How face threat sensitivity affects proactive negotiation behavior

Abstract

Purpose

Face threat sensitivity (FTS) has been found to influence objective negotiated outcomes when the threat to face is activated. The purpose of this study is to extend that research by testing whether FTS – which is defined as a propensity to act – is associated with the outcomes of negotiators when the threat has not been specifically activated. Face theory specifies that face threats can cause individuals to take proactive steps to avoid threats before they might occur.

Design/methodology/approach

Drawing on face theory and social role theory, the authors conduct a negotiation experiment and use hierarchical regression to test hypotheses concerning the relationship between FTS for sellers and buyers on negotiated outcomes in both distributive and integrative negotiations. The authors also use moderated regression to test if gender moderates the relationship between buyer and seller FTS and negotiation outcomes.

Findings

Results show that, when the threat is not activated, high FTS buyers pay more than low FTS buyers. Consistent with face theory and social role theory, this effect is moderated by gender, with the association being stronger for women buyers than for men buyers.

Originality/value

This paper exhibits that FTS can influence negotiator behavior even when FTS is not activated. This is valuable to negotiation scholars and practitioners who are interested in the role that individual characteristics play in negotiation behavior.

Citation

Miles, E.W., Schatten, J. and Chapman, E. (2020), "How face threat sensitivity affects proactive negotiation behavior", Organization Management Journal , Vol. 17 No. 1, pp. 2-14. https://doi.org/10.1108/OMJ-05-2019-0725

Introduction

White, Tynan, Galinsky, and Thompson (2004) introduced the concept of face threat sensitivity (FTS) to the negotiation literature. In a series of studies, they found that when FTS is activated, negotiators are more likely to reach an impasse and are more likely to reach agreements that do not optimize created value. White et al. focused on the negotiation effects of activated FTS – when individuals actually perceive that their face has been threatened in a negotiation. However, these studies analyze one side of the coin – when FTS is actually activated. White et al. describes that when FTS is activated it can produce a negative effect or increased competitiveness, which can result in an impasse in the negotiation, even when an agreement is more desirable. In contrast, from the perspective of the negotiation counterpart, a negotiation can breakdown because it is perceived as being not worth the effort to work with someone who is a difficult counterpart.

FTS is an individual difference that is stable over time (Tynan, 2005). Negotiating is an activity that often does bring together a number of elements that threaten face (Brown, 1968; Mohanty & Mukherjee, 2018; Small, Gelfand, Babcock, & Gettman, 2007; White et al., 2004; Wilson, 1992). As negotiations cover a vast range of potential issues (e.g. real estate, economic sanctions and a fence between neighbors), there will naturally be negotiations for which FTS remains not activated even though individuals vary in their sensitivity to face threat. Without the actual threat being activated by actions of the negotiation counterpart, will individual differences in FTS be associated with differences in negotiated objective outcomes?

Theoretical development

Face theory

Goffman (1967, p. 5) defines face as “the positive social value a person effectively claims for himself [or herself] by the line others assume he [or she] has taken during a particular contact.” As such, the face is not only solely a characteristic of the individual but also a characteristic of the individual’s interaction with others. Because it is not solely controlled within the domain of the individual, it is possible for the individual to “lose face.” Goffman (1967, p. 10) observes that a person’s face is “on loan to him [or her] from society; it will be withdrawn unless he [or she] conducts himself [or herself] in a way that is worthy of it.”

Because the face is a highly prized possession, individuals will go to great lengths to maintain face (Goffman, 1967; Petriglieri, 2011). They will avoid behaviors that are inconsistent with the face they have presented over time. They will also choose to engage in behaviors that uphold the face that they have established, even if such behaviors are not preferred or come at a personal cost, such as paying a high price for a brand that they do not value but view the purchase as integral for face savings efforts (Siu & Kwan, 2016) or by paying directly to preserve their own self-image (Eriksson, Mao, & Villeval, 2017). In any particular social interaction, individuals must be concerned not only for upholding their face in that particular interaction but also for the implications to their face in the broader social context beyond.

