# Effective ways to communicate margins of error through pop-up windows

In part 1 from this blog series, we learned the importanthis to use margins of error (MOE) in our mapping projects. Since survey data such as the American Community Survey (ACS) contain some kind of sampling and no sampling mistakes, especially in small geographic areas like groups of blocks and leaflets, it’s our job as card makers to communicate margins of error when necessary.

the we Census office has done us a huge favor by offering margins of error (MOE) for their The data estimates, and othe only method of presenting MOEs is to use the pop-up window of the map. Pop-ups allow the card hearing at dive deeper into a the subject of the map to see a detailed breakdown of the data.

This blog to brings various techniques and suggestions for how to create effective pop-ups on ACS estimates and MOE. You can use these techniques together or on their own. There is no one right way to communicate MOEIt is therefore up to you, as a cartographer, to assess the important information to be transmitted.

Before even adding an MOE to our pop-up, let’s talk about effective pop-ups in general. We often have see pop-ups with a list of attributes, letting the card reader decipher what is important. The easiest way to create an effective pop-up window is to a table of data into something digestible using easy-to-read instructions with a personalized pop-up. If you are not familiar with the Web map pop-ups, this blog talks about the basics, and this blog introduces custom pop-up views (which are used everywhere this blog).

For example, the pop-up at the top requires our card reader to scroll through a long list of attributes, while the one at the bottom includes the information. important for the map. This example even matches the colors of the map at the attributes in the arise.

Some tips when creating pop-ups on ACS and MOE:

- Think about which attributes you need to help your card tell an informative and concise story. Don’t let your card reader do all the work by including each attribute of a table.
- Include the “base” (AKA the denominator) somewhere in your pop-up. If you are mapping the percentage of the population aged 25 and over without insurance, include the total number of the population aged 25 and over somewhere in your pop-up.

- Not all attributes will necessarily need the included MOE. Think about how your audience plans to use the map and if the MOE is important to them.

Once you know the key information you want communicate to your audience on the map, land’s see at some simple ways to create impactful ACS pop-ups with our estimates and MOEs.

One way to communicate this ACS The data contains sampling Errors without causing alarm is at wee li wordske “approximately” or “estimated” when talking about the estimate himself. Notice how the pop-up below uses these words to help communicate that the number is not correct, without calling attention to the error itself.

If you want to communicate the estimate with the MOE, consider using +/- or ±. This helps to show that there is more than just an estimate to consider when using the data.

Note how this pop-up also uses the language suggested in the previous section. This is an example where more than one technique was used.

Some readers may want to know more about MOEs and what they are. One way to help them learn more is to include information about EOMs. directly in the pop-up itself. The pop-up below Links to this document on understanding the error of the ACS Manual. This technical allows the reader to explore MOEs further if they wish.

NOText, let’s go explore some of the more advanced ways to include EOMs. These will require additional calculations, but luckily Arcade allows us to do this on the fly..

As explained in this blog, the MOE tells us that there is a range of values in which the estimate could fall realistically. This range is known as the confidence interval.

We can communicate this range of values in our pop-up in many ways without using the words “confidence interval. “ The example below shows how the range can be used in a sentence related to the subject of the map.

To create this range, we add and to subtract the MOE to estimate. In the pop-up window, create a new expression for the upper and lower limits of the range. For the upper limit, just add the MOE to the estimate.

*Note: the example below uses variables and a return statement, but you can also just put $**functionality[**ESTIMATE_FIELDNAME] + $**functionality[**MOE_FIELDNAME] and it would also return the upper limit.*

Then we can create another expression for represent the lower bound, which is the estimate minus the MOE. When creating the lower limit, it is possible that the MOE is greater than the estimate, which could make the lower bound a negative number. To make sure that we do not try to pretend that there are –31 mobile homes in an area, we can use the When() function to manage cases less than 0.

The second line of this expression says “When the lower limit is greater than 0, returns the calculated lower limit.” Else (if the value is negative), returns 0. “

You will now have two expressions that you can use like any other attribute in your data.

Another way to understand the reliability of an estimate and its MOE is to use coefficient of variation. This essentially tells us the error as a percentage of the estimate. Here is an example that transforms this reliability number in three levels reliability: high reliability, medium reliability and low reliability.

In this example, we are using three categories, which are the the same reliability thresholds used in Esri Business Analyst products. If the ME is less than 12% of the estimate, it is considered **high reliability**. If it is between 12 and 40%, it is considered **medium reliability**, and anything above 40% is considered **low reliability**.

TThe equation below can be used to calculate ee coefficient of variation at 90% confidence level:

*Note: to get a 95% **confidence level, trade 1.645 for 1.96. *

As we showed in the previous example, we can use Arcade calculate new values on the fly for our pop-ups. First, you will calculate the coefficient of variation using the equation above. Then you can use a When () statement like we did in the previous example to turn the numbers into readable text.

We can then use this expression like any other attribute in our pop-up to communicate the reliability of the data:

Using the same equation as the previous example, we can set the color to appear in our pop-ups to associate with reliability. In the Example below, MOE is highlighted in red, green or yellow to indicate high / medium / low reliability. NOTotice some other technicals we haveen throughout this blog.

In this Arcade statement, instead of words, the statement produces hexadecimal color values associated with each reliability category.

In Map Viewer Classic, we can use this expression in the contextual HTML code to color the MOE. This blog explains how to insert the expression into your HTML.

This blog contains different methods for presenting EOMs in your card pop-ups. These methods can be used together, on their own, or in any combination that suits your card’s needs. Not all cards will need MOEs explicitly stated, while others will benefit greatly from the methods presented in this blog. Consider your card audience, their understanding of ACS data, and how the card will be used by the end reader. These factors will help you decide which method (or methods) is the most appropriate for your card.

Blog – The Census Bureau gives you margins of error, we help you map them

Blog – The importance of margins of error and mapping

2020 ACS Manual – Chapter 7. UNDERSTANDING THE ERROR AND DETERMINING THE STATISTICAL IMPORTANCE

Blog – Answers to your Arcade questions

the Web Examples of cards: