Skip To Content

Create a policy map

A policy map is a map that shows areas where policy intervention should occur. You'll create a policy map that shows low birth weight by county across the United States. Then, you'll focus on the state of Colorado. This map will emphasize the best counties for intervention programs to improve newborn health.

The Centers for Disease Control and Prevention (CDC) set a goal that by 2020, no more than 7.8 percent of newborns will have a birth weight below 5.5 pounds. Your map will show counties that are close to reaching this goal and counties that are the biggest outliers, where intervention may have the most impact.

Add data to a new map

First, you need data. The ArcGIS Living Atlas of the World has lots of authoritative GIS data, so you'll search it for county data on birth weight.

  1. Sign in to your ArcGIS organizational account.
    Note:

    If you don't have an organizational account, you can sign up for an ArcGIS free trial.

  2. At the top of your organization home page, click Content. Click the Living Atlas tab.

    Living Atlas tab

    This tab contains available content from the Living Atlas.

  3. Click Search Living Atlas and type County Health Data.

    Living Atlas search bar

    The search returns a large number of items (92 in the example image; your results may vary), including web apps and image layers. You want to work with data rather than view a finished product, so you'll restrict the item type to feature layers.

  4. Collapse Categories and Regions. Under Item Type, click Layers and click Feature Layers.

    Item Type set to Feature Layers

    The search results are filtered. Far fewer results appear, and all that do are feature layers, which can be opened in a new map. One of the results, County Health Rankings 2018 by JenniferBell_UO, might have the data you want.

  5. Click the title of the County Health Rankings 2018 result.

    County Health Rankings 2018 search result

    The feature layer's details page opens. This page contains documentation about the data and its source. The data comes from County Health Rankings & Roadmaps, so it should be trustworthy and reliable. But does the layer have data on birth weight?

  6. Click the Data tab.

    Data tab

    The data attributes are displayed as a table. The table contains a lot of data, so you'll search a list of attribute field names for what you want.

  7. In the upper right corner, click Fields.

    Fields button

  8. In the search bar, type Birth.

    The search returns eight results. Four of the results are about low birth weight percentage. You'll be able to make your policy map with this layer. You'll bookmark the layer in case you want to return to it quickly later.

  9. Click the Overview tab. Below the thumbnail, click Add to Favorites.

    Add to Favorites

    Tip:

    To access your favorites, at the top of the page, click Content and click the My Favorites tab.

    One of the links in the description leads to a list of downloadable Excel spreadsheets. The 2018 County Health Rankings National Data spreadsheet is the one the layer is based on. If you download it and open the Ranked Measure Sources & Years tab, you'll learn that the birth weight data is for 2010 to 2016.

    It's important to understand your data and where it comes from before you use it in a policy map. Always read the details page or metadata. Now, you're ready to add the data to a new map.

  10. On the details page, click Open in Map Viewer.

    Open in Map Viewer button

    A new map opens, showing the county boundaries of the United States.

  11. If necessary, zoom to the contiguous United States (the states except Alaska and Hawaii).

    Contiguous United States

    Tip:

    Click a county to view that county's pop-up. The pop-up shows all the information in the layer associated with that county.

  12. On the ribbon, click Save and choose Save As.

    Save As option

  13. In the Save Map window, enter the following parameters:
    • For Title, type Low Birth Weight by County, 2010-2016.
    • For Tags, type Birth Weight, Newborns, and United States.
    • For Summary, type This map shows counties with a percentage of low birth weight above the CDC's 2020 goal of 7.8 percent.
    • Leave Categories unchanged.
  14. Click Save Map.

Discover nationwide patterns

Before you focus on your state of interest, you'll learn about some of the patterns nationwide to better understand the birth weight situation. When looking for patterns in your data, it's helpful to try styling the data in different ways.

First, you'll change the basemap. By default, the map uses the Topographic basemap, which was designed as a reference map. To emphasize your data, you'll use the Light Gray Canvas basemap, which has fewer topographic features.

  1. On the ribbon, click Basemap and choose Light Gray Canvas.

    Light Gray Canvas basemap

    The basemap changes. Next, you'll change the style of your data so that counties with higher and lower percentages of low birth weights have different colors.

