Abstract
Theorists posit two fundamental tensions in negotiation. One is strategic: the tug-and-pull between creating value and claiming one's share of it. The other is interpersonal: the tension between asserting one's own interests and, at the same time, empathizing with the feelings and needs of other parties.1 This research report analyzes how negotiators experience these tensions in practice. Specifically, their self-perceptions about their relative competence in several key areas allow us to see how strength along one dimension (like getting the maximum) is correlated with other important skills. Some of the authors’ findings confirm familiar models. For example, people implicitly feel that being successful at asserting their own interests imposes a cost with respect to understanding others. There were some surprises, however: most notably, people who rated themselves as strong value claimers also saw themselves as good value creators. The authors explore some of the implications of their findings for both practice and teaching, and foreshadow a follow-up report they plan. They also note how other researchers can access their data for their own studies.
People negotiate with others in the hope of achieving results superior to what they could obtain on their own. They need to negotiate, however, because what is best for them is less than ideal for other parties. This requires negotiators to ride two different horses at the same time: they must cooperate to reach a mutually acceptable outcome, but because of their differing interests, they must compete over exactly where they settle in the zone of possible agreement. David Lax and James Sebenius explain this tension:
No matter how much creative problem solving enlarges the pie, it must still be divided; value that has been created must be claimed. And, if the pie is not enlarged, there will be less to divide; there is more value to be claimed if one has helped create it first. (Lax and Sebenius 1986: 33)
Lax and Sebenius maintain that this tension is inevitable and cannot be erased by good intentions or deft process skills. Specifically, value creation depends on uncovering information about priorities and potential trade-offs, but unreciprocated disclosure can leave a negotiator vulnerable to exploitation. The pie will be expanded in such cases but others will grab the lion's share.2 Correspondingly, if the parties are coy about their interests, they may forfeit mutually beneficial trades and leave potential value on the table.
More recently Robert Mnookin, Scott Peppet, and Andrew Trumello have characterized a parallel interpersonal tension between empathy and assertiveness.
This is an experiential tension. Often negotiators feel as if they must either assert themselves or listen to the other side, but they can’t do both. Skilled negotiators know that both of these very different interpersonal skills are critical to effective negotiation. (Mnookin, Peppet, and Trumello 2000: 9–10)
This second tension is behavioral, not strategic, and emanates from the way people see themselves in relationship to others.
Once I understand your view — and show you I understand — holding on to my own perspective will become too difficult. . . . Conversely, if I try to assert myself in this negotiation, it's going to be tough to demonstrate an understanding of how you see things. Our views are just fundamentally different. If I advocate for mine, I can’t also advocate for yours. It's one or the other, not both. (Mnookin, Peppet, and Trumello 2000: 50–51)3
Even though the creating–claiming and empathy–assertiveness dichotomies are staples of negotiation analysis and fundamental to teaching, we have little data on how people recognize these theoretical tensions in practice. Specifically, do they really feel that they face hard choices about how they negotiate, or do they believe that they can excel at seemingly contradictory tasks? Understanding how people see these tensions (correctly or not) is critical for those of us who teach. If students come to us understanding that creating and claiming are mutually exclusive endeavors, that is one thing. If they think otherwise, we may have to do some groundwork to make our standard models credible.
This preliminary report analyzes the self-assessments of veteran negotiators. A survey asked them to identify their own relative strengths and weaknesses at the bargaining table, particularly in regard to cooperative and competitive moves. The instrument had an important twist: instead of simply asking respondents to rate their skills qualitatively, independent of one another, it required them to allocate a fixed number of points. Allotting more to one skill necessarily meant cutting the rating for at least one other category. The instrument thus compelled subjects to make comparisons. They could not casually say that they were outstanding in every regard (or inept, for that matter). Our intent also was to get people to distinguish how they ranked their specific skills from how they might judge their competence generally.
Some of what we discovered is consistent with negotiation theory. Our data indicate a perceived tension between empathy and assertiveness, for example, but there were puzzles and surprises as well. Rather than expressing a tension between creating and claiming value, respondents tended to see them as positively linked: people who felt that they were strong in expanding the pie also felt that they were particularly capable at securing a goodly slice of it. We offer some tentative explanations for this divergence between theory and practice and encourage other negotiation specialists to mine our results more deeply.4
The Self-Assessment Survey
The survey that generated the data for this study was originally designed for teaching purposes.5 It asks participants to rate themselves in regard to eight different negotiation skills and functions. Specifically, they had to allocate a total of eighty points across the following eight categories:6
Asserting your interests and point of view
Managing your emotions
Understanding the motivations and feelings of other parties
Recognizing and capitalizing on opportunities to create value
Establishing your trustworthiness
Dealing with people who have different negotiating styles than your own
Getting the maximum possible in the agreement
Managing the negotiation process
Embedded in the list are measures of both creating and claiming (items 4 and 7, respectively) and empathy and assertiveness (items 3 and 1).
In the classroom, a simple grid allows students to plot their own profiles and to consider what balance of skills would be desirable in various circumstances. In Figure One, for example, empathy and assertiveness are measured on the horizontal axis, and creating and claiming, on the vertical. Counting out from the origin in each direction, a student with identical scores on each of the four highlighted questions would describe a perfect square rotated forty-five degrees.
It would be splendid to excel in all eight regards, of course, but none of the participants in our sample described themselves as perfectly balanced with respect to all the skills and very few described themselves as balanced across the four that are highlighted. Everyone reported being stronger in some dimensions than in others. These self-reported relative strengths and weaknesses may express a sense of people's individual capabilities and temperament. It may also reflect the relative importance they attach to the different categories (i.e., people might want to believe that they are good at what matters most). Depending on the context, some might prefer it if their skills were skewed in one direction or another.
A separate part of the survey asked people to rate their overall negotiation ability relative to other negotiators they typically encounter. In our ongoing research, this question will allow us to explore whether more confident negotiators see themselves as having a different skill profile from those of others who are less sure of themselves. We will also be able to analyze whether perceived strengths and weaknesses vary with age and gender.
This particular instrument has the same general virtues and vices of most self-assessment surveys. Self-reflection is an inherently subjective process. People are not necessarily the best judges of their own strengths and weaknesses. Some may express their aspirations — the traits they would like to have — rather than describe their less admirable qualities. Others may be unduly harsh in ranking their own abilities. We administered the survey anonymously — the form has no place for people's names — in the hope of dampening possible worry that it might somehow be evaluative.7
We further acknowledge that self-assessment is especially tricky in the negotiation domain. First, it is a “wicked learning environment” in that it provides incomplete, even distorted, feedback when it comes to judging success.8 A basketball player practicing free throws sees immediately if she has missed or scored, and can make real-time adjustments in her technique. By contrast, a negotiator cannot ever be sure about how much value he has claimed unless he somehow discovers how much his counterparts would have conceded. Likewise, even if he knows in his own heart that he is trustworthy and empathetic, how well he communicates those traits to others is at least as important.
The interactive nature of negotiation makes self-assessment even more problematic. An individual's effectiveness is heavily dependent on the skill level and attitudes of his or her counterpart. Ginger Rogers was a better dancer with Fred Astaire than she ever could have been with Mickey Rooney. (Astaire would not have been so graceful with him either.) Context comes into play too. A buyer might be hard-nosed in dealing with a car dealer, but be a sweetheart selling the same vehicle to a kid in the neighborhood a few years later. Our instrument did not distinguish such contexts. Instead, it asked subjects to assess their skills generally, even though they might feel more competent in some circumstances than in others. In short, our survey simply records people's professed self-conceptions, not their actual behavior in specific cases.
Methodology
We wished to explore the attitudes and perceptions of experienced negotiators rather than those of the general population, so we chose participants in executive education workshops offered by the Program on Negotiation (PON).9 We infer from their enrollment in a specialized course that they regard negotiation as an important facet of their work — important enough to justify both the expense and the time away from the office.
Most participants in the PON course are from the United States, although a smattering of foreign countries are also represented. Most work in the private sector but others are employed in government, education, and other not-for-profit enterprises. Their range of substantive interests is broad. Some are engaged in labor relations and intraorganizational issues, while others deal with transactional negotiations, deal making, partnerships, and strategic alliances. Still others focus on settling litigation with customers, vendors, and regulators.
A survey of a more homogenized group (such as people who work for one organization or handle the same kinds of negotiations) might reveal different tendencies than those we found in this mixed population. It is impossible, of course, to constitute a truly “representative sample” of all types of negotiators because negotiation itself has vague boundaries.10 Nevertheless, the respondents here were speaking from direct experience, as varied as it certainly must be. We took care, moreover, to administer the survey at the beginning of each program before the presentation of any substantive material that might have influenced their answers. There may be a self-selection bias, of course, among people who chose to attend a PON course instead of some other training, as some may associate the Program with the “principled negotiation” approach championed by Roger Fisher, one of its several founders.11
Analysis of the data required sophisticated statistical tools. An extensive note on the methodology is available on line, as are the data themselves. Readers here will be spared most of the technical background. In a nutshell, the nature of the survey precluded conventional regression analysis. The device of allocating a fixed number of points across the set of skills meant that by definition, everything was correlated. Fortunately, specialists at Harvard Business School's Research Computing Services crafted ingenious statistical tools that we are using in our ongoing research.12
Overall Perceived Strengths and Weaknesses
In the aggregate, respondents were most confident about their ability to establish trustworthiness and to assert their interests. By contrast, they rated themselves notably lower with respect to dealing with different styles, getting the maximum, and managing the process. Table One summarizes the responses in descending order of confidence. An average rating near 10 suggests a skill that is not distinctly strong or weak. The right hand column, showing the standard deviation for each item, highlights a great range of results. Even for skills that most people felt were strengths, some other respondents rated them as personal weaknesses (and conversely).
Summary of the Responses in Descending Order of Confidence
| Skill (Item Number) | Mean Rating13 | Standard Deviation14 |
|---|---|---|
| Establishing trustworthiness (#5) | 12.94 | 4.11 |
| Asserting interests (#1) | 11.96 | 4.39 |
| Understanding others (#3) | 10.49 | 3.88 |
| Creating value (#4) | 9.97 | 3.25 |
| Managing emotions (#2) | 9.65 | 3.88 |
| Managing the process (#8) | 8.61 | 2.97 |
| Getting the maximum (#7) | 8.55 | 2.95 |
| Dealing with different styles (#6) | 7.82 | 2.69 |
| Skill (Item Number) | Mean Rating13 | Standard Deviation14 |
|---|---|---|
| Establishing trustworthiness (#5) | 12.94 | 4.11 |
| Asserting interests (#1) | 11.96 | 4.39 |
| Understanding others (#3) | 10.49 | 3.88 |
| Creating value (#4) | 9.97 | 3.25 |
| Managing emotions (#2) | 9.65 | 3.88 |
| Managing the process (#8) | 8.61 | 2.97 |
| Getting the maximum (#7) | 8.55 | 2.95 |
| Dealing with different styles (#6) | 7.82 | 2.69 |
That people regarded “establishing trustworthiness” as their greatest skill is consistent with research on self-serving biases generally. In particular, it is consistent with studies showing that people tend to see themselves as being more cooperative than those with whom they deal.15 Being trustworthy is a celebrated virtue, one that most of us would like to claim, although our good intentions may not always be obvious to others.
At first blush, the other clearly felt strength — asserting interests — seems to be of a very different nature. While trustworthiness appears central to cooperation, self-assertion is individualistic, perhaps even competitive. The common element in both skills, however, may be a belief in our ability to make ourselves clear to others. Because we know our own feelings and intentions, we assume that they are transparent to others (even when this is not the case). “Keenly aware of our own emotions, we often suffer the illusion of transparency. We presume that our feelings are leaking out when we’re opaque. Ditto for our social blunders and mental slips. . . . For better or worse, others just aren’t as focused on us as we are” (Myers 2002: 88). It is noteworthy here that respondents did not rate their own capacity to understand others as high (10.49) as they rated their ability to make their trustworthiness and interests clear to people across the table.
The gap between the transparency that negotiators seem to feel about their own mindset and their confidence in reading others is underscored by item 6, “dealing with different styles,” which respondents rated as their greatest weakness. People may be comfortable dealing with someone who seems to look and act like them, but baffled by what makes other kinds of people tick.
Respondents also felt relatively unsure about their ability at “getting the maximum” in negotiation. This may not be surprising given the uncertainty about the potential size of the pie to be divided, but it is an intriguing contrast to people's confidence about asserting their interests. Perhaps they recognize that asserting alone is insufficient to guarantee success. It may need to be coupled with other factors, such as one's persuasive ability, the capacity to deduce the other parties’ limits, or even one's willingness to walk away from mediocre deals. The low score on this item may also reflect a kind of negotiator's remorse, an internal second-guessing where the buyer ruminates that if only she had pressed a little harder, the seller might have dropped the price.
This reported lack of confidence seems to contradict the research by Richard Larrick and George Wu indicating that people typically under-estimate the zone of possible agreement and overestimate the portion that they capture (Larrick and Wu 2003). One explanation for the diverging results may lie in the fact that the Larrick and Wu studies are of specific negotiations. After reaching an agreement, subjects in their experiments were asked to estimate the other party's walk-away price. Not surprisingly, people tend to believe that they have pushed their counterparts pretty close to the limit (otherwise, why settle?). Our data suggest that rationalization in specific transactions may not translate to confidence about securing the maximum as a general matter.
The third area in which respondents in our survey reported relative weakness was in “managing the negotiation process.” This phrase was intended to represent the procedural side of negotiation, as distinct from its strategic (value creating and maximizing) or interpersonal (understanding and trustworthiness) aspects. People could have reasonably read different meanings into the wording, or treated it as a soft category from which they could deduct points in order to allocate them to more crisply defined skills about which they felt more confident.
Overall, respondents felt neither skilled nor clumsy in regard to the remaining three categories: “managing emotions,”“creating value,” and “understanding others.” As our database increases with successive rounds of the PON workshop, we will be able to say with greater statistical confidence that even these items are above or below the mean of 10. The sample size is already large enough to give us assurance that these skills will cluster in the middle and not, on average, be seen as distinct strengths or weaknesses. As with the other five factors, individual scores ranged widely. This variability may have been affected by how they construed the questions. For example, some people may have taken “managing emotions” to mean dealing with mercurial counterparts while others might have been thinking about their capacity to cope with their own anxiety and frustration. Lengthier definitions in the survey might have reduced such ambiguities somewhat, but at the likely cost of inadvertently loading the questions.
Even granting the unavoidable vagueness of language, the negotiators in our sample tended to see themselves as markedly more competent in some areas than in others. For practitioners, the averages can serve as a benchmark against which they can gauge their own self-conception. Likewise, those of us who teach can see where students think they need the most help — and where they may be falsely confident. For our research, however, we were most interested in whether people's perceptions about their effectiveness at creating and claiming, and empathy and assertiveness were consistent with negotiation theory — specifically, whether people recognized the classic tensions and trade-offs. We also wished to explore how effectiveness along these dimensions was related directly or inversely with any of the other specified skills.
Tensions and Synergies
We began our analysis by looking for possible correlations among the eight skills listed in the survey. Figure Two presents the results graphically. A heavily shaded cell indicates a strong correlation; a lightly shaded cell indicates a statistically significant correlation, although a somewhat weaker one.16 The plus or minus sign indicates whether the relationship was positive or negative. Cells are left blank where there is no statistical relationship between the skill listed in the row and the corresponding skill in the column.
For example, reading across row 1 we can see that asserting is negatively correlated with managing emotions (column 2), empathy (#3), and dealing with different styles (#6). The correlation is strong in the first case, less so with the latter two. The same row shows that asserting is positively correlated with establishing trustworthiness (#5) and getting the maximum (#7), the latter particularly so. The blank cells in columns 4 and 8 indicate that there was no significant statistical relationship between asserting and either creating value or managing the process.
Assertiveness and Empathy, and Related Abilities
We found a moderate negative correlation between empathy (#3) and assertiveness (#1) in the group as a whole. This is consistent with the empathy–assertiveness tension postulated by Mnookin, Peppet, and Trumello (2000). People who perceived themselves to be relatively effective in one of these respects tended to see themselves as less so in regard to the other.
Tracing out the relationships of empathy and assertiveness illuminates the nature of this tension while also raising several intriguing issues. Not surprisingly, empathy was strongly correlated with both establishing trustworthiness (#5) and dealing with different negotiation styles (#6). People who believed that they were effective at empathizing also reported being good at the other two skills, and vice versa. (The correlation also represents people who rated themselves low in these three categories, of course.) Thus, both establishing trustworthiness and dealing with different styles seem to be linked to empathy. However, no correlation was found between establishing trustworthiness and dealing with different styles. This suggests that while empathy is related to these two skills, they are not directly related to one another.
Curiously, there was no apparent correlation between empathy and creating value (#4). One might well assume that understanding others is both important in its own right and a means for discerning their substantive interests. Having that understanding would plausibly enable people to respond to other people's feelings appropriately, and further, to craft creative and sustainable deals. However, the perceived relationship between empathy and value creation was essentially random. It is important to remember, of course, that we are analyzing self-assessments not objectively measured behavior. Not everyone who says that they are empathetic may be all that sensitive. In turn, others who are truly empathetic may underestimate their skill at value creation.
One answer to this puzzle may lie in the strong negative correlation between empathy and maximizing value (#7). People who see themselves as empathetic apparently feel that they pay a substantive price for their skill. Just as important, value maximizers perceive that they make a relational trade-off. This inverse relationship seems like a stronger form of the empathy–assertiveness tension. As we will discuss in the following section, we found a positive correlation between getting the maximum (#7) and creating value (#4). Therefore, the negative correlation between empathy and value creation may be at least partially due to the negative correlation between empathy and value maximizing.
Looking across the asserting column (#1) reveals both expected and unanticipated results. Not surprisingly, asserting was strongly correlated with getting the maximum (#7). The empathy–assertiveness tension that Mnookin, Peppet, and Trumello (2000) identified is thus seen as not just relational, but also linked to outcomes. People who felt that they were effective in this regard believe that their skill is rewarded. Those who see themselves as unskilled believe they suffer as a result.
Asserting had a strong negative correlation with managing emotions (#2). Perhaps people who are forceful recognize the risk of provoking others. On the other side of the coin, people who keep their feelings under wraps may know that they do not always advocate successfully for themselves. Likewise, people who are not assertive may dread confrontations with those who are more forceful, which may partially explain the negative correlation between assertiveness and dealing with different styles (#6).
Worth comment is the moderately positive relationship between assertiveness (#1) and establishing trustworthiness (#5). This relationship is particularly interesting in view of assertiveness's negative correlation to empathy and managing emotions, both of which have strong positive correlations with establishing trustworthiness. This association with trustworthiness may reflect simple hubris on the part of assertive types who assume that they can ask you for the shirt off your back and get you to say thanks, while you are disrobing. Yet for this positive correlation to arise, there also have to be shrinking violets who sense that their forbearance does not necessarily win them trust. While there is a stronger association between establishing trustworthiness and both empathy and managing emotions, it seems that respondents in our sample understand that assertiveness need not be a vice. Straightforward statements about one's needs and options can build mutual confidence. Censoring one's thoughts and feelings sometimes can undermine trust.
Creating and Claiming Value, and Related Abilities
While the relationships between empathy and assertiveness generally conformed to familiar negotiation theory, the data on creating and claiming did not. Instead of the tension that Lax and Sebenius (1986) might predict, the responses revealed a strongly positive correlation between the two skills. Thus, many people saw themselves as having relative strengths in both categories — or if they felt that they were weak in one, they also saw themselves as weak in the other.
In classic television ads, a beer can taste great and be less filling too. In Saturday Night Live parodies, an aerosol foam can be both a floor wax and a dessert topping. By the logic of negotiation analysis, however, it is hard to have it both ways at the bargaining table. Cooperative and competitive moves are often antithetical, especially in regard to information disclosure.
There are several plausible explanations for why some people we surveyed nevertheless believe they can be good at both. First is the fact that the instrument asked people to assess their skills generally, not in a specific case. Some respondents might have felt confident they could create value or claim it, depending on what the circumstances called for. Second, value maximization can be achieved by expanding the pie. Respondents who felt that confident (or not) about their effectiveness in regard to value creation might have reasonably associated their degree of talent with output. Finally, asking about success at creating and claiming may elicit the same general sense of competence. People who feel good (or bad) about their basic skills may give themselves the same grade in each category.
All of that said, it still is intriguing that the same respondents who expressed a tension between empathy and assertiveness did not report a parallel tension with creating and claiming value. It is noteworthy that Figure Two shows “getting the maximum” (#7) is strongly correlated with all the other factors but one (#6), dealing with different styles. Perhaps not surprisingly, it is positively associated with asserting (#1). As the saying goes, “you don’t get if you don’t ask.” The statistical corollary is that if you do ask, you do get. Getting the maximum is also positively linked to “managing the process” (#8). This seems credible, too. People who assert themselves explicitly influence the direction of the conversation. Those who feel shy in this regard may feel that they follow instead of lead.
The strong negative correlations with getting the maximum are revealing as well. Managing emotions (#2), empathy (#3), and establishing trustworthiness (#5) all are seen to suffer if negotiators push to the limit, and conversely. Respondents thus see a powerful link between substantive results and important relational issues.
The correlations (or lack of them) between creating value (#4) and the other factors are more perplexing. Figure Two shows no link, one way or the other, between value creation and five other items, specifically, asserting (#1), managing emotions (#2), empathy (#3), dealing with different styles (#6), and managing the process (#8). The latter three in particular would seem to be features of a value creating process.
Most surprising is the strongly negative relationship between value creation and establishing trustworthiness. We can only speculate about what may be going on here. One possibility is that some people feel that making it clear that they can be trusted leaves them vulnerable to exploitation. Others will expand the pie and grab it, too. It is harder, however, to construct a rationale for the inverse, that is, the notion that appearing other than trustworthy actually promotes value creation. We should acknowledge the possibility that some individuals hold inconsistent beliefs. It is odd nonetheless that a statistically significant portion of our sample apparently does not connect value creation with the positive relationship building steps advocated in many popular negotiation texts.
Conclusion
We undertook this project to see if the classic tensions in negotiation are perceived in practice. The answer that we found is, yes and no. As Figure Three illustrates, people on average feel that if they move toward empathy they must forsake something in the way assertiveness. As noted earlier, empathy was negatively correlated with maximizing gain, so the figure shows that something is lost in this dimension as well. The connections among these perceptions are not necessarily reality, of course. Mnookin, Peppet, and Trumello (2000) argue convincingly that it is possible to excel at both asserting and empathizing. The web of connections thus may be more pliable than most negotiators believe.
Figure Four describes a different sort of shape, one where excellence in the direction of creating value is positively associated with excellence in seemingly the opposite direction of value creation. Here we see a confident negotiator, but we can also imagine a flattened diamond, wider than it is high, for someone else who feels unskilled at both creating and claiming value.
In our ongoing analysis of the data, we plan to look for clusters of skills and traits and see if they correspond with or contradict familiar negotiation “styles” or types. We will also look at how peoples’ responses vary with age and gender, as well as basic self-confidence. In addition, we will give special attention to respondents whose skill sets were especially lopsided, that is, very strong or very weak in a particular area. We are curious about whether the general profiles of such people are similar to the sample at large, or whether they perceive different trade-offs.
Finally, we will continue to build our data set as more participants go though the PON workshops and as we gather data from other groups as well. We encourage colleagues to use the survey in their own classes and to contribute their data to the pool. We are especially interested in contrasting responses of sets of negotiators from a range of fields, organizations, and cultures.
Notes
There are other tensions in negotiation, of course, notably the potential conflict between the interests of agents (such as lawyers and brokers) and the principals that they represent (see Mnookin, Peppet, and Trumello 2000, Chapter 3).
Lax and Sebenius describe this as “the negotiator's dilemma,” borrowing the term from game theory and suggesting that strong strategic pressures often stifle value creation (Lax and Sebenius 1986: 38–41).
Extending the argument, the authors note, “[w]hen faced with conflict, we tend to either advocate forcefully — often too forcefully — our own view or focus on the other side's view, rather than moving nimbly from one skill to the other. We each assert our own story and listen to the other side only for the purpose of constructing a ‘Yes, but’ response. We cycle through argument, and counterargument, never demonstrating understanding or really communicating very effectively” (Mnookin, Peppet, and Trumello 2000: 50).
The data discussed in this report, as well as a detailed description of our statistical analyses, can be found at http://www.pon.harvard.edu/tensions
The instrument, along with an interpretative grid and teaching note, can be ordered as Case #902218 from Harvard Business School Publishing via http://www.harvardbusinessonline.hbsp.edu
This allocation device was chosen for several reasons. First, pedagogically it stimulates students to think about their learning priorities. Second, for research purposes, it tempers the problem of differing subjective standards of performance (as where one person might be boastful about all his skills and another, unduly modest). Third, it compels respondents to make trade-offs, which then can be analyzed in the aggregate for relationships and patterns.
We acknowledge that the sequencing of the questions may influence how people allocated their choices, although it is practically impossible to administer the survey in all its possible permutations. Moreover, there is no easy way to track the processes by which people distributed their points. Those who look at the assignment globally might budget their points differently from those who hop from one item to the next. Likewise, people who start by allocating “bonus” points to their relative strengths might demonstrate a different spread from others who began by deducting points in areas of perceived weakness. These are legitimate design issues that could be raised in connection with most self-assessment surveys.
For a comparison of “kind” and “wicked” learning environments, see Hogarth 2001: 89–90.
N = 239. Several hundred surveys were collected at PON workshops in 2002 and 2003. More than 80 percent of participants submitted forms. Approximately 5 percent of those were not included in the analysis, as the respondents made mathematical errors in allocating their points.
It is fair to assume that the people who chose to attend the workshop understood that negotiation spans more than just hammering out contract terms or settling lawsuits, but they might differ about whether the concept extends to leading teams, implementing business strategy, and management generally.
We are very grateful to William Simpson, Angela Tong, and Deborah Strumsky for their assistance, and to the Division of Research at the Harvard Business School for its support of this project.
Mean ratings that are displayed in bold were found to be significantly different from 10 (p< 0.05), and thus reflect the group's average perceived strengths and weaknesses. Mean ratings that are not displayed in bold were not found to be significantly different from 10. The mean rating for item 3 — understanding others — is marginally significant (p= 0.055).
The standard deviation for each item indicates how widely dispersed the responses to that item were. Therefore, as we see from these numbers, there was a greater variability of responses for items that were found to be relative strengths than for those found to be relative weaknesses although this is probably a function of the mean being closer to the lower bound (0) than to the upper (80).
For a discussion of positive illusions and their effect on decision making, see Bazerman 1998: 94–99.
“Strong correlation” corresponds to a one-tailed p-value of 0.05; “moderate correlation” corresponds to a one-tailed p-value of 0.025. By contrast, a “weak correlation” would correspond to a one-tailed p-value of 0.01, although none are reported here.
