Prevodb
Jump to navigation
Jump to search
Culture, gender, organizational role, and styles of conflict resolution:A meta-analysis$ Jennifer L. Holta,�, Cynthia James DeVoreb aHolt Enterprises Consulting Services, Minneapolis, MN, USA bInver Hills Community College, 2500 East 80th Street, Inver Grove Heights, MN 55076 Abstract The popularity of self-report five-style conflict resolution instruments, spawned by Blake and Mouton’s [(1964). The managerial grid. Houston, TX:Gulf Publishing] dual concerns theory, resulted in a plethora of research studies examining possible differences in culture, gender and organizational role. Using the Managerial Grid, dual concerns theory postulates that conflict involves balancing the desire to meet production goals (x) versus concern for personal relationships (y). Five styles of managing conflict are then revealed:smoothing, withdrawing, compromising, problem-solving, and forcing. Numerous studies using instruments derived from this theory validate its basic premises, but results have provided confusing results. Given the disparity of results, a meta-analysis was conducted to provide a clearer overall picture for the variables of culture (individualistic versus collectivistic), gender, and organizational role (superior, subordinate, and peer). Based upon 123 paired comparisons within 36 empirical studies, the results of the meta-analysis indicate:(1) individualistic cultures choose forcing as a conflict style more than collectivistic cultures; (2) collectivistic cultures prefer the styles of withdrawing, compromising, and problem-solving more than individualistic cultures; (3) in individualistic cultures, compromising is endorsed more frequently by females; (4) females are more likely to endorse the use of compromising than males, regardless of culture; (5) males are more likely to report using forcing than females in individualistic ARTICLE IN PRESS www.elsevier.com/locate/ijintrel 0147-1767/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijintrel.2005.06.002 $Questions about article content should be directed to Jennifer Holt at drjenniferholt@yahoo.com. Meta-analytic questions should be directed to Cynthia DeVore, University of Minnesota Department of Psychology, N218 Elliott Hall, 75 E. River Road, Minneapolis, MN, 55455; devo0023@umn.edu. �Corresponding author. Tel.:+1 763 588 1234; fax: +1763 588 1334. E-mail addresses: drjenniferholt@yahoo.com (J.L. Holt), cdevore@inverhills.edu (C.J. DeVore). cultures; and (6) with regard to organizational role, males are more likely than females to choose a forcing style with their superiors. Further research is needed, particularly on the variable of cultural status. r 2005 Elsevier Ltd. All rights reserved. Keywords: Conflict; Conflict resolution; Conflict resolution style; Gender; Culture; Organizational role; Meta-analysis; Managerial grid; Dual concerns theory; Conflict styles; Blake and Mouton; Conflict management survey; CMS; Rahim organizational conflict inventory; ROCI-I; ROCI-II; Employee conflict inventory; ECI; Thomas and Kilmann; Management-of-differences exercise; MODE 1. Introduction Multiculturalism, to flourish, relies on effective, expedient management of disputes. Addressing conflict effectively becomes more urgent as social change accelerates. (Duryea, 1992, p. 1) On a global level, people are increasingly concerned with creating and maintaining peace. Understanding conflict and how to resolve it are two issues directly related to accomplishing this goal, given that resolution of conflict helps to sustain peaceable relations (Blumberg, 1998). Cultural differences both within and across countries can result in conflictive communication; therefore, communication strategies such as conflict resolution may provide an important means of bridging diverse cultural perspectives (Dubinskas, 1992; Gabrielidis, Stephan, Ybarra, Pearson, & Villareal, 1997; Hofstede, 1983; Holt, 2000; Ting-Toomey et al., 1991; Ting-Toomey et al., 2000). Goodall (1996), for example, states: Professionals and academics are being called upon to articulate some new revolutionary ‘communication’ breakthrough capable of teaching us how diverse peoples can learn to live together meaningfully without destroying each other and—in the process—the planet itself. (pp. 1–2) Perhaps never before has this been more important, given the September 11, 2001 attack on the United States’ World Trade Center and Pentagon and its consequences (Kliman & Llerena-Quinn, 2002; Schuman, 2002). Clearly, conflict and violence are at the heart of the world’s problems, on both microcosmic and macrocosmic levels; thus the study of conflict resolution vis-a` -vis culture is an important endeavor. In addition, gender would appear to be a significant way that human beings differ in relationship to conflict resolution style. Similarly, within the workplace, how one chooses to resolve a conflict may be affected in large part by the status of the other party—whether superior, subordinate or peer to oneself. Despite the fact that intuitively, individuals from various cultural backgrounds, of different genders, and within the workplace would appear to solve conflicts in very different ways, there are no conclusive findings. In fact, the results of myriad studies using one of the many five-style conflict resolution instruments and measuring the variables of culture, gender, and organizational role, whether alone or in combination, yield confusing results. Therefore, this study used meta-analytic techniques to contribute a more