I’ve written a chapter for a forthcoming book edited by Harry Boyte (Democracy’s Education: A Symposium on Power, Public Work, and the Meaning of Citizenship, Vanderbilt University Press) in which I summarize evidence that colleges and universities can improve the economy by teaching their students civic skills and by being good institutional citizens, participating in local networks for community development.

I think the evidence is reasonably strong. But, like all empirical claims, it has exceptions and caveats. And even if civic education and engagement really do pay economic dividends, something else could work even better: for example, distance-learning, educational video games, or installing surveillance cameras in schools.

Thus we must be careful about how we generate, interpret and use empirical findings. This is where the emerging idea of Civic Studies provides guidance.

Often, social scientists presume that their job is to study a real-world practice that is already fully developed to learn whether—and why—it “works.” Usually they define success in terms of the objectives of the practitioners or their funders. In this case, we would ask whether college-level civic education and engagement generate what politicians demand: jobs.

But nothing simply “works.” Success always requires experimentation, assessment, adjustment, reflection, and new experimentation, in an iterative cycle. By the same token, many things can work if they are developed properly. One could start with civic engagement or with surveillance cameras in schools and improve either one until it enhanced students’ employment prospects.

In the best cases, the researchers who study a given practice are part of the reason that it works. They contribute to its development by offering their data and insights. They choose to work on this practice rather than something else because of a more fundamental commitment. I, for example, have studied deliberation. I want deliberation to work and I hope that the research that I produce will contribute to its success. I have no such commitment to surveillance cameras. I would not study them or strive to improve their impact.

The reason for my hope in deliberation is fundamentally moral. I think a world in which people reason and work together is better than one in which they achieve the same levels of security, income, or welfare without freely collaborating. Deeper down, I believe in a theory that the good life is a life of freedom, reflection, and mutual commitment.

Thus I hope that civic education and civic engagement boost employment because I am fundamentally committed to civic values. My colleagues and I seek evidence of economic benefit to persuade policymakers to support what they should support anyway. If the economic evidence is favorable, we will use it strategically to expand support. If not, our values and commitments should encourage us to improve civic education until it enhances democracy and also produces jobs. Regardless of the empirical results we find, we owe a public explanation of our core values.

A public defense of our values also yields criteria by which to assess the practices that we have been studying empirically. For instance, my chapter for Harry Boyte’s book is about college-level civic engagement (an input) and jobs (an output). I discuss the empirical link between the input and the output, as they exist today. But both are subject to criticism and change.

Today, many civic programs basically take the form of volunteering. But civic education can be reconceived so that it is less about volunteer service than about working on public concerns, where “working” implies serious commitment and accountability for results. “Public work,” in the phrase championed by Boyte and colleagues, means work that is done in public, by diverse citizens, on common issues. Reconfiguring civic education at the college level to look more like public work would satisfy core values that Boyte and colleagues have defended well. It might also strengthen the impact of civic education on jobs and careers. Students would be more likely to learn skills useful for employment if their civic experiences in college were more like paid work.

Meanwhile, jobs could become more public. A given job might serve only the interests of the employer and deny the worker any scope to address community problems in public with diverse other citizens. But even if the employer is a for-profit firm, the job can promote and encourage public work. For example, I presume that the corporate executives, government officials, and labor leaders who attended meetings of the Lehigh University board contributed insights from their daily work to the conversations about Lehigh and Allentown. They then brought ideas from those discussions back to their jobs. If that is true, they were doing “public work” in the Lehigh boardroom and in their own offices. Public work is obviously harder for low-paid service workers and low-ranking bureaucrats, but within many industries and professions, a struggle is underway to recover their public and democratic traditions.

If we made civic education into public work and also created jobs of greater public value, then the alignment between civic education and employment would be stronger and we would find more impressive evidence of economic impact. The data would then satisfy governors and presidents who want to see colleges produce jobs. More importantly, we would be building a better society and the educational system to support it.