Group 6: Friendship Works
Katelyn Donn, Narayan Sharma, Ronit Thummaluru
Friendship Works logo (via FW website)
Friendship Works logo (via FW website)
The motivation behind our project is to demonstrate different visualizations that can help represent the tremendous impact that Friendship Works is having on our community. From helping older adults create and maintain a friendship, to inspiring volunteers to do good for our community, FriendshipWorks has cultivated a meaningful, genuine experience for everyone that partakes in it. Organizations such as FriendshipWorks rely heavily on grant funding, which means that every cycle, they need to apply for grants in order to keep their organization running. Their proposal includes many different ideas, but they mostly seek data visualizations to help show the tangible impact that they have on the community. Sometimes, it may be hard to show the quantifiable parts of a given friendship, but we hope to include different visualizations that will allow Friendship Works to raise more money so they can continue to make such a lasting impact on our community.
Our data primarily consists of information regarding older adults and their demographics as well as information about the volunteers signed up. Our data is broken into a multitude of datasets, each containing very specific data relating to our visualizations. One of the datasets we plan on utilizing tells us about the older adult and volunteer locations in Massachusetts. In addition to that, we were also planning on using more selective data which relates to the specific program each older adult is signed up for. This data helps to support our goal of identifying the demographics in each area in order to determine resource allocation. Closing the gap between the difference in volunteers and older adults in each district will help FriendshipWorks become much more efficient in the long run.
In terms of location, we are covering all of Greater Boston because the goal of our data is to highlight deficiencies in the different districts in the area. Using the two different data sets based on location, we will have a toggle that allows the user to see a visualization of the specific dataset they are interested in. If no district is selected by the user, the default visualization will display the demographic data of all of Boston in one chart.
The data was given to us in a very clean fashion with only minimal imputation required, but there were a lot of rows with missing data. All of the data we received from FriendshipWorks was categorical, so we could very clearly see if there was data cleaning required on a dataset. We performed imputation on the "state" column in order to standardize the data. For example, we had instances where one row would have Massachusetts as the state while another would have MA. Standardizing our data made it much easier to plot out the data in a cleaner fashion as it was all grouped correctly.
The interview was a really eye-opening experience as we gained more insight into not only what data visualizations we needed to make for the organization, but also the work and genuine passion the organization has for the community it serves. As their work mainly surrounds the care of older adults and the feelings of contentment, gratefulness, kindness, and gratitude the adults feel after each visit, it’s hard to tangibly see the results that their work has. However, since their organization is also a non-profit, they receive most of their money from grants and/or outside funding, which often requires real and substantial data in their funding process.
One thing that I thought was really interesting was that the organization has an extensive vetting process for both the volunteers and the older adults before accepting them into the program. On the older adults side, they vet them to make sure they don’t have any real medical conditions like dementia (in which the organization directs them instead to a medical facility so they can receive hands-on care) and that their home is safe for their volunteer partner to enter and visit. On the volunteer side, they vet them to make sure that the volunteers have a genuine want to serve the community and conduct background checks on their criminal history and whatnot before accepting them.
After creating multiple graphs and charts analyzing the data for our data report we were able to find multiple interesting results based on what we observed from the observations. We found that per each mailing city there is a distinct count of each age which shows a close proportion between males and females. It can be inferred that there is no specific gender demographic that is preferred over another which conveys the idea that FriendshipWorks is truly for everyone of any gender and that there’s no specific preference towards one.
We also were able to find the correlation between the number and length of matches and overall, the number of matches has a negative correlation with the length of matches, which means that a lot of matches last for a shorter amount of time. In order to find any discrepancies among the older adult to volunteer ratio we plotted the number of volunteers per zip code and found that zip code 02134 clearly has the highest number of older adults residing within that area. The hope is that the amount of volunteers also displays similar numbers within each respective district. If there is a large disparity then we can quickly identify it. These were some key insights we were able to make based on observations made at a high level.
Our design process was relatively straightforward when it came to deciding which designs to move forward with. With the motivations that were given to us, we wanted to create a quantitative visualization that depicted some sort of ratio between the recipients and volunteers of the friendships. We thought that portraying this ratio would clearly show how unique each friendship is. Although a straight ratio between these two demographics would be useful, it would not be a thorough enough demonstration of how truly unique each friendship is.
So to dive deeper into this quantitative state, we chose to divide this ratio even further into the type of program participation. We believed that this would portray the exact demographic of the most common types of friendships throughout a given dataset. The next question we had to answer is how exactly we would seperate each demographic into smaller subsections. To do so, we thought to combine our different demographics into an embedded image of map to help visualize exactly where these special friendships were happening and how often they were happening relative to the other types of friendships in a given area. The idea behind this was to help visually show any specific correlations between types of visits throughout different neighbors in Boston.
Combined visualizations embedded of Types of Visits Per Neighborhood
We realized that this was not the most optimal way to show how each neighborhood compared to another as each pie chart pertained to the specific neighborhood that was selected. To overcome this obstacle, we created a different visualization that shows the relative size of each neighborhood that also showed the program distrubtion rate embedded within another visualization.
First Visualization(s)
This visualization shows the difference between each gender population and how prevelant each one is depending on the age. The purpose of this visualization is to see if there are any common patterns between each demographic or if there was any outlying data for a specific demographic. Overall, the most predominant area is an older aged female adult but as the map is showing, there is a strong correlation between age and prevelance of the number of recipients. Bascially, as the age increases there are more adults within that given age range. We are showing this through two type of visualizations in order to highlight this pattern.
Second Visualization
The top graph shows the number of recipients in each neighborhood and it gives us a clear image that brookline has the most recipients meaning that they should be receiving more funding than compared to neighborhoods with a small population of older adults. In addition to that, we have a pie chart that is linked to our first visualization and it gives us a more in-depth look at the programs that the older adults take part in in each neighborhood. This gives us even more specific data as to which programs seem to be the most popular in addition to which regions see the most activity.
Third Tableau Visualization of Medical Providers
This visualization depcited through Tableau is a size chart to show the relative counts for each Medical facility. Being that the greater Boston area has many different medical facilities, we thought that the best way to show where the most traffic is occuring is through a relative size chart shown through this bubble chart. Idealy, this chart should help show users who may be potentially interested in participating with FriendshipWorks where the most predominant locations seperated through zip code.
Through the thorough extent of this project, our group has been able to provide visualizations that visually articulate the tremendous impact that Friendship Works is making on their wonderful community. The original problem that we were tasked to solve was how we could quantify the numerous friendships that were being created as it is quite difficult to be able to present the happiness felt by an older adult.
To overcome this barrier we were able to design and implement visualizations that show exactly how prominent these friendships are and how important they are to these older adults. Ranging from Dorchester to the Roxbury area, our visualizations show exactly how much older adults are benefitting from these amazing program. Through the future, we hope that the welcoming leaders of Friendship Works receive more grant funding to continue this successful, heartfelt organization.