Why Connected Matters: A Data-Driven Exploration

Bowling Green, Kentucky (June 20, 2023) - When you work for a nonprofit, you wear a lot of hats. Each project is different, and you’re challenged to create innovative solutions that best achieve the goal at hand. As a result, my job responsibilities as a Research Assistant at Connected Nation change often.

At the end of 2022, I worked primarily on data analysis for the Connected Community Engagement Program (also know as the Connected program). Launched, in 2011, this program aims to craft comprehensive action plans to bolster broadband infrastructure and digital inclusion in participating communities. These policy prescriptions come straight from the data — acquired through surveys and interviews with community members.

Given that the Connected program has existed for so long, we have plentiful data from diverse populations nationwide. I originally sought to use it to illustrate trends, but ended up with a deeper understanding of why the program exists in the first place.

Identifying broadband assets

The Connected Program offers communities an opportunity to explore their current broadband environment and improve residents’ everyday lives. We partner with local champions and other community stakeholders to implement surveys and gather data. While our residential survey generates the most responses, we also reach out to businesses, local government agencies, schools and universities, and key economic sectors like agriculture and health care.

The surveys touch on many topics, including internet adoption rates, satisfaction with internet services, digital literacy, access to internet-connected computing devices, and using online resources — both for leisure and in several employment sectors.

Alongside this data collection effort, our mapping team determines the availability of broadband infrastructure and other assets in the community. CN’s Broadband Solutions Managers then transform this information into accessible deliverables with policy recommendations that we present to stakeholders. Finally, we create an online portal for residents and policymakers to reference later.

Our granular, community-specific approach ensures that every partnership we forge gets the attention that it deserves. In the process, we collect some incredibly useful data that I originally sought to analyze and highlight some general trends. Instead, I quickly realized that I was comparing apples and oranges.

The metrics of comparison

The biggest issue in comparing Connected communities over time is that the communities surveyed vary significantly from month to month. Connected surveys typically remain open in their communities for one to four months; as a result, the communities surveyed in January 2021 might differ substantially from communities surveyed in October 2021, for example.

Changes in metrics like internet adoption over that span might indicate improvement or decline, but they could also illustrate that communities surveyed later in the year had different attributes than those surveyed earlier in the year.

This possibility became evident once I started graphing trends over time — I could not tell whether observed changes were real or a product of the data. More importantly, comparing communities based on one or two data points can minimize important differences between them that could be lost if we rely too much on general trends.

Two Connected communities, Ellis and Knox counties (both in Texas), personify these concerns well.

Comparing these two counties on common broadband metrics would illustrate that they are largely the same. With respect to access, the new FCC broadband map suggests that 87.1% of households in Knox County and 87.6% of households in Ellis County can subscribe to wired or licensed fixed wireless home internet at speeds of 25 Mbps download and 3 Mbps upload (often referred to as 25/3).

Home internet adoption rates are also remarkably similar. Based on survey data gathered through the Connected program, 94.9% of responding households in Knox County and 96.4% in Ellis County subscribed to home internet of any kind. Both rates are much higher than the average from all Connected communities (82.7%).

Based on this information alone, you might expect that these two communities generally have the same broadband environment, and therefore solutions and recommendations for improvement should be similar.

However, relying on these numbers alone fails to recognize the nuanced differences between these communities.

Understanding unique challenges

Having internet access does not necessarily say anything about residents’ ability to use it. To that point, the Connected surveys for these two counties contain several indicators that measure digital literacy – asking about people’s familiarity with various digital hardware, software applications, communication tools, and other online activities.

These questions ask respondents to indicate their skill level from five response categories. I assigned these categories values between 0 and 3, where 0 means “I need to learn,” 1 indicates “I know a little,” 2 indicates “I’m comfortable with it,” and 3 indicates “I could teach this.” “Not interested” responses were changed to missing values because they do not gauge an individual’s skill level directly.

I used this data to construct additive indices of digital literacy skills across the four broad categories, with higher scores indicating stronger digital skills in the aggregate. The results paint a wildly different picture than looking at the access and adoption data alone.

With regard to hardware digital literacy, average scores for each Connected community range from 16.2 to 20.7. Out of all the communities that we surveyed between 2020 and 2022, Ellis County had the highest score (20.7). Meanwhile, Knox County had one of the lowest scores (17.3).

The gap between scores illustrates that Ellis County residents are more technologically savvy on average than Knox County residents. This trend holds up when looking at other digital literacy indices. For software, average scores for each Connected community range from 6.9 to 11.7. While Knox County had one of the lowest scores (9), Ellis County had one of the highest (11.6).

For digital communications, the average literacy scores ranged from 8.1 to 13.1 — again, Knox County had one of the lower scores (10.8) while Ellis County had the highest (13.1). These two communities looked quite similar from the onset but digging deeper, it’s clear that they have different needs and might want to consider allocating broadband funds toward different purposes.

Developing custom solutions

What does this mean for the future of the Connected program? By and large, this data exercise illustrates just how valuable the program is.

In partnership with active local champions, we have the ability to gather fine-grained information about the broadband context in these communities to inform decision-making and policy. Much of this data (like the digital literacy scores) isn’t readily available elsewhere. Furthermore, this exercise highlights that each community is unique and should be treated that way.

While broadband action plans at the state level are instrumental in closing the Digital Divide, they cannot fill every pothole on the road to digital equity. Every community has its own needs, its own priorities, and deserves its own solutions. Working with the research team, I’m proud to contribute to that effort.

About the Author: David Nunnally is the Connected Nation Research Assistant. David is responsible for using qualitative and quantitative techniques to interpret survey data, in addition to collecting data from secondary sources to help support those findings. David works with internal and external stakeholders to help develop research and provide critical information in support of the Connected Nation mission.