What happens when you have a discussion with Professor Edgar Schein, arguably the #1 workplace culture expert in the world and a strong critic of culture surveys, and Dr. Robert A. Cooke, creator of the most widely used organizational culture assessment in the world? It was exciting to see this discussion unfold to a point where both were “blown away” by the amount of agreement and “common ground” that exists between the approaches they advocate.
Culture is a hot topic but remains an elusive concept for many leaders. The ambiguity and confusion are in part driven by the lack of consistency across experts and thought leaders in the culture field. Nevertheless, there were 12 key areas of common ground across qualitative and quantitative culture assessment and development approaches that came out of the discussion. These areas should help leaders, consultants, and others to more effectively manage change efforts.
#1 – Leaders must start by being clear about the business problem or purpose of their change effort.
Edgar Schein: I think the leader short-circuits something that they feel is not right by jumping directly to culture and not being precise in defining what he or she feels is not right. So, when someone approaches me with culture questions or culture proposals, I immediately say why and try to push them toward what problem are you trying to solve. And work a lot on getting very clear in terms of: if we solve this problem, what would future behavior look like? Once we’ve done that, we can say, alright, let’s look at how the existing culture you’ve got will aid or hinder movement toward this new behavior that you think will solve your business problem. So now we need culture diagnosis, not before we’ve identified the future.
Robert Cooke: I think I would mainly reinforce that position, rather than adding to it. I love to see culture surveys used but not if they’re unnecessary and certainly not at the very beginning of an initiative or change program. I think there’s got to be discussions as Ed said about the organization itself–what is the mission, what is the purpose–and discussion about why are we looking to change something, why are we looking to change culture? Is that the right place to look? What problem are we trying to resolve? What initiative are we trying to facilitate? What strategy are we trying to implement? I really do agree that the use of a survey should come a little bit later in the process, rather than at the very beginning.
#2 – A culture survey may be useful under the right circumstances and will only measure some aspects of the culture.
Edgar Schein: The survey requires you to define a set of items and, as I encounter a culture, until I’ve talked to a lot of people and a lot of groups, I wouldn’t have any idea of what a survey might look like that would be relevant to that particular organization in its particular situation. So that’s why I think an upfront statement that I have a complete culture survey just doesn’t make sense. Culture is too broad. However, I’m very respectful of surveys that have taken what we do know about individual and group behavior and how organizational systems work and said okay we can build surveys from theory and we even know a lot about what works better and worse in organizations. So if a survey is built on what we think we know, is generally true, and we’ve looked at the problem that this organization has, then I can see a complete reason to say ‘well then let’s look at some general surveys that will enlarge your perspective on your problem’. You will be exposed to a whole bunch of dimensions that may or may not relate to whether you can improve the situation that you have. Some of those dimensions will indeed be irrelevant, others will be central, but you’ll have more to work with.
Robert Cooke: We’re actually very careful to say that the Organizational Culture Inventory (OCI) measures SOME of the behaviors that might be expected within organizations and which can have either functional or dysfunctional consequences. I don’t think we pretend to measure the entire array of behaviors that might come up in an organization including, in certain cases, the most important ones. But, you know, we have a pretty good conceptual framework with three distinct factors, three distinct types of cultures and related behaviors: constructive, passive and aggressive. I do feel that we do a pretty good job at measuring those. What we need though, and what I like to see consultants do, is to add to the survey before it’s administered, and to do so in an informed way on the basis of their discussions and on the basis of the questions they’ve asked to see if there are additional behaviors that are particularly relevant to the organization in which the survey is being administered. We add them into the OCI; we have a specific section for doing that.
#3 – The client should make the decision on whether to use a survey or not.
Edgar Schein: The point about the client being involved in the decision to do the survey seems to be crucial for another reason that hasn’t been brought out yet. If I’m an employee and I suddenly learn that I’m going to do a survey, I don’t know if I’m going to be very serious about it or even do it at all. So if part of the survey is we want as honest an answer as we can, and we want as many people as possible to answer it, then it seems to me the introduction of the survey becomes part of the change process and the consultant and the client have to jointly own that and figure out how are we going to get our employees involved in taking this thing so that 1) They’ll take it seriously and 2) They’ll all answer. Because small samples or casual throwaway answers are gonna hurt. I see the people who jump into the survey have no way of knowing whether the answers are any good or not and I presume that if we insert it in a process, then the existing management can say to its employees: we really need you to take this seriously, we really care, we want these results, and promise feedback.
