation methods
Sum of positive standard deviation classes
Instead of adding all the reclassed values of the indicators, you can create an index that only sums the reclassed indicators that were within a positive standard deviation range (0.50 < Std. Deviation) for each feature.
The Reclassify Field tool also created fields that end with “_RANGE”, which allows you to see the standard deviation ranges and values for each indicator.
Count of positive standard deviation classes
Another method is to count the number of indicators that were in the a positive standard deviation range (0.50 < Std. Deviation) for each indicator within a feature.
Considerations
Creating an index using standard deviation values accounts for the degree of differences in values across each indicator, which is helpful for capturing cumulative impacts of multiple indicators on a particular issue. It will be important to consider whether this method will be appropriate if there are indicators with non-normal distributions.
Quantile
The final method that will be discussed in this story to develop an equity index is to reclassify each attribute by quantile and calculating the sum or average of the reclassified values. Depending on your project target area and intervention, it may be valuable to develop an index using quantile classes if the primary goal is to identify a prioritized group, regardless of degree of differences within an indicator.
In a quantile classification , each class contains an equal number of features.
A quantile classification is well suited to linearly distributed data. Quantile assigns the same number of data values to each class. There are no empty classes or classes with too few or too many values.
Because features are grouped in equal numbers in each class using quantile classification, the resulting map can often be misleading. Similar features can be placed in adjacent classes, or features with widely different values can be put in the same class. You can minimize this distortion by increasing the number of classes.
Reclassify Field tool
For the quantile method, the Reclassify Field tool will also be used, but this time, you will set the Reclassification Method to Quantile . By doing so, the tool will reclassify each indicator into by an equal number of values in each of the specified Number of Classes parameter values. For example, if there are 50 values and the number of classes is 5, each class will have 10 records. This method is useful when you want to understand where each value falls in the ranked values. For example, you want to understand the locations in which average annual income falls in the top and bottom of 10 quantiles.
To create an index, you will run each indicator you want to include in your index through the Reclassify Field tool, setting the Reclassification Method to Quantile. For Number of Classes, choosing 10 will make it simpler for assessing percentile. For the Output Field Name, name the new field in a way that will help you easily identify the field for calculation later. The tool will also automatically add “_CLASS” at the end of the fields with the generated class values.
Sum of Quantile Reclassed Values
When you have run the Reclassify Field tool for all your indicators, you will create a new field to represent the values for your index.
Use the Calculate Field tool and Arcade to build an expression to sum the reclassified fields.
Average of Quantile Reclassed Values
You can also calculate a quantile index by averaging the reclassed values. To do this, you would create a new field for the index value and use the Calculate Field tool and Arcade to create an expression that averages the reclassed quantile values for each indicator.
Considerations
Creating an index using quantile classes is useful for identifying a localized prioritized ranking scale, without needing to be concerned with outliers. But it my not be appropriate for indicators with non-normal distributions.
Compare index maps by methods
Explore the app below to compare the different index maps produced by the methods discussed in this story.
Compare index maps created with methodologies discussed in this story.
Additional index methods
The methods and use of index maps are already widely used in a variety of applications, including government agency decision making processes. Here are a few examples of other methods local governments are using to create equity indices.
CalEnviroScreen
State of California
CalEnviroScreen is a screening tool used to help identify communities disproportionately burdened by multiple sources of pollution and with population characteristics that make them more sensitive to pollution.
"Disadvantaged communities in California are specifically targeted for investment of proceeds from the State’s cap-and-trade program. CalEPA designated the top 25 percent of census tracts in CalEnviroScreen 4.0 as disadvantaged communities in May 2022, among other categories, for the purpose of investing cap-and-trade proceeds."
CalEnviroScreen Methodology
The overall CalEnviroScreen score is driven by indicators that account for pollution burden, such as environmental effects and exposures, and population characteristics, such as socioeconomic factors and sensitive populations.
A formula to calculate scaled component scores of each indicator group results in the final CalEnviroScreen Score, which is a percentile value.
MiEJscreen
State of Michigan
Note: An error message map appear in the app to the right. Click OK to close the window to continue exploring MiEJscreen.
"MiEJscreen is an interactive mapping tool that identifies Michigan communities that may be disproportionately impacted by environmental hazards. The map allows users to explore the environmental, health, and socioeconomic conditions within a specific community, region, or across the entire state.
These data sets can be viewed individually or combined into a final MiEJscreen score that allows users to understand how communities experience environmental justice impacts relative to others. These results are depicted in the form of maps so that different communities can be compared to one another.
A census tract with a high score is one that experiences higher pollution burden and vulnerability than census tracts with low scores. MiEJscreen ranks census tracts based on data that are available from state and federal government sources."
MiEJscreen methodology
The MiEJscreen methodology is very similar to CalEnviroScreen. It also divides indicators into categories to calculate composite scores for Environmental Conditions and Population Characteristics, resulting in the final MiEJscreen score.
Denver Human Services Index
County and City of Denver, CO
Click the story to the right to explore the Denver Human Services Index and indicators.
At Denver Human Services (DHS), we envision a healthy community where people are connected, supported, safe, and well. This updated Denver Human Services Index for 2022, aggregates 16 key indicators by neighborhood into one summary map, which can be used by DHS decision-makers and community partners to inform programs, practices, services, and investments across the Denver community.
Denver Human Services Index methodology
For the Denver Human Services Index, each indicator was transformed to a standardized (z-score) value. The index rank value is the sum of the standardized indicator score.
Try an index lesson
Ready to learn how to make an index map? Try one of these Learn ArcGIS lessons that will walk you through a real-world scenario to create an equity index.
Create a social equity index to improve public health
Use the Social Equity Analysis solution to apply an equity lens and prioritize schools to host a parent health education program. Time: 45 minutes
Customize a climate resilience index
Map where heat health interventions will be most impactful in your community by adding demographic variables that fit your local context.
Time: 45 minutes
Shade equity - Determine tree planting locations with suitability analysis
Use site suitability analysis to find optimal locations for tree allocation, promoting social equity and reducing the impact of climate change. Time: 1 hr, 30 min
Analyze COVID-19 risk using ArcGIS Pro
Create risk maps for transmission, susceptibility, resource scarcity, and risk profiles for targeting intervention areas.
Time: 1 hr Start the lesson