The previous post on this digital inclusion series looked at multiple broadband speed thresholds in six states of the upper Midwest across four types of neighborhoods (completely urban, mostly urban, mostly rural, and completely rural). The primary focus was to gain a deeper understanding of broadband infrastructure access in the region.
Results indicated that the advertised 25 Megabit per second download and 3 Mbps upload footprint, or 25/3 for short and current Federal Communications Commission (FCC) broadband definition, improved significantly in the region from 19.8 million housing units in 2014 to 21.4 million in 2017. Unfortunately, there were significant 25/3 access gaps that persist between urban and rural areas.
Focusing solely on broadband access, however, overlooks a larger and more complex context under which digital inclusion unfolds. Digital inclusion refers to the meaningful use of broadband applications that improve quality of life. For this reason, it is important to keep in mind the type of devices and internet subscriptions that can augment or undermine broadband applications and their impact on a community’s or individual’s quality of life. By focusing on these other key elements, we are able to develop a more comprehensive picture of digital inclusion.
For this analysis, we use a new metric called digital distress. Digital distress is defined as census tracts that have a higher percent of homes not subscribing to the internet or subscribing only through a cellular data plan as well as a higher percent of homes with no computing devices or relying only on mobile devices, no laptops or desktops.
Homes that rely solely on cellular data or that do not subscribe to the internet are unable to benefit fully from digital applications, either as a result of limited data plans or because internet access can only be secured in places outside the home (such as a public library, community center, or fast-food establishments). On the other hand, homes relying on mobile devices only or not owning any computing devices can find it hard to leverage digital applications due to smaller screens or no screens at all – factors that place these homes in digital distress. The data we employ to measure digital distress metric are based on the 2013-2017 American Community Survey. Figure 1 below lists the variables utilized to identify areas in digital distress.
Figure 1. Digital Distress Variables
Given that cellular/internet and mobile/no computing devices are not mutually exclusive, simply adding them to obtain a final percentage is not possible. For this reason, variables 1 & 2 that are mutually exclusive (see figure 1) were compiled into one percentage (subscriptions) while variables 3 & 4 that are also mutually exclusive were compiled into another percentage (devices). Z-scores were then calculated for both subscriptions and devices, added together, and normalized to a score from 0 to 100, where a higher number denotes a higher digital distress. Tracts were considered digitally distressed if their score was greater than 50.
Figure 2 shows the share of overall tracts analyzed (13,105) by type as well as the share of those in distress. While 62.5 percent of tracts were completely urban, we found that 85.1 percent of those in digital distress were completely urban. In other words, data suggests that digital distress—as measured here—is mainly an urban issue. Regarding completely rural areas, they accounted for 13.1 percent of tracts analyzed, yet their digital distress share was 9.1 percent.
Figure 3 maps the tracts in digital distress across the six-state region. Notice all six states had tracts in digital distress, and this was especially evident in southern Illinois, northwestern Minnesota, and southeastern Ohio, to name a few. The list below shows the percent of tracts in digital distress by state. Ohio had almost one-third of all digital distress tracts in the six-state region:
|State||Digital Distress Tracts|
Figure 3. Digitally Distressed Census Tracts in the upper Midwest
Overall, about 8.1 percent of the region’s population (or 4.3 million people) lived in digitally distressed areas as of 2017. Figure 4 examines each of the digital distress indicators for the region and for areas in digital distress. Take note of how digitally distressed areas had a higher share of homes relying only on cellular data or mobile devices, not subscribing to the internet, or not owning any computing device.
Now that we know that digitally distressed areas exist in the six states and that most are concentrated in urban areas, we turn our attention to the socioeconomic characteristics of these distressed places.
AGE: No major differences are evident between the region in general and areas defined as digitally distress when it comes to age groups, as shown in Figure 5. The share of the population under 18 years old, though, is higher in digitally distressed areas (26.6 percent) than in the upper Midwest (22.9 percent) overall.
RACE/ETHNICITY: It is clear, from Figure 6, that the share of minorities is larger in digitally distressed areas compared to the region overall. Almost three-quarters of residents in the region analyzed were white non-Hispanic (74.9 percent) compared to 38 percent in digitally distressed areas.
EMPLOYMENT: The employment status of the working age population (ages 16 to 64) in the region versus digitally distressed areas is shown in Figure 7, with clear differences emerging. The unemployment rate is almost double in digitally distressed areas (9.9 versus 4.9 percent) and only 54 percent of those 16 to 64 years old are employed in digitally distressed areas, compared to almost 71 percent in the region. Lastly, the share of working age individuals not in the labor force is more than one-in-three (36.1 percent) in digitally distressed areas compared to less than one-quarter (24.1 percent) in the region.
COMMUTING PATTERN: Regarding employment and commuting (not shown), less than one-fifth of workers ages 16 and over worked outside their county of residence in digitally distressed areas versus 27 percent in the region. Although no major differences were evident between the region and its digitally distressed areas when it came to employment by type of industry (not shown), a slightly higher share worked in the arts, entertainment, and recreation industry in digitally distressed areas (12 percent) compared to the region (9 percent). Finally and as expected, a higher share worked from home (non-agriculture) in the region (4 percent) compared to digitally distressed areas (3 percent).
EDUCATION: An interesting pattern emerges when we turn our attention to educational attainment. As shown in Figure 8, just under 30 percent of residents ages 25 and over in the region had a bachelor’s degree or more. In digitally distressed areas, the figure was significantly lower at 10.5 percent. Likewise, the share of adults with less than high school was more than double in digitally distressed areas (23.5 percent) compared to the region in general (10.1 percent).
MEDIAN HOUSEHOLD INCOME: Similar to the pattern found by education, stark differences were seen with regard to median household income, weighted by population (Figure 9). The median household income in digitally distressed areas was half the income of the region in general ($30,137 versus $61,447).
To conclude, the nature of digital inclusion in the region is seen in a new light when viewed from a digital distress lens. The previous broadband-related blog post in our series found that broadband access has improved in the region over 2014 and 2017. Nevertheless, the urban-rural gap persists.
When explored through the digital distress lens, however, a different picture emerges. First, digital distress is mostly an urban rather than a rural phenomenon. This is an important finding given that lack of broadband access is often labeled as a rural issue; yet, digital distress is clearly an urban issue. In other words, digital inclusion is a more robust, complicated issue when looked at it holistically (broadband access plus digital distress).
Digitally distressed areas in the upper Midwest have a higher share of children and minorities, more likely to work in their county of residence (limiting their ability to potentially commute to better paying jobs), be employed in lower-paying jobs in arts, entertainment, and recreation, be less educated, have lower incomes, and a smaller share of working age residents in the workforce.
In this increasingly digital landscape, digitally distressed areas are at risk of falling further behind. As such, it is important to understand and address the factors that contribute to digital distress in various areas of the Upper Midwest region.
Some possible strategies include: (1) targeting distressed areas with programs and technical assistance activities that work to improve device and internet access/ownership; (2) enhancing the digital skills and literacy of residents of these areas; (3) and collaborating with local government, businesses, educational, and nonprofit institutions to build and implement a sound plan that systematically addresses the socioeconomic factors that limit the capacity of digitally distressed places to realize their full economic potential.
UPCOMING: The next and final post in this digital inclusion series will analyze an important related measure: the Digital Divide Index and discuss the community, economic, and workforce development implications for the region.