Robert Cooke: Yes and there should be discussions within the organization about some critical factors. Ed mentioned sampling versus population, for example. Other factors have to do with things like confidentiality and whether people are mandated to take the survey on one hand versus, on the other hand, strongly encouraged to take the survey and answer as honestly as possible. You know the list goes on and on. And organizations that tend to not respect their employees so to speak also have a tendency to jump into surveys and to administer them incorrectly.
#4 – Leaders should give assurance the feedback will be shared and acted upon.
Edgar Schein: If you don’t promise feedback and act on it, you just say do this and we’ll see what we get, a lot of employees will give you casual meaningless answers.
Robert Cooke: Administering them incorrectly without assurances of confidentiality, without assurances that the data are going to be used, leads to more problems within the organization including basically invalid results on the survey itself.
#5 – A team should be engaged to help identify how the survey will be set-up and used.
Edgar Schein: Maybe one thing that can come out of a discussion like this is some brand new ideas. I’ve always worried about promising confidentiality because that tells the employees that we got a problem here. If you feel that if you give me a straight answer…I punish you…that itself is the problem. So what if we said employee engagement is our task and we’ve got some very good surveys but let’s do it this way. We’ll start with talking about it, employee engagement, and we’ll create an employee group that will decide how this will be administered and how much confidentiality do we really need? Involve them, use the survey to start the process of engagement. Why go to all this super structure instead of saying employees, here’s our problem, we want good answers so why don’t you employees tell us how to administer…do you do that?
Robert Cooke: Some of the best administrations of the survey have come about in organizations that have put together an organizational change team, for example, where you have a diagonal slice of the organization represented on the team. We did a utilities company, I’ll never forget this, they were talking about the very issue that you were just addressing and the change team members [including linemen and operators]were coming in special for this meeting… They discussed things like how much confidentiality is needed, how much demographic information are we willing to share on the survey [in the background information section], should we be able to opt out of a question… I think most importantly, that works. It leads to involvement, it leads to better data, it sort of sets the stage for the acceptance and the use of feedback once it’s presented.
#6 – Only survey as part of a broader change effort that also includes qualitative approaches (though some disagreement remains around treating the survey as a scientific tool versus making it part of the intervention).
Edgar Schein: You’re never going to get perfect scientific data in the human system because of the nature of the system. So, therefore, you should always think as an interventionist, not as a scientist. You should do as much as you can to be logical and consistent and stuff, but you should always be thinking…how can I intervene to help? And how, therefore, how can I design a survey, and administer a survey, and use a survey to be helpful? Yes, I want it to be reliable because it’s got to be logical and people will want to know. And yes, they will want to know that it relates to some other things, but they sort of take that for granted. What they really want to know is why the hell am I doing this? What is in it for me to answer 140 questions unless it’s presented to me as a helpful tool, to solving some problem that I experience?
Robert Cooke: There is no survey I know of that can replace the consultants, the leaders, and the members of the organization getting into the problem and discussing it, and figuring out what’s going on. The strict adherence to the statistical results on a survey…on a feedback form saying that these two things are the best and these two things are the worst…only goes so far and that feedback does not necessarily mean that it’s most appropriate for the organization to focus on those two factors that they got the lowest scores on. It might be a starting point for discussion, but that’s only what it should be. It should be to open up the discussion and talk about what really matters here. What is going to have the greatest impact? What is feasible to change? What will get the acceptance of everyone at various levels? Can’t do that purely through numbers.
Next Week – Part 2 of this post with common ground areas #7 – 12. See the full video from this discussion below (covering blog post part 1 and part 2).
New Announcement – We’ll be holding the 1st Annual Ultimate Culture Conference on September 29 at Willis Tower (Sears Tower) in Chicago and Edgar Schein will be one of our keynote speakers. Watch this site and HumanSynergistics.com for more details about this amazing event soon.
The complete culture discussion video: