Stanford Business School Study Finds Teamwork Produces Better Decisions.

STANFORD, Calif. -- Good managers know that the proverbial "yes man" isn't much of an asset. But when it comes time to build work teams, many business managers, either by design or through inattention, staff them with people who are prone to think alike. In effect, they've built "yes teams."

As a result, those teams fail to make the best decisions possible--and the organization suffers.

Recent research by Deborah Gruenfeld of Stanford's Graduate School of Business suggests that teams encompassing at least two separate points of view on a particular question make better decisions because the pressure of the minority forces the majority to think more complexly and consider diverse evidence. Gruenfeld gained some of her evidence by analyzing decisions made by the U.S. Supreme Court.

Moreover Gruenfeld, associate professor of organizational behavior, also found that close majorities tend to be more open minded in their reasoning than majorities holding a larger balance of power. This, of course, suggests that token diversity of opinion is probably less effective than true diversity.

She is not the first researcher to look at these issues, but her work adds to previous knowledge by examining the very complex relationship between political beliefs and the balance of power within groups. She also attempts to correlate the nature of decisions--for example, whether they uphold the status quo or overturn it--with the other factors.

Gruenfeld has studied the psychology of power for more than eight years and has published a number of papers, including her Ph.D. dissertation, on the topic with data garnered from decisions of the U.S. Supreme Court.

Why the high court? "A lot of research tends to focus on the specific personalities of leaders. I was interested in showing that the dynamics of relationships among people who work together in groups are a stronger determinant of their behavior than personality. The court is small; you can get a sense of the alliances and allegiances and factions in the group," said Gruenfeld.

She said it is possible to generalize from research based on behavior of the Supreme Court to other fields. She has, in fact, found similar results in studies of student behavior. However, she cautions that the court is by nature a very conservative institution with a built-in bias toward upholding the status quo.

Gruenfeld believes her work contradicts a notion popular in some political circles (and supported by research done by Philip Tetlock of the Haas School of Business at the University of California, Berkeley, among others) that those who support parties and politics on the right are more rigid in their thinking and more intolerant of ambiguity than those on the left.

When groups in a democracy make decisions, the level of complexity in their thinking depends more on group dynamics than on personal ideological preferences, she says. When groups gain power (either political or organizational), they not only act differently than when they were a minority or smaller majority, they reason differently.

"Tetlock's finding that liberal opinions were more complex than conservative opinions was replicated only when the data were drawn from an era when liberal justices held a majority on the court. When data were drawn from the more recent, conservatively dominated era, conservative opinions were more complex than liberal opinions," Gruenfeld wrote.

Her early research on this topic won dissertation prizes from the Society of Experimental Social Psychology and the American Psychological Association, and was reprinted in Key Readings in Group Process (Levine & Moreland, 2001).

Grading the Justices

Gruenfeld's innovative research methods--and the subtleties of the findings--are well illustrated by "Upending the Status Quo: Cognitive Complexity in U.S. Supreme Court Justices Who Overturn Legal Precedent," published in August 2000 in the Personality and Social Psychology Bulletin and written with Jared Preston.

Data for the study are the verbatim records of published opinions written by high court justices. The researchers examined the more than 1,000 decisions handed down by the court between 1953 and 1993, which have been published online (http://www.icpsr.umich.edu/cgi-bin/archive.prl?study=6987&path=ICPSR) at the Inter-University Consortium for Political and Social Research.

Using a variety of filters, the researchers pared down the cases to 115 rulings that overturned legal precedent. They then selected 32 cases for which a majority and minority opinion was written--all dealing with either civil liberties or economic activity. Perhaps the best known of the cases is the landmark Miranda v. Arizona of 1967, which by a vote of 5 to 4 redefined the rights of suspects in police custody.

Each opinion was divided into three equal sections for analysis and scoring. Quotes and simple descriptions were not scored.

The researchers were looking for evidence of both open-mindedness and complexity in the text.

Evaluative differentiation is the recognition of more than one valid perspective, while conceptual integration corresponds to recognition of the trade-offs inherent in different perspectives.

Higher levels of differentiation indicate awareness that there are reasonable arguments on at least two sides of a controversy. "High integrative complexity corresponds to the acceptance of multiple worldviews and the experience of value conflict that comes with understanding the trade-offs among them (e.g., 'Liberal policies protect our right to equality; conservative policies protect our right to freedom')," she explained.

In addition to seeing that majorities reasoned more complexly when confronted with a minority, Gruenfeld and colleagues also found that Supreme Court justices in the majority reasoned with even greater complexity when defending the status quo than when upending it. When voting to overturn precedent, the decisions were more dogmatic.

One explanation: Majorities making major changes are, in a sense, more accountable than majorities defending the status quo. And because they feel more accountable, they may tend to bolster their own positions defensively, rather than engaging in more complex reasoning. In any case, said Gruenfeld, these nuances will probably be an area of further research.

A more recent paper provides a lesson that may be even more useful to generalize from: Examining similar data drawn from the court, Gruenfeld and Peter Kim of the Marshall School of Business at the University of Southern California found that as justices gain power, become the chief justice, or become part of a larger faction, their written opinions become less complex. "They exhibit less consideration of multiple perspectives and less discussion of possible outcomes (i.e., lower cognitive complexity) in their written opinions," wrote Gruenfeld and Kim in a recent paper.

Summing up her work, Gruenfeld said, "Our work on the psychology of power ... not only gives credence to the old adage that power corrupts, but it explains why this occurs. Whereas the classic perspective provided by Machiavelli suggests that power's effects are mostly premeditated and strategic, our research suggests that when power corrupts, it can be without conscious awareness."