February 2006
Vol. 3, No. 8
This newsletter is a basic HTML file, giving you control over the appearance through your Web browser's options and window controls.
From the Editor: In December, before I made my mistake and missed our January "meeting," I suggested that each of you start noticing a negative reaction to some event, a reaction you would like to change. As an individual, it could just be a thought; in a team, it could be words or actions the team tends to show in response to some event. Now that you've had an extra month, you should have a really good idea of what triggers the event you chose. I wrote, "Only by changing those thoughts can you then change the actions you take. The sequence is always this: stimulus event; internal response to the stimulus, which becomes the stimulus for your external response; then the responding action. The opportunity for change lies in that middle step, in controlling how you respond internally to the initial stimulus."
The next step is to start specifying how you would rather react. The next time the event you chose occurs, think of what you would prefer to have happen. For example, I mentioned I am trying to cut down on my negative thoughts by recognizing the unmet desires that cause them. Now, each time one of those thoughts comes along, I name the desire and come up with an alternate thought. Recently I saw someone driving the wrong way down a one-way street and had the unexpected thought that they were stupid. I stopped myself mentally to ask what desire had caused this. I recognized that because my next work contract is not yet lined up, I had an unmet desire to feel secure. This in turn led me to put this person down to try to build myself up. So I inserted a new thought that I would prefer to be my response the next time I see someone make that mistake, which happens fairly often as I walk along downtown Seattle's many one-ways. The thought I would prefer is, "I've done that before—I'm sure he's upset."
Putting this into a team context, let's say your team gets hit with frequent changes to a customer's orders (defining "customer" as whoever receives your team's output, in or outside of the company). We'll also assume they lose some time and goodwill by griping about it amongst themselves. What would you like them to say instead? Maybe, "Cool—A new chance to show off our customer service!" Or at least, "No problem—we get paid the same either way." This, then, is your next step. Take the item you came up with from the December lesson and spend some time identifying a couple of reactions, internal or spoken, that you would prefer over the ones you usually get. When the trigger event arises, don't try to force the new response: just say in your head the new one after the old negative one occurs. If it's a team issue, don't raise it with the team yet. But you can start considering ways you could suggest the new response in a positive manner. Kindness is always a good way to go.
Wish I went with that more automatically myself. But I'm working on it!
What if every scientist who ever studied how to make groups of people work together better could sit in your offices or factory for several days or months and tell you everything they know? How much stress could you get rid of? The next best thing is now available on a consulting or contract basis: me! Contact me today to find out how I can make your working life better.
Study: Ongoing coaching of drug research teams over eight months lead to large improvements in team operations and performance in a study at pharmaceutical giant AstraZeneca in Sweden. The authors, project management (PM) professor Kina Mulec at the Chalmers Univ. of Technology, and Jonas Roth of AstraZeneca's PM Support Office, point out that use of "coaching" in the business world has grown, defined at AstraZeneca as "'the ability to take the opportunities presented by the job itself and use them in a conscious manner to improve the knowledge, skills and performance of the learner.'" But few scientific studies have been done on coaching, and fewer still on long-term results or on coaching of teams as opposed to individuals.
Mulec and Roth arranged to provide coaches to two project teams at AstraZeneca over eight months. Each team was given an external professional coach paired with someone inside the company. The coaching did not involve the standard offsite meetings usually used for "teambuilding." Instead it was done "at regular team meetings while the project team did their normal work, with almost no extra time being reserved for coaching interventions." The coaches also provided individual coaching to the team leaders. Data was gathered using: questionnaires at the start and end of the study; observations of the team meetings; interviews of the coaches and team members; and team workshops.
The researchers claim the results "showed that the 8 months of coaching had resulted in a stronger and more supportive team environment." Specifically:
Respondents specifically stated that the coaching was responsible for changes. For example, "Many of the improvements during their meetings were related by the respondents to the coaches taking on the responsibility of intervening, pointing to obstacles such as getting stuck in discussions or lines of reasoning." But the behavior changes from the coaching were evident even when the coaches weren't involved. Several team members noted a meeting in another country where the other attendees "were not actively participating in the meeting, but were instead sitting along the wall, whispering and commenting. The team members explicitly objected to this behaviour, inviting them to sit at the table." This would not have happened, the members said, had it not been for the team's shared value about full participation that had been created through the coaching.
Application: This is another one of those studies I report nervously because it so fully supports what I say in my marketing about short-term "teambuilding" exercises versus team coaching. (I don't want you readers to think I go looking only for studies that support my sales efforts.) That caveat given, it does fit perfectly with my observations as both the recipient of teambuilding and a provider of team coaching: A period of regular interventions using a team's real-life challenges are required to effect lasting change in team behavior. By observing and intervening as part of the team's normal interactions, the coaches were able to help the team recognize patterns in the ways members worked with each other and the impact those patterns had on teamwork. During the study period, behaviors "hindering learning, creativity, change and innovation" were changed into patterns that supported those team traits.
The study also provides support for my contention that improving meetings is a key element of fixing overall team communications. Many of the respondents' comments focused on the meetings, with this one pretty typical: "Before, we had lousy project meetings in my opinion, such as last minute changes to the agenda… delegates who did not come, (changes to) the purpose of the meetings, the meetings were long…" Another said that before the coaching, the team made timid, unclear decisions during meetings that were not understood the same way by all members and often not implemented, but that after the coaching the team made real decisions and followed through on them.
I have to note some major limitations to this study before I offer any advice. It did not compare coached teams versus similar non-coached teams, so we can't be sure non-coached teams would have been any different after the eight months. Also, the researcher was heavily involved in coaching the coaches, so the results may be biased, and the article does not show the data, so I can't double-check the authors' conclusions against it. In short, although it is a step above anecdotal evidence, it's a far cry from hard science.
That said, it nonetheless is another bit of evidence supporting the conclusion I gathered after looking through hundreds of studies to write The SuddenTeams® Program: Short-term teambuilding exercises do nothing to solve long-term challenges. If you really want your team to change, take time out regularly to implement real changes in your meetings like those you read here in TeamResearch News. If you don't take time out for fire prevention, you will always spend all your time firefighting.
Source: Mulec, K., and J. Roth (05), "Action, Reflection, and Learning: Coaching in Order to Enhance the Performance of Drug Development Project Management Teams," R&D Management 35(5):483.
Study: Teams that liked their internal processes and their work tasks had members who were absent from work less and helped each other more, in what appears to be the first study of group-wide attitudes toward group tasks.
Australian researchers Claire Mason and Mark Griffin of the Queensland Univ. of Technology realized that team job satisfaction is usually measured by asking how each individual member of a team felt about his or her job. They believed that asking members how the team as a whole felt about things would bring different results, and that turned out to be the case. A formal work group can develop a shared view of a topic, through discussions and daily interactions, which each individual would not otherwise have identified on their own.
The findings came from surveys given to 514 employees in 71 groups in nine employers. "The organizations came from a variety of different industries operating within both the public and private sector, and the functions of the work groups varied widely, from management to customer service to the replenishment of stock on a factory floor," the study says. Among their findings were:
Be aware that team task satisfaction was different from the team's general emotions about its work (such as enthusiasm and distress). Task satisfaction included "the group's satisfaction with the task itself; the group's internal work environment; and the group's satisfaction with facets…such as rewards, the physical working conditions, and the wider organizational environment," the study says. Overall task satisfaction was not directly related to manager perceptions of team performance. But the authors suggested that might be a problem with the questionnaires, not an accurate reflection of reality. It is very hard, they noted, to come up with an accurate way of objectively comparing performance when the teams involved do so many different things.
One other point of interest is that the researchers did not necessarily consider these groups to be "teams" as the term is defined by scientists (see "Definition of a Team" on the TeamTrainers™ site). But use of the term is so prevalent—that is, employers were so determined to say they had "teams," not just "groups"—that the researchers had to change the word "group" to "team" on their surveys.
Application: Several results from this survey captured my eye. First, the strength of positive emotions people perceive their team members to have was highly correlated with task satisfaction as defined above (+0.67), but not perfectly linked (not +1.00). More importantly, negative emotions weren't related at all. So you simply cannot assume that because no one is griping, everyone is happy about their deliverables, processes, or work environment. Or, that because everyone says they enjoy their work, you cannot do more to improve their team satisfaction.
Second, although the way the study measured group performance might have masked a direct link between that and overall satisfaction, there was a strong link to the helping and collaboration behaviors that many other studies have shown to improve performance. A key finding for me was the importance of the work environment within the team: the correlations to satisfaction with the internal working environment were +0.59 for individual job satisfaction; +0.53 for the perception that other members were happy with the team; and +0.33 for sportsmanship. This means you ignore things like creating team goals, project plans, and internal procedures (such as procedures for hiring and for mediation of member conflicts) at your team's peril.
Finally, just because you call a group a "team" doesn't make it one. They are not a team if, for example, they:
Source: Mason, C., and M. Griffin, "Group Task Satisfaction: The Group's Shared Attitude to its Task and Work Environment," Group & Organization Management 30(6):625.
