Sitting here at the Microsoft Patterns & Practices Summit (West 2007) and just heard a presentation by Grigori Melnik. His topic was primarily around empirical evidence (surveys from the wild) regarding agile software development adoption and results. However, one of the data points he mentioned really struck me as incongruent with one of my long-held beliefs.
A simple slide up on the pair of giant screens said, "Happy Teams -> Productive Teams". He asked how many people believe that to be a true statement, and then clicked his mouse - a big red X dropped over the arrow along with the text, "Myth".
Apparently, Grigori's work with organizational behavior experts and analysts has convinced him that, in fact, there is no demonstrable causal relationship between a team's happiness and their productivity. However, they have demonstrated a direct relationship between happiness and turnover. In other words, if people aren't happy, they will work productively until they leave, and when they do leave, the churn will not negatively impact the team's overall productivity.
I'm still trying to fully grasp that notion and figure out if I can synthesize it with my personal experiences and values. For the time being, sitting here in this room with another presenter about to take the stage, I am going to have to politely disagree with Grigori - regardless of how much data he has to back it up. I'll dive into this more.