Single-Loop Learning, Double-Loop Learning and Homelessness – Part I – Housing First

One of the most useful concepts in organizational learning theory is the idea of single-loop learning and double-loop learning, first introduced by Chris Argyris and Donald Schon in the 1970s. According to Argyris and Schon, an organization engages in single-loop learning, “when the error detected and corrected permits the organization to carry on its present policies or achieve its presents objectives... Double-loop learning occurs when error is detected and corrected in ways that involve the modification of an organization’s underlying norms, policies and objectives.”

 

In other words, and as this excellent infographic illustrates, in single-loop learning, we do not question the underlying assumptions (“Why”), under which we operate, only the strategies and techniques in use (“What”), whereas in double-loop learning, we question the actual underlying assumptions.

The idea of challenging the governing variables, under which we operate, can be extremely disconcerting. Padgett, Henwood and Tsemberis recount how, in one of their first attempts to seriously study the idea of Housing First, they had to contend with this.

At that time, the idea of Housing Readiness, that those experiencing homelessness had to be gradually prepared for housing, was widely accepted. The idea at the core of Housing First, that those experiencing homelessness could be housed immediately, and that the housing itself would have a stabilizing effect, that could then help them address the challenges in their lives, at their own pace, radically questioned the underlying beliefs prevalent at the time.

 

As one might expect, to study the idea of Housing First, Tsemberis and his team randomly assigned some clients to a traditional Housing Readiness program and some clients to the new Housing First program. This where they ran into a problem. Case managers involved in the study raised an ethical question. Since they “knew” that Housing Readiness worked, their values could preclude them from participating in a program that assigned some clients to a program that did not operate under this framework. Tsemberis and his team had to promise case managers that any client that “washed out” of the Housing First program, would be immediately offered a spot in a traditional Housing Readiness program.
 
The “rest of the story”, as radio broadcaster, Paul Harvey, might put it, was that Tsemberis’ double-loop learning paid off. Housing First was shown to be much more successful. In fact, at a certain point in the study, case managers came back to him, and once again raised an ethical concern. They could not, in good conscience, continue to participate in the study if the clients participating in the Housing Readiness program were not promised a spot in the Housing First program, once they “washed out” of the traditional program!
 
Continued study of the idea of Housing First led to it becoming the accepted model by the consensus of researchers in the field, and hence federal administrations across the political spectrum. The resistance to this idea having not died out, though, exhibits the strength governing variables, especially those rooted in strong beliefs and values can have. Double-loop learning is hard.