Social Capital and Social Distancing – Part 2

In Part 1, we introduced the concept of social capital, which refers to connections between people that generate trust, norms of reciprocity, and participation in civic life.  In general, people that join groups and interact regularly in social networks are cooperative, obliging, and inclined to do things for others.  

We suggested this community-mindedness would be well suited for effective social distancing.  After all, a key motivation for distancing is the health of our communities including, friends, neighbors, and vulnerable others.   

The descriptive data presented did show that Kansas counties high in social capital received substantially better social distancing grades than low social capital counties.

Estimated effects

We now scrutinize the relationship with a statistical model that isolates the effects of voter turnout from other variables expected to influence distancing – such as county population, income, percent minority, percent health insurance, and confirmed number of Covid-19 cases. 

The Table below lists only statistically significant variables.  Grades are derived from model forecasts that return the estimated impact of each predictor – holding other variables at typical levels.  

For example, the model predicts a C+ for the least populated county and a D for the most populated. In sum, smaller population counties post higher grades than larger ones.    

Similarly, the poorest county realized a better grade C+ that the wealthiest C-.     

Why would affluent counties grade lower than poorer ones?   After all, many well paid professions can work remotely and the wealthy can “afford” an extended lockdown.  Without more detail, and individual-level data, we can only speculate.   

Model estimates of key predictors of social distancing grades

Estimates derived from OLS multivariate model

In addition, counties with sizeable minority populations performed better than counties with less minorities – moving from a grade of C to a B -. 

While the national statistics, and figures in certain states including Kansas, show minorities account for a disproportionate share of Covid-19 infections and deaths, this does not imply a failure to distance.  Rather, in Kansas counties where minority populations are significant, social distancing grades are better than where minorities are absent.      

Finally, our measure of social capital – voter turnout – produced the largest impact on distancing.  The model returned a social distancing score of D + for the lowest turnout county.   And for the highest in turnout, the grade advanced a full letter and then some to a B -.


Would millions of Americans adhere to social distancing guidelines?  This question lingered as elected officials authorized lockdowns across the nation.  The Kansas data suggests observance varied, and substantially so.  Many Kansans did obey the guidelines, and the county grades show substantial reductions in mobility and human interaction.  However, many counties received average to failing grades.   

The counties that tended to score higher, and thus comply with distancing guidelines, possessed notable stocks of social capital – high voter turnout. 

Social capital nurtures civic engagement which is a means to address public concerns and protect community values.  For many, voting is an obligation and a responsibility to fellow citizens and to government.  The regard for community and a tangible cooperative spirt presents an ideal environment for successful social distancing.      

Critics however may counter that compliance turns on fear of grave illness and death – motives centered on self, not on community nor vulnerable others. 

Yet if self-interest sent people inside their homes, why did counties reporting no infections comply so successfully?  Some even achieved the highest grades.  For example, 29 Kansas counties did not report a single Covid-19 infection.  The average distancing grades among them was a B -, 6 scoring B’s and 3 at an A-.    

Fear surely increased in counties reporting substantial outbreaks and deaths.   But it did not seem to translate into an effective reduction in activity. In many counties that suffered unusually high outbreaks, compliance remained modest to poor.  In fact, the number of confirmed Covid-19 cases did not influence social distancing grades.[i]  

Finally, nearly two decades ago Putnam cautioned about the decline of social capital.  The pandemic underscores his warning and draws attention to the quality of our social fabric.  Among the variables examined, only social capital can be developed and renewed.  A healthy public might actually depend as much on a vigorous collective spirt as on medical advancement.  

[i] As expected, Covid-19 infections are strongly correlated with population (r = 0.71).  If we remove population from the model, Covid-19 infections reach statistical significance showing a negative association with social distancing.  In other words, counties that report low infection rates receive better grades than counties with more infections.  However, population must be considered in the model.  In addition, Covid-19 as a proportion of population does not affect social distancing. 

2 thoughts on “Social Capital and Social Distancing – Part 2

  1. If I understand your concept of social capital correctly, it’s essentially someone’s ability to act with empathy. My question is why your model doesn’t consider actions that empathize with those people who are out of a job and have no savings and money coming in? They NEED to work. How do we consider policies that demonstrate empathy for those small businesses that operate on a shoestring? I heard a rumor that some models expect up to 40% of our small businesses could go bankrupt due to social distancing regulations. As government-paid salaried employees, it’s easy for us to miss factors that could totally mess up one’s perspective on what should be done and why. Maybe the criteria used to determine an A grade would actually be closer to a C than the C grade because the actions of the population that earned a C may be attempting to balance the need for social distancing with the need for a working economy.

    This is a precarious balance, because it seems so callous to make decisions based on money. Maybe that’s why I hear little from people about it. But, in the long run, we may need to have the courage to risk losing some lives (those who are generally older or with pre-existing conditions) in order to save so many others at risk of suicide, overdose, abuse, and stress related diseases because of the stress caused from imminent poverty (those who may be younger and healthier). Isn’t it interesting that the ones making all these regulatory decisions are salaried and not at risk of the imminent threat of losing all the things that make a life worth living? Rather than being governed by fear (as you mentioned so much in the blog), our decision makers may need to learn about balance to and act with courage.


    1. Steve, thanks for the questions.
      First, social capital does not necessarily specify empathy, though that may be included with the focus on community and the larger collective. At the highly aggregated level of the data used, I can only speculate about empathy. Second, the grading system utilized applied percentage reduction in activities and travel levels to determine a grade. It did not consider the psychological states that produced the behaviors. However, your point about trade-offs is well-taken and indeed leaders across the U.S. are struggling with those difficult decisions. And, it appears these decisions will vary tremendously in the absence of a federal mandate. Indeed, federalism at work.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: