Duplicated and Unduplicated Counts
You will often see the terms duplicated and unduplicated counts when discussing nonprofit data. They are two important ways of looking at the same data. We'll start with a very generic definition.
Duplicated Count - the total number of units your nonprofit touched over a specified time period.
Unduplicated Count - the total number of unique units your nonprofit touched over a specified time period.
For human service organizations, the “units” are individual people or families. For an animal rescue, the “units” might be dogs or cats or leatherback sea turtles. Those units could also be outcomes that you achieved (e.g. family acquired stable housing or child reads at grade-level).
The distinction between duplicated and unduplicated is that the unduplicated units are “unique”. In other words, each unit is only counted once, even if it shows up multiple times in your data.
With duplicated counts, you might count the same person, family, or cat more than once, but with unduplicated counts, you count them only once, even if you serve them numerous times.
Why Duplicated and Unduplicated Counts Matter
As you’ll soon see, duplicated and unduplicated counts can be (vastly) different, especially when you have multiple programs or provide services repeatedly to the same people, animals, or whatever else your "units" are.
We need to know how duplicated and unduplicated counts differ so that we can clearly and truthfully describe our programs to our supporters. Foundations and governments are particularly interested in unduplicated counts of participants and outcomes. Even foundations that provide general operating support will often want unduplicated data in year-end grant reports.
Duplicated and Unduplicated Counts - Shelter Nights
A simple example from a shelter might help us. Table 1 shows 5 nights in a small overnight shelter. The shelter can host 5 people per night. There are a maximum of 25 bed-nights available in the 5 nights of data.
We have 3 outputs to consider:
Total Bed Nights: the total number of beds that are filled during the time period, regardless of who is in the bed.
Duplicated Participants Served: the total number of participants who stayed in the shelter during the period. People who stay multiple nights are counted multiple times.
Unduplicated Participants Served: the total number of unique participants who stayed in the shelter during the period. People who stay multiple nights are counted once.
Duplicated Count - Shelter Participants
Table 1: Shelter Stays By Day (Monday - Friday)
Monday | Tuesday | Wednesday | Thursday | Friday |
Vanna W. | Martin Q. | Jose R. | Omar A. | Omar A. |
Pedro P. | Kevin F. | Lee Z. | Maria S. | Tiffany L. |
Brian Y. | Chris B. | Martin Q. | Tiffany L. | Jack B. |
Martha R. | Martha R. | Tonya H. | Martin Q. | |
Chris B. | Carlos S. | | Lewis H. | |
In Table 1, we can see that Total Bed Nights = 22. On Wednesday and Friday the shelter has open beds.
We can also see that the following people stayed multiple nights:
Martha R. - 2 nights
Chris B. - 2 nights
Martin Q. - 3 nights
Tiffany L. - 2 nights
Omar A. - 2 nights
If we want the duplicated count of the participants who stayed in our shelter, we simply count the total number of people who used the shelter, regardless of whether any individuals stayed multiple nights.
Monday = 5
Tuesday = 5
Wednesday = 4
Thursday = 5
Friday = 3
Duplicated Count of Participants at the Shelter Monday - Friday = 22
Unduplicated Count - Shelter Participants
Some of our participants stayed multiple nights. The duplicated count of participants adds each participant to the total, even when the participant has already been counted on a prior night. This value (22 duplicated participants) is important, but it also does not tell the whole story. If we want to know how many unique participants stayed in our shelter, then we need to determine the unduplicated count.
Table 2 below visualizes how we could get the unduplicated count of participants staying in the shelter. As you can see, we excluded the names of participants after their first night. That allows us to count each participant one time. For example, Martin Q. stayed 3 nights, but we only count him as a unique participant on Tuesday, but Martin Q. is excluded on Wednesday & Thursday.
Table 2: Shelter Stays by Night - Unduplicated
Monday | Tuesday | Wednesday | Thursday | Friday |
Vanna W. | Martin Q. | Jose R. | Omar A. | excluded |
Pedro P. | Kevin F. | Lee Z. | Maria S. | excluded |
Brian Y. | excluded | excluded | Tiffany L. | Jack M. |
Martha R. | excluded | Tonya H. | excluded | |
Chris B. | Carlos S. | | Lewis H. | |
With those adjustments, we can determine the number of unique participants each night.
Monday = 5
Tuesday = 3
Wednesday = 3
Thursday = 4
Friday = 1
Unduplicated Count of Participants at the Shelter Monday - Friday = 16
Reporting Duplicated and Unduplicated Counts - Shelter Participants
You might look at this and think “the shelter isn’t getting credit for those bed-nights if you remove people from the unduplicated count.”
If you only report the unduplicated count of people staying at the shelter, then you won’t have a complete picture of the work the shelter does. To give a complete picture of the work the shelter is doing, you might report the following:
16 unduplicated participants stayed at the shelter during the Monday - Friday time period.
Those 16 participants stayed for a total of 22 bed-nights during the Monday - Friday time period.
5 adults stayed more than one bed-night during the Monday - Friday time period.
In this small shelter, with only one week of data, the difference between the duplicated and unduplicated counts seems modest. However, imagine a shelter with fifty (50) beds and is open 365 nights per year. Suddenly, we have the possibility of more than 18,000 bed-nights. The difference between duplicated and unduplicated counts of participants in the shelter will probably get very large.
Organization-Wide Duplicated and Unduplicated Counts in a Multi-Program Organization
By now, you probably have a good sense of the difference between duplicated and unduplicated counts. We’ll use one more example to help you think about how they can impact your data collection and reporting.
Imagine your organization has the following 3 program: Basic Needs, Housing Stability, and Financial Literacy. You can see program enrollment for the last calendar year in Table 3 below.
Table 3: Participant Enrollment in the Basic Needs, Housing Stability, and Financial Literacy Programs Last Calendar Year.
Basic Needs | Housing Stability | Financial Literacy |
Martin Q. | Jose R. | Omar A. |
Kevin F. | Lee Z. | Maria K. |
Tonya H. | Martin Q. | Tiffany L. |
Martha R. | Tonya H. | Tonya H. |
Carlos S. | Don D. | Martha R. |
In this example, we can look at organization-wide duplicated and unduplicated counts of program enrollment. Since we are looking organization-wide, rather than at a single program, we will include all of our programs in our participant counts.
We have three metrics to consider
Duplicated Count of Organization-Wide Participant Enrollment - Total number of participants enrolled in any of the organization's programs. It is obtained by adding total enrollment of each program and does not take into account whether a participant is enrolled in multiple programs.
Unduplicated Count of Organization-Wide Participant Enrollment - Total number of unique participants enrolled in any of the organization's programs. Participants enrolled in multiple programs are only counted once.
Unduplicated Count of Participants Enrolled in Multiple Programs - Total number of unique participants who are enrolled in more than 1 program. This metric helps the organization understand program use patterns.
Duplicated Count of Participants Organization-Wide
To determine the duplicated count of participant enrollment organization-wide, we simply add up the total number of people enrolled in each program (see Table 3 above).
Total Participant Enrollment By Program
Basic Needs = 5
Housing Stability = 5
Financial Literacy = 5
Organization-wide duplicated count of participants enrolled in programs last year = 15
Unduplicated Count of Participants Organization-Wide
However, it’s clear that some of our participants (Martin, Tonya, and Martha) are enrolled in multiple programs. In the duplicated count, we count these participants more than once. We can only count unique participants to get the unduplicated count, so we only count them one time.
Table 4, below, helps us visualize how we could determine the organization-wide unduplicated count of participants.
Table 4: Determining the Unduplicated Count of Organization-Wide Program Enrollment
Basic Needs | Housing Stability | Financial Literacy |
Martin Q. | Jose R. | Omar A. |
Kevin F. | Lee Z. | Maria K. |
Tonya H. | excluded | Tiffany L. |
Martha R. | excluded | excluded |
Carlos S. | Don D. | excluded |
Number of Unique Participants Enrolled By Program
Basic Needs = 5
Housing Stability = 3
Financial Literacy = 3
Organization-wide unduplicated count of participants enrolled in programs last year = 11
Now, you might look at these numbers and think “wait, there are still 5 people enrolled in the Housing Stability Program and Financial Literacy Program, but it looks like you are saying there are only 3 people in that program!”