  2. In the Contents pane, point to the County Health Rankings 2018 layer and click Change Style.

    Change Style button

  3. In the Change Style pane, for Choose an Attribute to Show, choose Percent Low Birthweight.

    Choose an attribute to show

    Because this field is a percent or rate, the color drawing style (instead of size or single symbol) is chosen by default and the counties are automatically styled with a high to low color ramp.

    Map styled by color

    The legend explains the distribution of the colors. In the darkest counties, over 10 percent of births have low birth weight. In the lightest counties, less than 6.1 percent do. Many counties in the southeastern part of the country have high percentages of low birth weights, as do several counties in New Mexico and Colorado.

    You can check out a histogram to better understand the distribution of values.

  4. For Select a drawing style, under Counts and Amounts (Color), click Options.

    The pane changes to show the histogram.

    Histogram showing default distribution of values

    The minimum value is 2 and the maximum is 27.

  5. Point to the x̅ symbol.

    This symbol indicates the average value, which is 8 percent. However, according to statistics published by the CDC, the national average in 2016 was actually 8.17 percent. Why is there a difference between what the CDC reported and the data?

    The average of your data treats all counties as equal, meaning it does not account for population differences between counties. The CDC's national average does, leading to the difference.

    The values of 10 and 6.1 on the side of the histogram represent one standard deviation away from the average. These values were assigned as the cutoffs for the darkest and lightest colors. While the default statistical classification of the data is okay, you can change the settings to uncover more patterns.

  6. For Theme, choose Above and Below.

    Theme changed to Above and Below

    The style of the map changes. Now, two distinct colors are used to depict values above and below the average (your colors may vary from the example images). The cutoffs for the darkest and lightest colors remain the same. However, a new handle has been added to the histogram with the value 8.1, an approximation of the average.

  7. Drag the 8.1 handle to 8.2 (or as close as you can get it).

    Adjusted histogram

    Tip:

    You can also adjust a handle by clicking its value and typing the number you want.

    The values of the handles on the histogram change and the colors of the counties adjust on the map. The central value is closer to the CDC average of 8.17.

  8. Drag the central handle to 7.8 (or as close as you can get it).

    This percentage is the CDC's 2020 goal for average low birth weight. By styling the map with this value as the average, you'll have a clearer idea which counties will need help to achieve the goal.

    The dataset has a long tail. That means that there is a large outlier (the maximum value of 27 percent). This outlier, and any other extremely high percentages, isn't styled any differently than counties that are only one standard deviation from the average. You'll change the data distribution to better view the counties that need the most help.

  9. Drag the top handle to 11.2 (or as close as you can get it).

    While dragging the central handle adjusted all three handles, dragging the top handle does not change the other two. Now, your map has more gradation between values above the CDC goal.

    Next, you'll change the color ramp. Your basemap is light, so you want a color ramp with a light center to draw attention to the darker extremes. Additionally, you want to deemphasize values below the CDC goal, because they aren't as relevant to your policy map.

  10. Click Symbols.

    Symbols button

  11. Choose the orange to gray color ramp (second row, third from the left).

    Orange to gray color ramp

  12. Click OK.

    Map styled to emphasize high values

    Now, the map visually indicates which counties are of highest concern.

  13. In the Change Style pane, click OK and click Done.

Configure pop-ups

When you click the data on your map, a pop-up opens with information about that data's attributes. It would be useful for your users if they could click a county and see the percentage of births in the county with low birth weight.

  1. Click any county on the map to open its pop-up.

    Default pop-up

    The default pop-up contains all the attribute information for the selected county. While the list includes the low birth weight percentage, you need to scroll down to find it. You'll configure the pop-up to show only the information relevant to your policy map.

  2. Close the pop-up. In the Contents pane, point to the County Health Rankings 2018 layer, click the More Options button, and choose Configure Pop-up.

    Configure pop-up option

    The Configure Pop-up pane opens. The default title shows the county name and state. You'll shorten the title by changing the state name with the state abbreviation.

  3. For Pop-up Title, delete {State_1} (the field for state name). Click the Add field name or expression button and choose {ST_ABBREV} (the field for state abbreviation).

    Pop-up Title configured

    Next, you'll change the pop-up's contents. Currently it displays a list of all fields. You'll create a custom display that includes some explanatory text and the field for low birth weight percentage.

  4. Under Pop-up Contents, for Display, choose A custom attribute display. Click Configure.

    Pop-up Contents Display parameter and Configure button

  5. In the Custom Attribute Display window, type (or copy and paste) the following text:

    The percent of babies born at low birth weight (5.5 lbs or less) between 2010 and 2016 in {NAME_1}, {ST_ABBREV}, was {Percent_Low_Birthweight} percent.

  6. Highlight {Percent_Low_Birthweight} and click the Bold button.

    Bolded Percent Low Birthweight field

  7. Click OK. In the Configure Pop-up pane, click OK.
  8. Click a county to open its pop-up.

    Configured pop-up

    The pop-ups now convey important information quickly to users. However, this pop-up is not appropriate for every county.

  9. Click a county that is colored white.

    Pop-up for a county with no data

    Counties that are white have no data for low birth weight percentage. There are a few of these counties, usually in sparsely populated areas that had too few births between 2010 and 2016 to obtain a valid rate. For these counties, the pop-up you configured doesn't make sense, because there is no percent value to display.

    You'll create an Arcade expression that causes the pop-up show a different message if the county has no data.

  10. Close the pop-up. Open the Configure Pop-ups pane for the County Health Rankings 2018 layer.
  11. Under Attribute Expressions, click Add.

    Add button

    An Arcade window opens. The default title of the expression is Custom.

  12. Next to Custom, click Edit. Type Null Values Exception and click Save.

    Next, you'll add a when expression. A when expression indicates that when a certain condition is met, a certain outcome occurs. Your expression will indicate that when there is no value for percent low birth weight, appropriate text will be displayed. In all other cases, the percent low birth weight will be displayed as usual.

  13. For Expression, type when(. Under Globals, scroll through the list of fields and click $feature["Percent_Low_Birthweight"].

    Percent Low Birthweight field

    The field is added to the expression inside the parentheses.

  14. After the field name, but inside the parentheses, press Spacebar and type == null, "not available".

    Expression showing condition and outcome

    This expression means that when the low birth weight field has a null value, the text not available will be shown. Next, you'll add the text that will be shown when the low birth weight field does have a value.

  15. Type a comma and add the low birth weight field again ($feature["Percent_Low_Birthweight"]). Press Spacebar and type + " percent" (be sure to include the space before percent).

    Expression showing what happens in all other cases

    Note:

    If you have trouble creating the expression, you can instead copy and paste the complete expression:

    when($feature["Percent_Low_Birthweight"] == null, "not available",$feature["Percent_Low_Birthweight"] + " percent")

    The last part of the expression indicates that when the percent low birth weight field isn't null, the field's value plus the word percent will be shown. Next, you'll reconfigure your pop-up to use the expression.

  16. Click OK. In the Configure Pop-up pane, for Pop-up Contents, click Configure.
  17. From the end of the existing text, delete {Percent_Low_Birthweight} percent and type {expression/expr0}. If necessary, apply bold formatting to the expression.

    Pop-up custom attribute display text

  18. Click OK. In the Configure Pop-up pane, click OK.
  19. Open the pop-up for a county with data to confirm that it still displays as expected. Then, open the pop-up for a county with no data.

    Final pop-up

    The pop-ups are appropriate for both types of counties.

  20. Close the pop-up and save the map.

Identify counties for intervention

The national map you created is great context for the issue of low birth weights across the United States. However, decision makers in Colorado only have authority over the counties in that state. Furthermore, you want to explicitly highlight Colorado counties in which intervention programs should be implemented to help the state reach the CDC's goal by 2020.

To do so, you'll save a copy of your map and filter your data to only show counties in Colorado. Then, you'll change the way the counties are styled to highlight two types of counties in need of intervention: those close to the CDC goal (for which only a little intervention is needed) and those far from it (for which intervention can help a lot).

Note:

If you want, you can also perform this workflow on any state in the United States.

  1. On the ribbon, click Save and choose Save As. Enter the following parameters:
    • For Title, type Priority Counties for Prenatal Programs.
    • For Tags, replace the United States tag with Colorado.
    • For Summary, type This map shows priority counties within Colorado for increased prenatal programs.
    • Leave the other parameters unchanged.
  2. Click Save Map.

    You are now working with a copy of your original map. Next, you'll filter the data.