Study: Are you "the master of your own fate?" Or do you believe the Fates master you? How the members of a team answer those two questions can have a major effect on how they should be managed for maximum financial performance, according to a new study by three European researchers. One reason may be the effect this "locus of control" (sense of mastery) has on whether the team gathers the information it needs for optimal performance.
Teamwork scientists have already established that, as common sense also suggests, teams are more likely to meet their goals if they are better at getting the information needed to make good decisions. Unfortunately, the study, using working managers in a sophisticated business management game, found that helping a team learn to go after that information is not a simple issue. If most members of a team had an internal locus of control—believed they had significant control over what happens in their lives—the team was better at getting information, which in turn made them more successful financially. Not surprisingly, teams that felt like they didn't have much control over their lives needed a strong leader to help them get the information, which again helped them be more successful. Whether the leader had an internal or external locus of control didn't matter.
But average locus of control can look very different in two different teams. To illustrate, on a scale of 1 to 10, the average score for each of two four-member teams would be a 5 whether the members scored 4, 5, 5, 6 or 1, 3, 7, 9. In the first, every member is basically average. In the second, two members feel a lot of control and two feel very little. You'd think the latter team was more likely to need help getting information than the former, because of the wide variation in scores, and therefore would do better with a leader. But in fact, teams like that did better without a leader than did teams of more-similar people, "which is difficult to explain," the researchers admit.
This information came from the International Management Competition, a 30-year-old simulation game developed in the Netherlands. The study explains, "Organizations generally enroll teams of (young) managers for training purposes…" in the game, which is based on 1,400 business parameters to creates "a very sophisticated and realistic simulation of a multifaceted business environment" in which "small teams run fictitious multiproduct production firms" by making decisions about "a broad range of business issues." They compete in groups of five teams each, making decisions in 37 areas over six periods of two weeks each, getting back performance feedback at the end of each period to adjust their approach. They also could ask for additional specific information for which they had to pay a fee (I assume this was a fake fee from a limited "budget"); this action was used by the researchers to measure a team's information gathering. Personality tests were used to measure locus of control, and questionnaires at the end of the three-month game asked about the team's process quality. If at least half the team members said one member tended to lead the decision-making, that team was considered to have a "leader" in the analysis.
The authors gather several conclusions from the results. First, just as "internal" individuals—people who feel they control their own fates—are better at going after information needed to optimize a decision, the same is true for teams that are internal as a group. "Specifically," the authors write, "adding internals to a team…is likely to increase the team's information-processing capacity, resulting in more information acquisition behavior and, as a result, better team performance. Second, external teams clearly gain effectiveness from having leaders." However, when the team has a wide range of internal and external styles, a leader can somehow cause the team to gather less information (and thus be less successful as a team). Since no earlier study has looked into this issue, the best the researchers could do was to guess why this is the case. They suggested that a team with a range and no leader works extra hard to get enough information, to overcome the personality differences, but is likely to just rely on a leader to do that if there is one. Of course, I'll add, one person cannot collect as much information as several people working together could, so the team ends up with less data.
Application: The authors point out that "there does not seem to be a best way to structure a team." In this case, for example, whether to have a self-directed versus a managed team would depend on how individual team members scored on a test measuring locus of control. But there are a huge number of problems with applying that kind of thinking around the office. You rarely get that level of control over the makeup of a team; many people are on more than one team (when you throw in short-term project teams, at least); and other personality traits have different effects, all of which means you'll never get a team whose set of traits point perfectly to a particular set of team leadership structures and techniques.
The way around this, in my opinion, is to establish rules and processes for decision-making that incorporate best practices for any team's success. In this case, have your team adopt a formal procedure for decision-making that includes answering specific questions about the problem. Here's an example of questions related to process-improvement from my SuddenTeams® Program:
That way, team members who don't feel especially motivated to get information on their own can be helped by the formal guidance, motivated by positive peer pressure, and/or assisted by other members with external loci of control.
Source: Boone, C., W. Van Olffen, and A. Van Witteloostuijn (05), "Team Locus-of-Control Composition, Leadership Structure, Information Acquisition, and Financial Performance: A Business Simulation Study," Academy of Management Journal 48(5):889-909.
Study: Being the lone dissenter in a group is far more difficult than disagreeing when you have at least one supporter—even when you're a U.S. Supreme Court justice, according to a review of case votes over 50 years. In the 4,178 cases studied by sociologist Donald Granberg of the Univ. of Missouri and political scientist Brandon Bartels of The Ohio State Univ., only 10.4% of the votes were 8-1. In comparison, 13.7% went 7-2 (about 32% more), and around 20 percent went 6-3 or 5-4. More than a third of the decisions, by far the largest percentage, were unanimous.