ARTICLE IN PRESS 166 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 complete picture of possible conflict resolution style differences among cultures, between genders, and with regard to organizational role. 1.1. Background Conflict resolution, defined as ‘‘the process used by parties in conflict to reach a settlement’’ (Sweeney & Carruthers, 1996, p. 328), first gained professional interest in the 1960s due to seminal research conducted by Blake and Mouton (1964). Initially using a population of managers, then eventually extending their ideas to the general population, Blake and Mouton’s dual concerns theory proposed that individuals have two primary motivations with regard to interpersonal conflict:the desire to obtain one’s own goals (concern for production) versus the desire to retain interpersonal relationships (concern for people). By mapping these two concerns on the ‘‘Managerial Grid,’’ five discrete styles for resolving conflict resulted:smoot hing (high concern for people and low concern for production); withdrawing (low concern for both people and production); compromising (medium concern for production and people); problem-solving (high concern for production and people); and forcing (high concern for production versus low concern for people). For example, an individual who is ultimately concerned with meeting production goals, and is willing to sacrifice the desires of others (relationships) to reach these goals would fall under the ‘‘forcing’’ style of conflict resolution. At the opposite end of the grid, someone who is far more concerned with preserving the goodwill of others may choose not to press their particular goals in a conflict, resulting in the style of ‘‘smoothing.’’ Another person might feel both relationships and production are equally high in importance, exhibiting the style of ‘‘problem-solving,’’ in which win-win solutions are generated. On the other hand, for someone who dislikes conflict of any kind, neither meeting production goals nor retaining relationships may be important enough to risk engaging; the style of ‘‘withdrawing’’ would then be a probable choice. Finally, for someone who is willing to give up some of both—goals and relationship—in order to resolve conflict, there is a style in the middle referred to as ‘‘compromising.’’ ‘‘When these basic styles are understood, one can predict for each how a man [sic] operating under that style is likely to handle conflict’’ (Blake & Mouton, 1970, p. 419). In this article, ‘‘five-style paradigm’’ will refer to the set of beliefs Blake and Mouton (1964) and authors with subsequent conflict resolution instruments share: that conflict comes from the opposing forces of production (trying to meet one’s own goals) versus people (attempting to honor the needs of others), and that five basic styles of dealing with conflict are the result:smoo thing, withdrawing, compromising, problem-solving, and forcing. If dual concerns theory is valid, and if the instruments utilizing this theory are valid and reliable, then true differences regarding culture, gender, and organizational role should become clear through meta-analytic techniques. It is important to note, however, that one’s cognitive choices on a selfreport instrument are not the same as one’s behavior. For the purposes of this research, conflict choice is viewed as a cognitive orientation, and ‘‘all measures ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 167 including choice of conflict style will be assessed cognitively rather than behaviorally’’ (Sorenson, Morse, & Savage, 1999, p. 30). Blake and Mouton’s (1964) theory became a popular means of conceptualizing and simplifying a complex issue, given that the grid enabled numeric assignation to each conflict style. Several conflict resolution self-report instruments were subsequently spawned from dual concerns theory, the four most prominent being Hall’s (1969) Conflict Management Survey (CMS); Rahim’s (1983) Rahim Organizational Conflict Inventories I and II (ROCI-I and ROCI-II); Renwick’s (1975) Employee Conflict Inventory (ECI); and Thomas and Kilmann’s (1974) Management-of-Differences Exercise (MODE). While the labels provided for each conflict style vary among instruments (e.g., smoothing is also known as accommodating, obliging and yield-lose), the general principles of the grid (people concerns versus production needs) and basic descriptions of the styles appear very similar (Fig. 1). For purposes of simplification, the names of the styles as originally provided by Blake and Mouton will be utilized throughout this article. It is important to note an implicit value judgment within Blake and Mouton’s (1964) original theory (Dena, 1994). That is, problem-solving was considered the superior mode of solving conflict, being high in concern for people as well as production. This assumption formed the basis for many popular treatises on ARTICLE IN PRESS Smoothing (Blake & Mouton, Renwick) Accommodating (Thomas) Obliging (Rahim) Yield-Lose (Hall) Concern for People (Blake & Mouton) Party’s Desire to Satisfy Other’s Concerns (Thomas) Concern for Others (Rahim) Concern for Relationships (Hall, Renwick) Problem-Solving (Blake & Mouton) Confronting (Renwick) Collaborating(Thomas) Integrating (Rahim) Synergistic (Hall) Compromising (Blake & Mouton, Renwick,Thomas,Rahim,Hall) Withdrawing (Blake & Mouton, Renwick) Avoiding (Thomas, Rahim) Lose-Leave (Hall) Forcing (Blake &Mouton) Competing (Thomas) Dominating (Rahim) Win-Lose (Hall) Concern for Production (Blake & Mouton) Party’s Desire for Own Concern (Thomas) Concern for Self (Rahim) Concern for Personal Goals (Hall, Renwick) Fig. 1. Overlay of conflict resolution styles and authors derived from dual concerns theory. Although the majority of this figure is original, the idea was based on ‘‘Fig. 1, Composite of the Hall, Pruitt, Rahim, and Thomas two-dimensional models with associated conflict styles’’ (Sorenson et al., 1999, p. 27). 168 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 ‘‘win-win’’ business strategy, including Getting to Yes (Fisher & Ury, 1981). Blake and Mouton admit this style is more popular as a choice for respondents from the United States than any other countries; the ethnocentric bias appears to be very clear:‘‘Regar dless of national grouping, managers agree that the 9, 9 [problemsolving] Grid style represents the soundest way to manage’’ (p. 16). In other cultures, however, problem-solving may not be the preferred choice, nor the best means of solving conflict (Kilmann & Thomas, 1978; Lewicki, Weiss, & Lewin, 1992). For example, researchers have provided evidence that, depending upon the particulars of the situation, other styles may be preferred (Elangovan, 1998; van de Vliert, Euwema, & Huismans, 1995). By the late 1970s, the vanguard of conflict resolution researchers had begun to eliminate this bias toward problem-solving in updated work (Kilmann & Thomas, 1978). Initial research by Thomas (1976), one of the most prolific and wellresearched conflict resolution experts, stated: People who practice the withdrawing style tend to behave as if they were indifferent both to their own concerns and to the concerns of others. The withdrawing orientation is often manifested through nonassertive and uncooperative behavior. Those who avoid conflict tend to prefer apathy, isolation and withdrawal to facing conflicts. (p. 892) However, two years later, Thomas, in research with Kilmann, indicates opposition toward value judgments being placed on any one style (Kilmann & Thomas, 1978). Withdrawing, in fact, has been found to be the superior style for many Asians (Chua & Gudykunst, 1987; Ting-Toomey, 1988), perhaps because it is considered more respectful not to argue. Given the wide assortment of research studies that have examined conflict resolution styles, conducted over a span of nearly half a century, an overall evaluation to assess possible culture, gender, and organizational role differences is an important contribution to the field. This research is unique; to the authors’ knowledge, no research has yet been done using rigorous statistical procedures via meta-analysis to analyze the results of conflict resolution self-report instruments within the five-style paradigm. 1.2. Variables of interest Rankings of the five styles of handling conflict were examined with special emphasis on the following variables:cultural status (comparing intercountry, as well as among ethnic minorities of the United States; for the purposes of this research study, the term ‘‘Americans,’’ along with an ethnic identifier, is being used to denote United States citizenship, e.g. ‘‘African Americans’’); gender (including females and males); and organizational role (peer, subordinate, and supervisor—the three levels typically present in hierarchical work settings). ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 169 1.3. Cultural status With regard to cultural status, the growing diversity of the United States has resulted in a multicultural workforce of its citizens (Gabrielidis et al., 1997; Kozan, 1990; Lagao, 1996; Oetzel, 1998). Given this increasing multiculturalism, Yinger (1994) cautions, ‘‘How the United States develops as a multi-ethnic society will be of critical importanceyfor its own quality of life’’ (p. 35). Indeed, many researchers validate the fact that diversity can result in increased conflict. As individuals attempt to communicate and work together, they may react negatively to the cultural practices of others (Gudykunst et al., 1996; Pearson & Stephan, 1998; Ting-Toomey et al., 2000). Yet until recent years, this crucial element was ignored in studies assessing conflict resolution styles, according to researchers such as Gudykunst (1998): Most, if not all, of the cross-cultural [conflict resolution] studies comparing the United States with other cultures have focused on European Americans. There also are differences across ethnic groups in the United States. yFor instance, ythere are severalyareas where European Americans’ and African Americans’ styles of communication may be problematic when they communicate with each other, particularly in a conflict situation. (pp. 253–254) The decrease in the popularity of conflict resolution self-report instruments, most ‘‘in vogue’’ in the United States from the late 1960s into the early 1980s, may be due, in part, to such a lack of cross-cultural inclusiveness. However, a number of research studies, primarily in the past two decades, have measured styles across cultures and/ or countries, thus providing the self-report conflict resolution five-style paradigm with current relevance, and an updated appeal (see D’Silva & Whyte, 1998; Elsayed- Ekhouly & Buda, 1996; Gabrielidis et al., 1997; Gudykunst et al., 1996; Lee, 1990; Lee & Rogan, 1991; Oetzel, 1998; Pearson & Stephan, 1998; Smith & Haar, 1990; Ting-Toomey et al., 2000; Trubisky, Ting-Toomey, & Lin, 1991). The concept of individualism and collectivism provides one means of distinguishing broad differences in cultural values (Hofstede, 1980). While many theories have branched off of this concept, such a distinction continues to be the basis of discussions concerning how styles of conflict resolution may vary across cultures (Elsayed-Ekhouly & Buda, 1996; Oetzel, 1998; Ting-Toomey et al., 1991, 2000). According to Ting-Toomey (1988), members of individualistic cultures prefer direct and assertive methods when resolving conflict. Typically, when comparing communication styles inter-country, such countries as the United States, Canada, Germany, Australia, and England are considered individualistic (Elsayed-Ekhouly & Buda, 1996; Hofstede, 1980, 1983; Trubisky et al., 1991). Individualistic cultures, characterized as more concerned with self than others, are hypothesized to prefer the conflict styles of problem-solving, compromising and forcing. Such styles involve strong verbal communication, less emphasis on internal aspects of communication, and less concern with the needs of others (Hofstede, 1983; Rahim, 1992; Rahim & Blum, 1994). ARTICLE IN PRESS 170 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 On the other hand, in collectivistic cultures such as China, Japan, Korea, the Middle East, and Mexico, the needs of one’s group are considered more important than oneself (Hofstede, 1980, 1983), and conflict communication will reflect this. Styles high in relationship preservation, such as smoothing and compromising, are thus hypothesized to be preferred over forcing (Elsayed-Ekhouly & Buda, 1996; Rahim, 1992; Rahim & Blum, 1994). Withdrawing may also be employed in an effort to ‘‘save face,’’ rather than embarrass others (Ting-Toomey, 1988). Several research studies corroborate these hypotheses. For example, Kagan, Knight, and Martinez- Romero (1982) found that subjects from Mexico (collectivistic) reported using withdrawing and smoothing more than European American (individualistic) subjects, who preferred more active, confrontational strategies such as forcing and problem-solving (see also Gabrielidis et al., 1997; Soto-Fulp, 1996). Pearson and Stephan (1998) found Brazilians (collectivistic) to be more likely to report the use of smoothing and withdrawing with members of their in-group, while United States subjects reported treating out- and in-groups the same. Other studies of conflict resolution instruments utilizing the five-style paradigm have yielded differing results. For example, Lagao (1996) found no significant differences in reported conflict styles between European Americans (individualistic) and Filipinos (collectivistic). Research results validate the postulation that ethnic minorities within the United States may not use the same conflict resolution styles as European Americans (Haslam, Turner, Oakes, McGarty, & Hayes, 1992; Pearson & Stephan, 1998; Ting- Toomey, 1986). African Americans have been hypothesized as individualistic in conflict style; however, regardless of job level status, an African American male may never feel comfortable using certain styles of conflict, such as forcing, due to fear of being negatively stereotyped (Firebaugh, & Davis, 1988; Haslam et al., 1998). Ting- Toomey (1986) conducted research comparing conflict resolution style choices of African American and European American subjects, finding African American males reported less use of forcing and problem-solving conflict strategies than European American males. Interestingly, African American females were more likely to choose the forcing conflict style than African American males or European Americans of either gender. Algert (1998), however, studied conflict style preferences with European Americans, African Americans, and Latin Americans, and found no significant differences among all three. Dena (1994) found significant differences between European Americans and Latin Americans, but the results were counterintuitive— the European Americans preferred the styles of smoothing and withdrawing. In Kim and Kitani’s (1998) research, a comparison of European American and Asian American students bears out individualistic versus collectivistic theory in that Asian American (collectivistic) students preferred smoothing and withdrawing; however, contrary to theory, they also preferred problem-solving. Comparisons with the United States and Middle Eastern countries, the latter of which are considered collectivistic, have yielded notable differences in conflict resolution styles as well. Elsayed-Ekhouly and Buda (1996) compared conflict styles between United States and Arab Middle Eastern [sic] executives, finding United States executives’ conflict style preferences to be forcing, in keeping with predictions, but also smoothing, and compromising. The Arab Middle Eastern executives ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 171 preferred withdrawing, considered typical of collectivistic cultures, but also problemsolving. Similarly, Kozan (1990) found that Turkish/Jordanian managers and United States managers all chose problem-solving as their primary style, but the former ‘‘prefer[red] obliging [smoothing] the least and in this regard differ[red] significantly from the US managers’’ (p. 179; see also Agee & Kabasakal, 1993). The contradictory results of such studies indicate the need for a meta-analysis to more thoroughly understand true cultural differences as measured by instruments within the conflict resolution five-style paradigm. 1.4. Gender Socially appropriate behavior differs for females and males in many countries around the world; thus, it is probable to assume that females and males would prefer to resolve conflicts with different conflict style choices (Shockley-Zalabak, 1981). In the United States, historically, males have been socialized to communicate in direct, confrontational ways, assuming the dominant power position; females have been socialized to take care of others, and play a more receptive role (Gilligan, 1977; Stockard & Lach, 1989; Zammuto, London, & Rowland, 1979). Kolb (1993) states: Existing research and our own experience suggest that the voices of women are often hushed in formal negotiation. Conflict and competition are important in formal negotiation, and therefore, it may not be a comfortable place for many women. (p. 139) Given this difference, styles such as forcing (high in production, low in relationships) or problem-solving (high in production and relationships) have been postulated as popular choices for males on conflict resolution self-report instruments (Mills & Chusmir, 1988). Females, for whom relationships may be of greater importance, and for whom aggressive behavior is less condoned (Ting-Toomey, 1986), would seem more likely to prefer such styles as smoothing (high in relationships, low in production), withdrawing (low in production and relationships), and compromising (medium in production and relationships). As Ting- Toomey (1986) states, ‘‘[M]ales typically engage in more direct, ‘up-front’ strategies. yFemales typically engage in either indirect, ‘smoothing’ communication strategies to diffuse the conflict topic, or engage in avoidance or withdrawal strategies’’ (p. 79). In the first two decades after the inception of conflict resolution self-report instruments, conflict studies primarily used a respondent base of males. Renwick (1977) was one of the first researchers to examine differences in conflict resolution styles between male and female management personnel in the United States. While Renwick argued from a feminist point of view that females ought to be no less apt to choose aggressive styles than males, her results indicated males tended to rate the forcing style higher than females. Mills and Chusmir (1988), studying managers in the United States, found similar results:‘‘[N] ot surprisingly, men were slightly more likely to compete [force] at work’’ (p. 307). Nelson and Lubin (1991) determined that females were significantly higher on smoothing, when asking United States politicians about their conflict styles. Content (1986) found female principals in ARTICLE IN PRESS 172 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 the United States reported higher use of the compromising mode than male principals. Cardona (1995) found females from a Midwestern university population in the United States to report more use of withdrawing than males. Rahim (1983) created norming data with one of his instruments (ROCI-II), using a population of 1219 United States executives. Interestingly, his data indicated males rated smoothing higher, while females preferred problem-solving, withdrawing and compromising (however, only 50 female subjects were included). Zammuto et al. (1979) reported supervisors in United States companies who were asked to rate their subordinates’ use of conflict resolution styles rated males as more frequently using compromising; females were rated as predominantly using forcing. Muir’s (1991) research, in which middle managers in the United States were studied, was intended to corroborate Rahim’s general findings. However, she was unable to duplicate his results—no significant differences were discovered. Likewise, no significant differences were found between genders by Shockley-Zalabak (1981), in studying managers at Colorado companies; or in Sternberg and Soriano’s (1984), and Sternberg and Dobson’s (1987) research with United States college students. Such contradictory results indicate a meta-analysis of all data on male and female conflict style choices may reveal true differences, if they exist. Correlating gender with culture may also prove important, given perceived gender differences within various ethnicities indigenous to the United States, as well as in comparison with other countries. 1.5. Organizational role Given the history of Blake and Mouton’s (1964) Managerial Grid, and its origins in analyzing company conflict, the plethora of studies exploring conflict resolution style differences within the organizational hierarchy (superiors, peers, and subordinates), is unsurprising (see Conrad, 1985; Harris, 1988; Mills & Chusmir, 1988; Musser, 1982; Oetzel, 1998, 1999; Rahim, 1986, 1992; Rahim & Buntzman, 1989; Renwick, 1975, 1977). Theoretically, given power differences, superiors are generally predicted to prefer problem-solving, compromising and forcing, peers are predicted to be less aggressive with superiors than each other, but more so with subordinates, and subordinates are predicted to tend toward the least aggressive styles—withdrawing and smoothing (Mills & Chusmir, 1988; Musser, 1982; Oetzel, 1999; Rahim & Buntzman, 1989; Renwick, 1975, 1977). That is, given inherent power differences, a subordinate may not be willing to engage in any conflict style that challenges a superior, while a superior may have more leeway to use aggressive techniques, particularly in order to meet company production goals (Rahim & Buntzman, 1989). Peers are considered most likely to use compromising with each other, given the equality of power. Research would appear to bear organizational role predictions out to a certain extent (Mills & Chusmir, 1988; Musser, 1982; Oetzel, 1999; Phillips & Cheston, 1979; Rahim & Buntzman, 1989). For example, Phillips and Cheston (1979) studied business managers, and found that superiors were more likely to choose the use of forcing with subordinates than vice versa, while compromising was the style most ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 173 likely to be chosen for use with one’s peers. Several researchers have found subordinates to prefer the styles of withdrawing or smoothing when in conflict with superiors, perhaps due to the risk of negative consequences such as job loss (Kahn, Wolfe, Quinn, Snoek, & Rosenthal, 1964; Phillips & Cheston, 1979; Rahim, 1986). However, Renwick’s (1975) research with United States business companies found the top three styles for superiors to be problem-solving, compromising and smoothing, contrary to theorized predictions. Renwick also measured subordinates’ conflict style rankings, finding the top three styles to be compromising, problemsolving, and forcing. Similarly, Rahim (1983) found subordinates most likely to prefer the use of problem-solving and forcing. Paulson’s (1986) research with middle managers from the United States, on the other hand, found no significant differences. Such conflicting results again indicate the need for a thorough overall evaluation of the findings via meta-analytic techniques. 2. Method 2.1. Location of studies In order to make a thorough assessment of studies, both published and unpublished, utilizing instrumental derivatives of Blake and Mouton’s (1964) Managerial Grid, electronic searches were conducted of Psychological Abstracts, the Social Sciences Citation Index, Dissertation Abstracts, The International Journal of Conflict Management and The Journal of Conflict Resolution. Search words used in the indices included:‘‘Man agerial Grid’’; ‘‘Blake and Mouton’’; ‘‘conflict resolution styles’’; ‘‘conflict resolution instruments’’; ‘‘mediation’’; ‘‘conflict styles’’; and the names of all authors with instruments utilizing five styles of resolving conflict. This initial search yielded over 3000 articles, but less than 200 could be considered empirical studies. A thorough examination was conducted of this grouping. In addition, ‘‘snowball sampling’’ was used:refer ences at the end of these articles were perused to obtain additional studies (Oetzel, 1998). The search included all studies published and non-published through 2002. Findings support the supposition that dual concerns theory’s five-style paradigm is one of the most frequently used theoretical paradigms in organizational conflict resolution research. 2.2. Exclusion criteria For the purposes of this meta-analytic investigation, a research study was included if (1) the study used dual concerns theory, the Managerial Grid, or the five conflict resolution styles explicitly or implicitly via related instruments, (2) the conflict resolution styles were self-reported, and (3) the study provided enough statistical information for the calculation of an effect size, the standardized difference in scores for two different groups, calculable from the means and standard deviations of scores and group sizes. ARTICLE IN PRESS 174 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 One hundred and one studies were considered for selection in the meta-analysis. When a study’s results were presented in two publications, as determined by sample sizes and means and standard deviations, the peer-reviewed study was included and the other excluded (three studies). When a particular study did not provide means and standard deviations or the equivalent, an attempt was made to contact the author(s). Thus, studies were excluded if the authors could not be found (nine studies), the authors did not respond to inquiries (13 studies), or the authors responded and no longer had the data available (five studies). Studies were excluded if they measured conflict resolution styles for a single group and did not provide the gender or cultural composition of the group with a t-test (22 studies). Studies were excluded if they were incompatible because they were not self-reported scores on a Likert-type scale for conflict resolution styles that could translate to the Blake and Mouton (1964) Managerial Grid (six studies). The factor analysis of one exemplar of this type of excluded study, Putnam and Wilson’s (1982) Organizational Communication Conflict Instrument (OCCI), resulted in three, rather than five styles. When a study did not report its non-significant findings or there were unexplained inconsistencies in sample sizes, it was excluded (seven studies). Of the excluded studies, 45 were from journals, 17 were unpublished dissertations or theses, and three were from book chapters. 2.3. Final data set Thirty-six studies constituted the final meta-analysis data set. Of the included studies, 21 were from journals, 13 were unpublished dissertations, one was a book chapter, and one was a conference paper. These 36 studies included a total of 123 groups, for which there were two through five means and standard deviations. Five studies reported results for pairs, providing six comparisons of either two or five effect sizes. 2.4. Coding of studies For each study, the conflict resolution instrument, the mean and standard deviation for each group measured (or alternately the t-value, F-value or correlation coefficient associated with two groups), the number of individuals in each tested group and the number of each gender in each tested group were captured. Codings were made for several additional elements as well, to facilitate testing possible interactions, including type of participant (student, worker, manager, teacher, principal) and the type of individual with whom the participant was in conflict. The first 30 studies were coded on coding sheets by three individuals. The first five studies were coded in common by all three coders for interrater reliability. The means and standard deviations reported by the three coders were the same for the five studies. The remaining studies were entered directly into either a Microsofts Excel spreadsheet or the Microsofts Access database. Once all studies were entered into the Microsofts Access database, the means and standard deviations were checked against the published studies. ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 175 2.5. Statistical moderators The type of instrument used to measure conflict resolution style and the year the study data were collected were possible moderators; however there were too few studies to conduct a moderator analysis of them. 2.6. Computational procedures A Microsofts Access database was used to enter the data for each study. The following data were collected:the instrument used to measure the conflict resolution style; the gender; and the context of the culture. In order to calculate d-scores, the means and standard deviations for each group were used or calculated, or, when necessary, derived from the t-values, F-values, or correlations associated with compared groups. After all studies were entered, groups were matched to form pairs for analysis. Three major variables were selected for further analysis:culture, gender, and organizational role. The decisions concerning placement of groups into individualistic and collectivistic categories were based on prior research (Elsayed-Ekhouly & Buda, 1996; Hofstede, 1980, 1983; Oetzel, 1998; Ting-Toomey et al., 1991, 2000). For each study in which culture was included, all possible individualistic/collectivistic culture combinations were generated, such that collectivistic cultures were the referent group. If the cultural classification of a group was not identified, the group was excluded. For each study in which gender was included, all possible female-male combinations were generated, such that males were the referent group. If a group included both genders, that group was excluded. For each study in which organizational role was included, all possible peer-subordinate, subordinate-superior and superior-peer combinations were generated, such that subordinates, superiors, and peers were the referent groups for the three comparisons, respectively. 2.7. Correction of artifacts Sampling error was calculated using Eq. (1) from Hedges and Olkin (1985): s2 e ¼ N1 þ N2 N1N2 þ d2 2ðN1 þ N2Þ . (1) The formula adjusts for differences in the sample sizes used to generate each d-value. Because the measurement instruments were imperfect and the differences in true scores were sought, it was necessary to correct for unreliability. Reliability information was available for eight of the twelve instruments in the usable studies. Test–retest reliabilities were available for seven of the twelve instruments, accounting for 92% of all effect sizes. Where the test–retest reliability was not available (as for Danes, Leichtentritt, Metz, & Huddleston-Casas, 2000; Gabrielidis et al., 1997; Kim & Kitani, 1998; Miyazaki, Moroi, & Stephan, in press; Renwick, 1977; Zammuto et al., 1979), an estimate of the test–retest reliability was calculated based upon the item ARTICLE IN PRESS 176 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 counts for the unknown instrument and the reliabilities per item for each known instrument (see Table 1). The mean effect sizes were corrected for unreliability using the following Eq. (2) from Hunter and Schmidt (1990): d ¼ dk P N1þN2 2 � �ðaÞ , (2) where a is the correction for the effect size and k is the number of comparisons. 3. Results 3.1. Interpreting meta-analytic results The point of a meta-analysis is to cumulate the results of several studies in order to determine true effects in the population. Applying meta-analytic results is not as straightforward, however, as generalizing to the entire population because more variables than the effect size need to be considered. Thus, additional values are reported to aid in the interpretation of meta-analytic results. For each of the conflict resolution styles and comparison types (culture, gender, organizational role), we reported the mean of the observed effect size (d) and the true effect size (d) corrected for sampling error and unreliability in the measuring instruments. The effect size is a measure of the standardized difference between the two groups. The variability of the effect sizes was reported in the observed and corrected standard deviations of the effect sizes (sd and sd, respectively). The number of groups used to calculate the mean effect size (k) is to a meta-analysis what N is to typical empirical studies; even though the underlying studies may have hundreds of participants, a meta-analysis is based upon k. Just as an empirical researcher is hesitant to make generalizations based upon four participants, so meta-analysts are hesitant to generalize results when ko5. We listed the actual number of studies from which the effect sizes were calculated (k0) because some studies supplied more than one comparison. The N from each empirical study is summed into PN. Three less familiar types of measures determine degree of confidence regarding meta-analytic results. First, when the