Goffman (1967, p. 13) goes as far as to say that “almost all acts involving others are modified […] by considerations of face.” A number of authors (Brown & Levinson, 1987; Carson & Cupach, 2000; Oetzel, Garcia, & Ting-Toomey, 2008; Leavitt & Sluss, 2015) have discussed typologies of these particular modifications (“face work”) that individuals use to maintain face. These writers note that the most obvious strategy – avoidance of the face-threatening act – is the most effective at saving face.

A key conclusion is that a concern for face entices people to engage in proactive behaviors to avoid situations that would cause their faces to be threatened. “In many societies, members know the value of voluntarily making a gracious withdrawal before an anticipated threat to face has had a chance to occur” (Goffman, 1967, p. 15). The strategy of avoidance is a proactive strategy.

Face and negotiating

Negotiation is an arena that can provide significant threats to face (Brown, 1968; White et al., 2004; Wilson, 1992). Metts (1997) observes that, in part, maintaining face involves others valuing a person’s possessions and opinions. Each of these is often challenged in negotiation. When we are sellers, we assert that our possessions have a certain worth; it is the role of the buyer to challenge our claim. The opinions we use in logical support for our negotiation position are often undermined by the negotiation counterpart who stands to gain advantage from the use of potential face-threating negotiation techniques.

It should be noted that many of the acts we call face threat are common competitive practices, and part of the “game” for some negotiators. Yet they are intrinsically threatening to face (White et al., 2004, p. 104).

Face threat sensitivity and negotiation

FTS has been proposed by Tynan (2005) as an individual difference. Tynan defines FTS as the degree to which an individual is likely to have a negative affective reaction to threats to the face the individual is attempting to project and maintain. By this definition, FTS is a propensity that varies among people.

White et al. (2004) introduced the construct of FTS to the negotiation literature. They conducted a set of studies, which found that, when high FTS individuals feel threatened in a negotiation context, the threat causes them to act in a defensive manner. This defensiveness increases the incidence of impasse and decreases the ability to cooperate to optimize joint gain when the integrative potential exists. White et al. hypothesized and found an effect by the negotiator role. They found that high FTS sellers negotiated agreements that resulted in lower joint gain than low FTS sellers while the level of buyers’ FTS was not significantly linked to joint gain. White et al. explained this dichotomy as being a result of the seller being more invested in the goods/services being negotiated. In one study, the negotiation involved selling a business that the seller had built and operated; the property was an extension of self for the seller but was simply a potential business transaction for the buyer.

In a multi-study investigation, White et al. (2004) chose negotiation scenarios intended to invoke threat to face for the negotiators. One was a scenario, which frequently ends in an impasse and indeed ended in an impasse for 41 per cent of their negotiating dyads. In another of their studies, results indicated that high FTS sellers were rated as more competitive than low FTS sellers, consistent with the hypothesis that they were acting defensively in the presence of a perceived threat.

Proactive behavior caused by face threat sensitivity

The work of White et al. (2004) shows that, when FTS is activated, it influences negotiator behavior and negotiator outcomes. In addition to these dimensions of FTS effects, we posit that high FTS also has an effect on people who have not experienced threats in a specific negotiation but are sensitive to the possibility of a threat to face. According to Goffman (1967), people tend to migrate to situations where their faces can be upheld and tend to avoid situations where the potential exists for their faces to be threatened. This avoidance strategy indicates that – even absent a specific threat – people who are concerned about potential threats to face will take proactive measures to prevent the possibility of that threat emerging. By definition, individuals high in FTS are more prone (more “sensitive”) to be concerned about threats to face; it is a propensity to perceive a threat. Therefore, we hold that the level of FTS will influence the degree to which individuals will engage in proactive measures to avoid a potential threat.

Miles (2010) has proposed that individuals high in FTS will be more likely to avoid negotiations altogether. While this pattern seems quite plausible, we also hold that, once individuals have entered a negotiation, they will engage in behaviors within that interaction to avoid the possibility of the threat emerging. Goffman (1967) notes that such behaviors include hedging in making statements; using belittling modesty; and making strong qualifications in expressing statements, opinions or assertions. In general, these individuals are less likely to “stick their neck out.” The underlying logic is that the individual does not want to present an unequivocal face from which retreating would be problematic. Unfortunately, such equivocality can be strategically sub-optimal in negotiating. For example, individuals engaging in a pattern of taking an equivocal face would be less ambitious in stating their desires and the strength of those desires. Because of this pattern of behaviors that are inconsistent with an optimal negotiation strategy, we suggest the following hypotheses:

H1a.

There will be a negative relationship between FTS for sellers and negotiated outcomes in a distributive negotiation.

H1b.

There will be a negative relationship between FTS for buyers and negotiated outcomes in a distributive negotiation.

Asymmetrical role effects

White et al. (2004) predicted asymmetrical role effects in situations where the threat to face was actually activated. Because sellers are more invested in their position (e.g. their goods or services), they will respond more defensively to a threat to face. Based on that same logic, we anticipate an asymmetrical effect – but not the same effect – when the threat is not activated. Under conditions of no activated threat, we believe that high FTS buyers will be more proactive than high FTS sellers in engaging in avoidance choices and, therefore, will receive fewer outcomes in distributive negotiation than other buyers. Because the seller is more invested, the buyer will anticipate the greater likelihood of provoking a defensive response from the seller. In a buying/selling context, the seller is typically providing goods or services of an indeterminant (i.e. less objective) value in return for the payment of a determinant value (usually monetary). Conversely, the buyer is providing payment of a determinant value in return for goods or services of an indeterminant value (Neale, Huber, & Northcraft, 1987; Kristensen & Gärling, 1997). In this context, there is more necessity for buyers to “stick their neck out” and declare what determinant value they wish to exchange for the goods or services of indeterminant value they would receive in return. Because the seller is more invested, a declaration from the buyer of the determinant value is more likely to draw a defensive response from the seller than a declaration of determinant value by the seller is to elicit a defensive response from the buyer.

In summary, we anticipate that FTS will be associated with objective negotiated outcomes. However, we anticipate an asymmetrical pattern that the effect will be particularly pronounced in buyers:

H2.

The negative relationship between FTS and negotiated outcomes in a distributive negotiation will be stronger for buyers than sellers.

Gender as a moderator

In addition to the asymmetry by role, we also predict that the relationship between FTS and negotiated outcomes will be moderated by gender. This prediction is rooted in social role theory (Eagly, 1987). In part, social roles provide a prescription for how individuals are expected to behave. Any person who violates expected social roles incurs negative consequences for not following the prescription (Carli, 1990; Cialdini & Trost, 1998; Rudman & Glick, 1999).

Specifically, with regard to gender, there is a social role expectation that men will act in a pattern that is assertive, rational and competitive. Women will act in a pattern that is passive, relationship-oriented and cooperative. Unfortunately, the pattern expected of men is more consistent with successful distributive negotiation while the pattern expected of women is less consistent with successful distributive negotiating (Bowles, Babcock, & Lai, 2007; Kray & Thompson, 2005; Walters, Stuhlmacher, & Meyer, 1998; Watson, 1994; Kugler, Reif, Kaschner, & Brodbeck, 2018). Therefore, violating gender role expectations to negotiate effectively is more likely to be a concern for women, but not for men. As noted by Goffman (1967), in a particular social interaction (e.g. a specific negotiation), individuals must be concerned not only with upholding face in that immediate interaction but also for the implications to their face in the broader social context beyond. Mazei et al. (2015) in a meta-analysis on gender differences in negotiation outcomes and their moderators call for future research into finding new moderators. We answer this call. We posit that violating gender roles to negotiate effectively should elicit concern for the possibility of losing face in the broader social context:

H3a.

Gender moderates the negative relationship between seller FTS and negotiation outcomes, such that the relationship will be stronger for women than men.

H3ab.