  3. In the Contents pane, point to County Health Rankings 2018 and click Filter.

    Filter button

    A filter uses an expression to determine which features from a layer to show. You can create expressions using specific fields and attributes. Your data has a field for states, so you'll use that.

  4. In the Filter window, for the first drop-down menu, choose State Name.

    The second drop-down menu is automatically set to is.

  5. Under the third box, click Unique. Choose Colorado.

    Expression for filter

    The full filter expression reads State Name is Colorado. The filter will only display features that fit this criterion.

  6. Click Apply Filter and Zoom To.

    Default map of Colorado

    Some Colorado counties are dark gray, some are dark orange, and some have no data. The state has a diverse range of low birth weight percentage by county. Next, you'll investigate the histogram of values for the state.

  7. Open the Change Style pane for the County Health Rankings 2018 layer. Under Counts and Amounts (Color), click Options.

    Histogram for Colorado counties

    The breakpoints you set at the national level remain, but the average has changed to 9 percent. While this average doesn't account for population differences between counties, the CDC also records in Table I-21 (you may need to scroll to find the table) that 9 percent of babies in Colorado are born at low birth weight.

    Decreasing low birth weight in counties already below the state average will likely require a lot of investment for not much payoff. It'll be more impactful for the state to focus on the counties with the highest percentages. Although these counties will require a lot of investment, the investment will make a bigger difference in the statewide average.

    It'll require fewer resources to help counties that are only slightly above the CDC goal of 7.8, so these counties should be highlighted for intervention too. Thus, your policy map will emphasize two groups of counties: Those high above the CDC goal, and those just barely above it.

    To display these counties on your map in a way that makes clear which ones should be targeted for intervention, you'll use classified colors for counties with specific ranges of values instead of a continuous color ramp that gradually changes color as values increase or decrease. This style will make the map more readable at a glance and tell a simple, clear story to decision makers.

    Because the classifications you use will automatically focus only on counties above the CDC goal, you'll use a high to low theme.

  8. For Theme, choose High to Low.
  9. Under the histogram, check Classify Data.

    Classify Data check box

    By default, the Natural Breaks statistical method automatically classifies the data into four categories based on the data's distribution. You'll instead create manual breaks in the data: one for the CDC average (7.8), one for the statewide average (9.0), and one for extreme values (14.0).

  10. For Using, choose Manual Breaks. Adjust the handles of the histogram to 7.8, 9.0, and 14.0.

    Manually configured histogram

    Next, you'll adjust the legend to make clear what each classification means.

  11. Click Legend. Click the > 14 - 17 label, type Over 14.0 percent - In most need of programs, and press Enter.
  12. Change the remaining labels in the following way:
    • Change > 9 - 14 to 9.01 percent to 14.0 percent - Room for improvement, invest here if resources allow.
    • Change > 7.8 - 9 to 7.81 percent to 9.0 percent - Could benefit from small interventions.
    • Change 5.4 - 7.8 to 7.8 percent and under - Meets the 2020 CDC goal.

    Labels for the policy map

    You'll also change the colors of the counties to yellow to red to indicate magnitude of importance.

  13. Click Symbols. Scroll through the color ramps and choose the yellow to red color ramp.

    Yellow to red color ramp

  14. Click OK. In the Change Style pane, click OK and click Done.

    Final map of Colorado

    This map indicates three counties in the greatest need of intervention, as well as clear categories of other counties where intervention may be advised.

    The Centers for Disease Control and Prevention defines low-weight births as babies born 5.5 pounds or less. One of the three Colorado counties in the greatest need of intervention, Jackson County, located along the Wyoming state line, led the state in low birthweight infants. According to County Health Rankings, 27 percent of Jackson County's births from 2010 to 2016 were categorized as low birthweight.

    It's unclear why this sparsely populated county leads Colorado in that category. However, amplifying the low-birthweight issue is recreational marijuana, which Colorado legalized in 2014. Studies have shown that, like tobacco, smoking marijuana during pregnancy significantly increases the likelihood of low birthweight infants. Also according to the CDC, between 15 percent and 18 percent of Colorado's adults smoke tobacco, a figure which puts that state within the national average.

  15. Save the map.

In this lesson, you created a policy map indicating to decision makers the counties in Colorado most in need of intervention to help the state reach the CDC's goal for low birth weights by 2020. In the next lesson, you'll share your results as a user-friendly web app that provides context and functionality.