You are correct. There are five (5) people enrolled in those programs. And, if you were asked to report the number of unique (i.e. unduplicated) participants enrolled in the Housing Stability Program, then you would, in fact, report that there are five (5) unduplicated participants enrolled.
However, we are looking at the unduplicated count of program enrollment organization-wide. The counting approach shown for Table 4 is simply a way to demonstrate how we would determine the unduplicated count of participants organization-wide. It helps us see that we count unique individual participants one time when when they are enrolled in multiple programs.
Unduplicated Count of Participants Enrolled in Multiple Programs
Looking back at table 3, we can see that Martin Q. and Martha R. are enrolled in 2 programs and Tonya H. is enrolled in 3 programs.
Three (3) out of 11 unduplicated participants are enrolled in multiple programs, or about 27 percent of the organization's participants.
Eight (8) out of 11 unduplicated participants are enrolled in a single program, or about 73 percent of the organization's participants.
So we can see that the vast majority of our unduplicated participants are enrolled in one program, but a sizable minority are enrolled in multiple programs.
Are duplicated counts ever the right way to present data?
Yes, especially in a low data collection context. Consider a soup kitchen that provides breakfast, lunch, and dinner. Some people who eat at that soup kitchen might eat multiple meals per day. Soup kitchens are often “no questions asked” programs that don’t get detailed demographics or IDs from the people they serve.
In this case, they can count the number of meals that they serve, and maybe even count the number of adults, children, and seniors, but they aren’t carefully tracking if Person A had breakfast and dinner or Person B only had lunch. That soup kitchen can say "we provided 500 meals yesterday" but can't say "we served 500 meals to 300 unduplicated participants".
The important things to keep in mind are:
What data does your audience need in order to understand the work you are doing, the benefits you are providing, the impact you are making?
What do you need to do your work effectively and efficiently?
If you and your audience don't need unduplicated data, then it probably doesn't make sense to collect it and present it that way.
Tracking and Reporting Unduplicated Counts Effectively
Your nonprofit needs to take into account the requirements of accurately tracking and reporting unduplicated participants, services, and outcomes, including:
You need to identify and count each participant uniquely.
You need to connect each output and outcome to an unique, identifiable participant.
You need to track when you served each participant (i.e. program start date and end date).
You need to "de-duplicate" your data when creating reports.
With this type of data, you can say things like:
“In 2022 we completed 500 tutoring sessions with 75 unduplicated children, and 50 unduplicated children achieved grade-level reading by the end of the program.” Or
"In the first 3 months of 2023, we provided computer training to 100 unduplicated adults, and 33 of them obtained full time jobs."
If you are using specialized data tracking and case management software, then that software will help you address these data needs much more easily than if you are using spreadsheets. However, if your human services org uses spreadsheets to track your work, you'll have to find a way to manually address the requirements noted above in order to provide accurate, unduplicated data.
Conclusion
Duplicated and unduplicated counts are essential concepts for nonprofits to understand and use for data collection and reporting. Whereas duplicated counts allow you to include duplicates (ha!), unduplicated counts require you to count each unique unit (person, animal, etc) only once.
Using both duplicated and unduplicated counts can give you and your supporters the clearest picture of the scope of your nonprofit's impact, and give you another tool to analyze your work.
This post is part of our nonprofit data bootcamp series. Check out the complete list of nonprofit data bootcamp topics with links to other published posts.
Reporting your impact is hard when you’re juggling spreadsheets. countbubble makes it easy so you can focus on your mission.
countbubble is nonprofit data management software simplified.
Learn how countbubble helps nonprofits track and report their participants, services, and outcomes. Email us at contact@countbubble.com or sign up for email updates on blog posts, product news, or scheduling a demo.
Founder, CountBubble, LLC
Comentarios