Gender moderates the negative relationship between buyer FTS and negotiation outcomes, such that the relationship will be stronger for women than men.

Materials and methods

Participants

Study participants were 130 undergraduate students (65 negotiating dyads) at a large state university in the USA. Students were taking principles of management course and received course credit for participating in the study. Two dyads were deleted because of missing data, leaving a sample of 126 participants (63 dyads). Of this final set of respondents, 55 per cent were female; the average age was 24.3, with a standard deviation of 5.8. The age range of participants was 19-47.

Procedure

Students completed a questionnaire prior to negotiating. This questionnaire, a seven-page document, contained several focal measures for the study. After completing the preliminary questionnaire, students negotiated the “Rock and Roll Band Movie” case [1]. Students were randomly assigned to a negotiation role – buyer (movie producer) or seller (band manager) – and randomly assigned to a negotiation counterpart. This negotiation is a one-issue distributive situation based loosely on the actual situation of the Beatles’s manager negotiating the terms of the first Beatles movie. A movie producer is interested in making a movie about a day in the life of a 1960s rock band. As stated in the case, all terms (e.g. dates of filming, location of filming and choice of the director) of the agreement have been settled in previous discussions except for the percentage of the movie profits that will go to the band. The two negotiators (band manager and movie producer) are given the task of negotiating that percentage. The case instructs the movie producers that they may not agree to more than 25 per cent and instructs the band managers that they cannot agree to less than 10 per cent. This information sets up a one-issue distributive negotiation with a bargaining zone (i.e. the range of possible agreements that are acceptable to both negotiators) of 10-25 per cent. Upon arriving at an agreement, students completed an agreement form, which recorded additional study variables.

Measures

Seller and buyer.

In this case, the band manager meets the criteria of the seller and the movie producer meets the criteria of the buyer (Neale et al., 1987; Kristensen & Gärling, 1997). The band manager is providing services of indeterminant value while the movie producer is providing monetary payment of determinant value. Therefore, the band manager will be referred to as the seller and the movie producer will be referred to as the buyer.

Face threat sensitivity.

One instrument in the questionnaire was the FTS measure. Tynan and colleagues have used two different measures of FTS in published research – a three-item measure (White et al., 2004) and a five-item measure (Tynan, 2005). The same format was used in both measures – each item was a nine-point semantic differential format with “1” anchored as “not at all characteristic of me” and “9” anchored as “very characteristic of me.” One item (“I don’t respond well to direct criticism”) was common to both measures. For our measure of FTS, we combined all items from both measures; however, we deleted an item (“I am pretty thin-skinned”) that was used in the White et al. (2004) measure. In the Tynan (2005, p. 234) measure, this item was deleted because, in the validation process, “several participants were unsure of what the [item] meant and asked for clarification.” This combination resulted in a six-item scale. Four of the items were reversed so that higher scores represented greater FTS. Items were averaged to obtain the FTS score. The six-item scale was sufficiently reliable in this sample with a coefficient alpha of 0.80.

Agreement (percentage).

Upon arriving at an agreement, the two negotiators completed an agreement form, which recorded the agreed upon percentage of movie profits, between 10 and 25 per cent.

First offer.

The agreement form also asked the dyad members to record, which individual made the first offer (i.e. offered a specific number before the other party mentioned a specific number). This variable was included as a control variable; evidence (Galinsky & Mussweiler, 2001) indicates that, under some circumstances, making the first offer can influence the end agreement, and we wanted to control that potential effect.

Satisfaction with process and satisfaction with outcomes.

Also, after arriving at an agreement – and before any debrief information was provided – participants individually responded to two items intended to assess whether FTS had been activated in the process of negotiating. White et al. (2004) applied three criteria to assess whether FTS had been activated. One was whether negotiators reached an impasse rather than an agreement. The other two were satisfaction with the negotiation process and satisfaction with negotiation outcomes. The two questions used in this study were intended to replicate the same satisfaction assessment. Both questions were in the format of a nine-point semantic differential. The first item assessed satisfaction with the negotiation process. It was phrased “how satisfied are you with the negotiation process that resulted in the agreement?” The second item assessed satisfaction with the end agreement: “how satisfied are you with the percentage (the end agreement) you received?” Each measure was scored so that higher scores indicated higher satisfaction.

Results

Descriptive statistics and variable inter-correlations are provided in Table 1. A one-sample t-test indicated no significant difference between the mean FTS for the two roles (t(62) = 0.80, ns).

In Table 1, the correlation between satisfaction with the negotiation process and FTS was not statistically significant for either buyer or seller. Additionally, the correlation between satisfaction with outcomes and FTS was not statistically significant for either buyer or seller. Because neither measure of satisfaction was associated with FTS nor these results suggest that FTS was not activated in this negotiation. Also, supporting that conclusion is the result that no dyads reached an impasse. White et al. (2004) intentionally attempted to activate FTS and selected a negotiation exercise that resulted in a 41 per cent impasse rate in one study. The fact that we have an impasse rate of 0 per cent is consistent with FTS not being activated. Consistent with the criteria of White et al. (2004), the totality of this evidence suggests that FTS was not activated in this negotiation situation.

The ability of FTS to predict the negotiated agreement was tested using hierarchical regression. The results appear in Table 2. The regression weights in Table 2 are standardized. In Model 1, the control variable (party making the first offer) and the four predictor variables were entered. The overall model was statistically significant, and the R² value was 0.18. As the data from Model 1 did not support the relationship between seller FTS and negotiation outcomes, H1a was not supported. The data from Model 1 did support the relationship between buyer FTS and negotiation outcomes, so H1b is supported. The positive and significant association between FTS and negotiated outcomes for buyers, but not for sellers, supports H2.

In Model 2, the interaction of seller gender and seller FTS was entered. The interaction was not significant and the increment in the R² value was not significant. Therefore, H3a of gender as a moderator of the relationship between FTS and distributive agreement is not supported for the seller role.

In Model 3, the interaction of buyer gender and buyer FTS was added to the variables in Model 1. In Table 2, the interaction effect is statistically significant and the increment in the R² value is also significant. To better understand this interaction, we graphed the interaction and tested regression models separately by the buyer gender.

Figure 1 provides the graph of the interaction. To graph this effect, median splits were used to create a high FTS group and a low FTS group for each gender [2]. The visualization in Figure 1 suggests that the significant interaction was caused by the group of higher FTS women buyers. To provide a statistical test for this suggested effect, Table 2 also provides regression models by buyer gender. Model 4 provides the regression results for women buyers; Model 5 provides results for men buyers. As anticipated from the pattern in Figure 1, the FTS of men buyers is not significantly related to distributive agreements. However, Model 4 indicates a very clear relationship. Model 4 has a large R² value of 0.30, and the overall model is statistically significant. One predictor variable, buyer FTS, is statistically significant. The positive regression weight is consistent with Figure 1 in showing that, for women buyers, greater FTS is associated with paying higher prices.

Discussion

The established literature on FTS and negotiations has centered on the effects of activated FTS. For example, White et al. (2004) suggest that when FTS is activated during a negotiation, counterparts are less likely to reach an agreement and that the negotiated resolution is less pareto efficient. In this paper, we take a different approach by analyzing the effects of FTS when FTS is not activated. As described by Goffman (1967), those who are high in FTS will draw on avoidance strategies, such that even without experiencing FTS activation, such individuals will take preemptive measures to mitigate the risk of potential threats. In this study, we hypothesized that FTS – as a propensity – is associated with negotiator outcomes even when the FTS is not activated in a particular negotiation situation. We also hypothesized that this effect would be asymmetrical by role – more prominent for buyers than sellers – and would be moderated by negotiator gender. Four conclusions seem warranted.

First, the sum of the evidence supports our hypothesis that FTS is associated with negotiation outcomes even in contexts where the FTS is not activated. This finding is a noteworthy and contrasting extension of the White et al. (2004) results showing that activated FTS influences negotiators. It is also consistent with face theory’s basic tenet (Goffman, 1967) that individuals concerned about avoiding threats to face will engage is an equivocal pattern of behavior; this pattern is inconsistent with optimal negotiation strategy. Thus, we showcase that FTS is a far more powerful and relevant force in negotiations than was previously established – even when it is not activated, it still plays a role in negotiations. We believe that this has significant implications for negotiators. Namely, that those who are higher in FTS might take proactive steps to avoid a threat to face, as they are more likely to be concerned that particular actions of a counterpart are threatening to face. Negotiators higher in FTS should be aware that these preemptive measures to save face might have a negative impact on their negotiation objectives.

Second, we hypothesized and found that the association between FTS and negotiation outcomes is stronger for buyers than for sellers. We had not anticipated that the association would be non-significant for sellers but that finding is consistent with our asymmetry hypothesis. Buyers are suggesting a determinant value (i.e. price) for the goods or services of the seller, and the value of these goods/services is less objective than the suggested price. Therefore, buyers should have greater concern about “sticking their neck out” and suggesting the appropriate determinant value. Findings are consistent with this logic. Of note, we found a significant correlation between buyer FTS and seller satisfaction with outcomes (r = 0.26). This seems logical that buyer FTS results in a higher price paid by the buyer so that the seller would be more satisfied with the outcome.

Third, according to the social role theory (Eagly, 1987), there are negative consequences for individuals who step out of their “ascribed” social roles. In contrast to women, men tend to have the social expectation of engaging in a manner that is dominant or aggressive, attributes that the negotiation literature has established as being optimal for distributive negotiations (Bowles et al., 2007; Kugler et al., 2018). Thus, we suggested that those higher in FTS would be less likely to depart from gender roles and that for women, this reluctance would be suboptimal for negotiation outcomes. We hypothesized and found that the association between FTS and negotiated outcomes is moderated by gender; however, this effect was only present for buyers. Women buyers with higher FTS paid more than women buyers with lower FTS. The degree of FTS was not a significant predictor of negotiated outcomes for men. Taken together, these findings support the thesis that women who are higher in FTS will be less likely to act in a way that runs counter to social roles, even though it would be beneficial in a distributive negotiation. This is a significant finding that showcases the relevance of FTS more broadly. A recent study by Pew Research Center (2018) found that while the gender pay gap has improved, women in the USA are still only paid 82 per cent of the salary of men. Some of this discrepancy has been connected to the findings that in contract negotiations women make lower salary requests than men do (Save-Soderbergh, 2019). Our results suggest that one possible explanation for why women make lower salary requests is related to gender differences in non-activated FTS.

Fourth, the results in Figure 1 indicate that women buyers lower in FTS did not pay more than men with lower FTS. This finding suggests that absent higher FTS, women will not differ from men in the amount of negotiated outcomes. We believe that the research question of whether men and women differ in the negotiation outcomes they obtain (Bowles, Babcock, & McGinn, 2005; Bowles et al., 2007; Kray & Thompson, 2005; Miles & LaSalle, 2008; Stuhlmacher & Walters, 1999) would benefit from a closer consideration of FTS and of avoidance behaviors in general.

Limitations and directions for future research

This study tested a single population with a single negotiation situation. Therefore, the ability to generalize broadly is limited. A number of situational factors can influence the degree to which gender differences in negotiated outcomes occur. These include whether the negotiators are negotiating for themselves or representing others (Bowles et al., 2005) and whether the negotiation context is gender-stereotyped (Miles & LaSalle, 2008). Therefore, it may be that under a different set of situational factors, the effects found in the current study could be either more evident or less evident.

As stated previously, a number of writers (Miles, 2010; Small et al., 2007) have researched the issue of the decision of whether to initiate a negotiation. Additionally, Volkema (2009) has written about the concern of sub-optimal initiation in the negotiation process. We believe that tying FTS to these discussions could provide a beneficial link. A basic tenet of face theory is that avoidance of threats to face is a strong motivator of behavior choices. Also, Goffman (1967) holds that almost all human interactions are modified by considerations of maintaining face for the initiator of the interaction, as well as other participants in the interactions (i.e. negotiators and counterparts in negotiation contexts). As the scope of negotiation research broadens to include the decision of whether to initiate a negotiation, FTS and related concerns of face theory should be a relevant topic in that discussion.

Conclusion

FTS has been proposed as a propensity that may or may not be activated. Previous research has shown that activated FTS influences negotiated outcomes. In this study, we contribute to the negotiation literature by also showing that FTS, which is not activated is associated with negotiation outcomes. This relationship was found for buyers but not for sellers. Also, consistent with gender role theory, this relationship is moderated by gender. Thus, we also shed light on gender differences in negotiation, showing that female buyers who are higher in FTS tend to pay more in negotiations. Furthermore, given the current interest in the pay gap between women and men, our findings could provide one explanation of this gap – women who have a higher FTS accept lower pay (making a greater concession) in an effort to “fit” with their stereotyped gender role of cooperativeness. They do not wish to upset the user by asking for more and avoid the threat altogether, often settling for sub-optimal outcomes. In conclusion, this study provides practical implications for buyers, sellers, women and men through the lens of potential threat to face in negotiations.

Figures

Figure 1.

Interaction of buyer gender and buyer FTS

Table 1.

Descriptive statistics for study variables

Variable	Mean	STD	1	2	3	4	5	6	7	8	9
1. Seller satisfaction with process	7.50	1.71
2. Buyer satisfaction with process	7.53	1.62	0.03
3. Seller satisfaction with outcomes	7.79	1.46	0.73**	−0.10
4. Buyer satisfaction	7.87	1.74	0.08	0.79**	0.03
With outcomes
5. Seller gender	0.42	0.50	0.00	0.04	−0.08	0.02
6. Buyer gender	0.50	0.50	0.16	0.03	0.06	0.06	0.13
7. Seller FTS	4.02	1.58	0.17	−0.06	0.05	−0.13	−0.18	0.16
8. Buyer FTS	3.80	1.32	0.16	−0.08	0.26*	−0.08	0.02	−0.34**	−0.20
9. Agreement (percentage)	17.16	3.27	0.11	−0.34**	0.16	−0.45**	0.04	−0.17	−0.20	0.34**
10. First Offer	1.55	0.50	−0.05	−0.04	−0.01	−0.02	0.00	−0.04	−0.09	−0.21	0.10

Notes:

N = 63;

*p < 0.05;

**p < 0.01; gender is coded “1” = male and “0” = female; and first off is coded as “1” = seller and “2” = buyer

Table 2.

Study 1: hierarchical regression models: ability of FTS to predict distributive agreement

Variable	Model 1	Model 2	Model 3	Model 4 Women Buyers (N = 32)	Model 5 Men Buyers (N = 31)
Control variable
Party making	0.21	0.21	0.26*	0.30	0.21
First offer
Predictor variables
Seller gender	0.01	0.12	0.00	0.02	−0.03
Buyer gender	−0.03	−0.03	0.70
Seller FTS	−0.09	−0.05	−0.07	0.05	−0.24
Buyer FTS	0.38**	0.38**	0.62**	0.63**	0.10
Interactions
Seller gender x	−0.12
Seller FTS
Buyer gender x			−0.71*
Buyer FTS
F	2.59*	2.14***	2.92**	2.94*	1.01
R2	0.18	0.19	0.24	0.30	0.13
aΔR2		0.01	0.06*
Adjusted R2	0.11	0.10	0.16	0.20	0.00

Notes:

N (number of dyads) = 63;

***p < 0.10;

*p < 0.05;

**p < 0.01; and

^aincrement to R² added beyond Model 1

Notes

1 Available upon request.

2 The cell sizes created by these median splits (cells of 15-17) suggest that the cells are too small to use for individual tests of differences (e.g., t-tests and ANOVA). However, because the hierarchical regression had found a significant interaction effect, the graphing was helpful, not as a statistical test, but to illustrate the form of the statistically significant interaction.