Help out this student by taking his survey

This week I got an email from Matthew Miller, a student from Loughborough University, with a request to put out the word about his questionnaire.

“This dissertation aims to investigate the level of success within IT projects within the last 3 years . The results will then be analyzed to find any common factors to success and failure, to see if some projects are destined to fail and to see if certain industries, type of projects or methodologies/models are more likely to fail. The final outcome will be statistically supported advice for IT project managers. This research is about both project success and failure. It is through understanding failure that we can begin to improve. As such, please do not view failure as a negative in this context and please include projects that have failed as these are very relevant to this study. “

If you participate, you will get a copy of the results.

Take the survey:

IT Project Managers
Anyone involved in an IT project

Bas de Baar helps people find ways to enjoy the diversity of human interaction in their organizations so that they can get out of their own way and achieve their goals. - Way-New Collaboration: What I Meant To Say is a post from: Project Shrink.

Photography by Dylan Parker

A couple of years ago I was asked during a sales presentation what I thought was the most essential ingredient for a successful project. My answer was “mutual trust”. People
in the room were staring at me like I was some kind of Softy Oozy New Age Treehugger. I switched very fast to the normal “plan-and-control” crap to get happy faces again.

Today I would provide the same answer to that question. Trust is essential to doing successful projects, and therefor a core concept for Project Management. But to avoid this “Treehugger” image, let me give you a view on how trust can be modelled, so we put some hands and feet to this rather abstract idea.

As a starting point we take the Prisoners Dilemma (PD) as discussed in this posting. In essence it is a situation where

• 1) if people cooperate both have success,
• 2) if one person is taking advantage of the other (defect) this person has an even larger benefit, but the other suffers a loss,
• 3) if both persons defect they loose both. In a situation like being in prison, you have only those two options, cooperate and defect.

However in more general circumstances, like projects, you can have a third: don’t play, walk away, just exit. For this posting I will simplify a project to a series of prisoners dilemma’s where the strategies can be cooperate, defect and exit.

Earlier I talked about the idea of “iterative PDs” where an enormous series of PDs are performed behind each other and the choice of strategy is based upon the outcomes of the previous iterations (remember Tit-For-tat? ). History plays an important role in choosing a strategy. Central question is “Do you trust the other party to cooperate?” “Trust” is defined as “have confidence or faith in” but also as “reliance: certainty based on past experience”. Based upon the things that happen in the past, you adapt your strategy.

Having Only One Shot

But what if you haven’t done a zillion iterations? What if you meet a person for the first time and you are confronted with a prisoners dilemma? Researchers call this the “one-shot prisoners dilemma”. In a situation like this, people are trying to determine the “trustworthiness” of others. They are trying to read “telltale signs”, look for behavior or other marks that they identify with trustworthiness. This might be as simple as being friendly and saying “hello” every time you see someone down the hall. Perhaps you have automatically more trust in someone wearing a suit, or a person with PhD behind his name. The idea is that you are trying to detect signs of trustworthiness, whatever
that my be for you.

Next to this detection, the projection of your own intentions plays a role in the decision of the strategy; if you want to cooperate you are more likely to be biased into “seeing” the other as trustworthy. So, we use projection and detection as a mechanism to compensate for the lack of history one has in one-shot Prisoner Dilemma’s.

How people detect the tell-tale signs of trustworthiness is not only based upon behavioral markers that society associates with it; it has also to do with the similarity of the other with you. Persons that are more viewed as being equal or “the same” or more likely to be considered honest and sincere towards you. Translated to terms of social networks: people closer in social networks are more likely to consider each other trustworthy than people further apart.

This is not a one dimensional thing, people are associated with multiple social networks and groups. And every social group has its own rituals and signs that communicate its uniqueness towards the world outside the group. If you have a lot of aspects associated with a certain social group, you will more likely be considered trustworthy by members of the same group.

The only question remaining is: do you trust this model?

This posting is largely based upon THE EVOLUTION OF TRUST AND COOPERATION. BETWEEN STRANGERS:. A COMPUTATIONAL MODEL

Bas de Baar helps people find ways to enjoy the diversity of human interaction in their organizations so that they can get out of their own way and achieve their goals. - Treehugger Project Management: Is Trust Important? is a post from: Project Shrink.

I love looking at situations through different glasses to get several perceptions. I love reading and thinking about human behavior. And I adore authors that take on a multidisciplinary approach. So I must be ecstatic when all three are combined. I am.

In Critical Mass: How One Thing Leads to Another Philip Ball explains to us what happens if you try to explain human behavior by using concepts and theories from physics. It is a treat. Ball writes clear and entertaining, and the ideas behind such a “physics of society” make your head spin. Not necessarily because you are staring at the ultimate truth, but just because you have never looked at your own behavior like that.

Physics cannot tell you how a certain individual particle is going to behave. However scientist can explain how a large group of particles, like a liquid, is going to act. Using this principle on humans you have to focus on collective behavior. The advantage is collective behavior smooths out individual unpredictability. Ball keeps emphasizing throughout the book: “the interdependence of people makes group behavior different from simple extrapolation of individual behavior”. He keeps hammering on the fact the global phenomena are caused by local actors. A large group of individuals interacting make up the “behavior” of the group and not some other way round.

The book explains a lot about the interactions of individuals. I want to highlight one in particular and apply it to the area of Project Management. Facing a certain situation, a person has to select a strategy to interact with another individual. Are you going to cooperate, or are you going to be egoistic? How does this strategy selection work?

How a strategy is exactly chosen, remains a little mystery. Some light can be provided though using Game Theory as treated in the book. In this part of mathematics probability is used to evaluate certain plays. A popular example in this context is that of Prisoners Dilemma. In this posting I will use my own version of this game, the Project Managers Dilemma.

There are two Project Managers and both of them need a certain programmer from a shared pool of resources. They have both two options:

• Share the resource, so they have them both for half the time or
• Claim the resource 100% exclusively for their own project.

In this particular game, if both Project Managers claim the developer, they loose both, as no one will get the programmer (each PM will only get 1 point, as the resource is still available). If they both share the resource, they will earn each a three point reward. If PM 1 claims, and PM 2 goes for option share, PM 1 gets the resource, and is rewarded 5 points, and PM 2 gets nothing and the sucker bonus of 0 points.

Of course, the goal of the game is to get as many points as you can. One round of this game is no fun, and will tell us not much. But playing multiple times this game is more interesting. Based upon previous outcomes the Project Manager can adapt its strategy. Computer simulations have shown (Ball, 2004) that one of the most successful strategies is called Tit-For-Tat. The first round the PM will share (to show he is willing to play nice). And after that, he will always use the strategy that his opponent chose in the previous round. In this way, if the other PM will keep on claiming, he will not be the sucker all the time. And if the other player wants to cooperate by sharing, the PM will go along, as that strategy is seem to be best for both of them.

Bas de Baar helps people find ways to enjoy the diversity of human interaction in their organizations so that they can get out of their own way and achieve their goals. - Prisoners Dilemma: Do You Cooperate In Your Project Or Are You Egocentric? is a post from: Project Shrink.