Edgar Schein on Values

In this article I will discuss the problem of dishonesty and talk about how Edgar Schein can help address the problem.

In my work I have states repeatedly that worldview drives all kinds of actions. People pursue their own interests necessarily and tautologically. Those interests are determined by worldview and context. Worldview is essentially no more than a stock of heterogeneous information which has been accepted by a mind or individual.

To make predictions in my systems model we need to understand this stock of accepted information. The easy way would be through survey. Many times this works just fine, but there is a problem of dishonesty which prevents many times from being all the time. One solution which many economists have embraced is simply to observe action and not to allow ethnography or survey, but this is often insufficient for an accurate picture. Other kinds of sociologists have developed tools which can address this issue to some degree.

Edgar Schein is a well known author in the areas of organization and sociology, although by no means the best author. In fact I think he is quite bad, but he is well known and discusses something I want to talk about in his book Organizational Culture and Leadership.

In that book Schein discusses a method of analyzing the organization of a culture at different levels. His framework can overcome the problem of dishonesty. There are 3 levels:

1. Artifacts
2. Espoused Values
3. Basic Assumptions

[caption id=\"attachment_2033\" align=\"aligncenter\" width=\"300\"] Click the link to be taken to a Wikipedia article for Edgar Schein. Image credit to Wikipedia.[/caption]

Artifacts are things which you observe such as things people have and the ways in which they use them. Language can to some extent be considered an artifact. For example, speaking with a southern accent, in Spanish or with excessive profanity would be considered artifacts. Espoused values are what people say they believe. While the line blurs, essentially the way people talk is an artifact while what they mean is espoused values. Basic assumptions are the things that go without saying. These are the pillars of culture in the collective entity, but they can also be seen as pillars of personality in the individual. If I wave at a taxi they may assume I want a ride, but if I wave at a friend they will assume I am only saying hello. Assumptions are the unobserved second half of the equation in communication.

Basic assumption analysis can to some degree reveal when a person is providing misinformation, whether that be through failure to communicate, communicating falsely or communicating truth in a less than optimal way. Economists already do this to a large degree. For example, if someone buys something in a free market the economist assumes they did so because they wanted to. This is an example of basic assumption analysis at the artifact level.

Basic assumption analysis can be applied at the espoused value level to come up with something like a rough honesty index. There are several ways to do this and some are better than others, but the rough idea is simple. Observe people, make notes and survey them. Then compare their actions to their survey answers and you can deduce the rough degree to which they are honest in their espoused values. This can sometimes also be done through a survey-only method involving clever trick questions. The observations need not be live action observation. Research data can be used for objective observation. One example would be to find a person's income on record and then survey them and ask them how much they earned in that time period.

One problem remains. It is the classic question on whether a person is a liar or just stupid for getting things like their income in a particular year wrong. The usual answer is both and for most applications it really doesn't matter. If it does matter then there are certain techniques which can be used to separate the two to effects to some degree. One technique would be to establish a person's average competency level in activity such as memory, presume that the level holds for the activity in question and assume that the difference, if any, is due to a combination of random variation and dishonesty. There are also more sophisticated techniques, but that is for another time.