For those readers outside the United States, the Supreme Court is the highest court in the American justice system. Its nine members ("justices") are appointed by the President and confirmed by the legislature's upper house, but then serve until retirement—and often, until death. Since the justice system has equal power to the Presidency and the Congress, and justices cannot be fired, there is no external pressure upon sitting justices to change their votes. Any "pressure" comes only through the force of persuasion from other justices, politicians, and the news media.
As Granberg and Bartels point out, a 5-4 decision has just as much legal weight as a unanimous one. "Many people believe that a unanimous decision by the Supreme Court will command more respect than a split vote," however. That's why the Chief Justice in 1954 lobbied hard to turn a simple majority into a unanimous decision in a famous case that ended apartheid in public schools.
Using simple math, each type of vote breakdown should happen 20% of the time, since there are five possible "scores." Using a more complex model of random chance—essentially, a computer flipping a coin on each possible way to get each possible score in each case—you would expect a 9-0 decision only 0.4% of the time, with each of the other scores more likely up to the score of 5-4 happening almost 50% of the time (see table below). However, this method of "predicting" scores includes every different way that, for example, an 8-1 vote could take place (Justice A dissents or Justice B dissents or Justice C dissents, etc.), which the authors say may not matter for these calculations. Here's the full comparison:
| Vote | Chance % | Even Split % | Actual % |
| 9-0 | 0.4 | 20 | 34.7 |
| 8-1 | 3.5 | 20 | 10.4 |
| 7-2 | 14.1 | 20 | 13.7 |
| 6-3 | 32.8 | 20 | 19.8 |
| 5-4 | 49.2 | 20 | 21.4 |
Using the "even split" model, which comes closer to the actual data, the authors point out that both unanimous votes and 7-2 votes (the third row above) happened more often than you would expect mathematically. Only the 8-1 votes happened significantly less often than expected from the math. Thus the Supreme Court, despite the lack of job pressure or outside power held over it, acts very much like you might expect any small group to: "There is pressure to conform, and Justices apparently find it easier to be in the minority if there is at least one other Justice seeing things their way," Granberg and Bartels write. In other words, when there were two initial dissenters, if one gave into the majority, the other one was more likely to join them than stand alone.
Application: In this newsletter and elsewhere I have often warned against the dangers of "groupthink," where the mistaken pressure to "be a team player" by going along with the team causes people to do so even when they feel strongly that the team is wrong. The authors cite further evidence from a study of American juries in criminal cases, where all 12 people must agree on the decision of guilty or not guilty to prevent a repeat of the trial, and members thus receive tremendous pressure to conform from judges and fellow jury members. In that study, 95% of juries ended up in unanimous agreement although only 31% were unanimous on the first vote.
The problem is, sometimes the dissenter is right, at least right enough that a compromise from the majority would create a better decision. I have witnessed a single individual changing the team's position to one that all later agreed was the best solution. U.S. President John F. Kennedy blamed his efforts against dissent for the disastrous decision to back an invasion of Cuba by mercenaries in 1961, and changed to a model supporting disagreements during the (thankfully) successful solution of a 1962 confrontation with Russia over nuclear missiles in Cuba.
Fortunately, it is easy to make sure a dissenter is heard by the team, this study suggests: just provide the support of one other person. Anyone on the team can do that, and without even necessarily changing their own position. If one or two people on your team seem to feel especially strongly about a position the rest oppose, show support for their dissent by saying something like, "I want to understand your position better. Could you please explain it to me in a different way than you did before?" This makes the dissenter feel heard, improving their satisfaction with the team's process (see the second article in this issue) and forces the team to stop pressuring the dissenter long enough to listen to them. And by asking for a different approach, perhaps you will guide the dissenter to explaining themselves in a way that more people on the team understand, leading to a better decision.
Source: Granberg, D., and B. Bartels (05), "On Being a Lone Dissenter," Journal of Applied Social Psychology 35(9):1849.
All content in this newsletter, including the title, is: Copyright 2003-6 by Jim Morgan dba TeamTrainers Consulting. All rights reserved.
"SuddenTeams" is a registered trademark (US Trademark #2,456,849) and "TeamTrainers" and "The Science of Teams" are trademarks of Jim Morgan dba TeamTrainers Consulting.
TeamTrainers Consulting makes no guarantee or warranty regarding the use of information in this newsletter by individuals not employed by or under contract to TeamTrainers Consulting and performing official TeamTrainers business.