lower and upper bounds of the credibility interval (CV) include zero, the operation of a moderator is likely. In other words, ARTICLE IN PRESS Table 1 Test–retest reliability of conflict resolution instruments Methodology Smoothing Withdrawing Compromising Problem-solving Forcing CMS 0.53 0.61 0.41 0.54 0.66 MODE 0.62 0.68 0.66 0.63 0.61 ROCI-II 0.81 0.79 0.60 0.83 0.76 van de Vliert 0.66 0.71 0.71 0.75 0.81 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 177 within the particular population being studied, there are likely multiple subgroups with distinct effect sizes. As a hypothetical example, within culture, males and females might provide distinct subgroups. The lower and upper bounds of the second measure, the confidence interval (CI), suggest the variability of the true effect size. When the confidence interval includes zero, the direction of the true effect cannot be determined. For instance, with a confidence interval including zero, there would be no difference regarding conflict style preferences between males and females. The size or amplitude of the third measure, the Failsafe N (Nf), is the theorized number of contradictory effect sizes that would be required to change the results’ interpretation from true to negligible differences between the comparison groups. In this study, a difference of 0.20 was chosen. The criterion for Failsafe N was set to the current number of comparisons (Nf ¼ k). Therefore, true differences existed between the compared groups when (1) the number of compared groups exceeded three, (2) the effect size exceeded 0.20, (3) the confidence interval did not include zero, and (4) the Failsafe N exceeded the number of comparisons. 3.2. Culture and gender Between 14 and 17 comparisons were used to determine differences in conflict resolution style by culture. All conflict resolution styles except smoothing exhibited generalizable results. Withdrawing (d ¼ �1:66) and compromising (d ¼ �1:19) showed the largest effects, signifying that persons in collectivistic cultures tend to choose withdrawing and compromising more than persons in individualistic cultures. In addition, both withdrawing and compromising had credibility intervals suggesting multiple populations, e.g., gender, were represented. Problem-solving followed the same pattern as withdrawing and compromising, with collectivistic cultures reporting more use than individualistic cultures. Forcing showed the opposite effect, with persons in individualistic cultures exhibiting more forcing than those in collectivistic cultures (d ¼ 1:13), as shown in Table 2. Based upon a probable moderator operating in the individualistic to collectivistic comparisons for withdrawing and compromising, the studies were divided in which gender of respondents was reported. For gender within individualistic cultures, the corrected effect sizes were between �0.42 and 0.73, with only one style, compromising, showing generalizable results. In individualistic cultures, females endorsed the use of the compromising style more frequently than males (d ¼ 0:73). Only two studies reported conflict resolution style scores by gender for collectivistic cultures, thus a meta-analysis was premature. Preliminary results suggested that withdrawing was self-ascribed more by males than females in collectivistic cultures (d ¼ �0:39) (Fig. 2). Gender provided the greatest number of comparisons for the meta-analysis. Between 27 and 29 comparisons were used to determine differences in conflict resolution style by gender. As shown in Table 2, the five corrected effect sizes ranged from �0.31 to 0.64. Smoothing and withdrawing showed essentially no effect for gender. Compromising was reported more for females than for males (d ¼ 0:64). Although the corrected sampling error accounted for all of the observed variance, as ARTICLE IN PRESS 178 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 the Failsafe N indicates, it would take very few studies with contrary results to possibly negate the findings for problem-solving (two studies) and forcing (16 studies). 3.3. Gender and organizational level Several studies examined conflict style preferences among peers, subordinates, and superiors in the workplace. Three sets of comparisons were utilized:pe ers to subordinates, subordinates to superiors, and superiors to peers. Table 3 and Fig. 4 show the results of this meta-analysis. In general terms, in comparing conflict resolution styles with one’s peers to one’s subordinates, withdrawing (d ¼ 0:93) and compromising (d ¼ 1:32) were used more with peers than subordinates, and forcing ARTICLE IN PRESS Table 2 Effect sizes for conflict resolution styles for gender and culture Conflict resolution style d d K k0 i PN sd sd CV CI Nf Culture (individualistic– collectivistic) Smoothing �0.12 �0.26 14 10 7 4648 0.69 0.00 {�0.26, �0.26} {�0.26, �0.26} 4 Withdrawing �0.73 �1.66 17 11 7 5694 0.88 0.74 {�4.50, 1.18} {�1.78, �1.54} 124 Compromising �0.46 �1.19 15 9 5 5349 0.78 0.58 {�3.41, 1.02} {�1.29, �1.10} 75 Problem-solving �0.27 �0.61 17 11 7 5694 0.68 0.00 {�0.61, �0.61} {�0.61, �0.61} 35 Forcing 0.49 1.13 17 11 7 5694 0.56 0.00 {1.13, 1.13} {1.13, 1.13} 79 Gender within individualistic culture (female– male) Smoothing 0.02 0.05 12 9 5 2626 0.14 0.00 {0.05, 0.05} {0.05, 0.05} 9 Withdrawing 0.08 0.18 11 9 5 2407 0.13 0.00 {0.18, 0.18} {0.18, 0.18} 1 Compromising 0.28 0.73 11 9 5 2407 0.23 0.00 {0.73, 0.73} {0.73, 0.73} 29 Problem-solving 0.02 0.05 11 9 5 2407 0.20 0.00 {0.05, 0.05} {0.05, 0.05} 8 Forcing �0.18 �0.42 11 9 5 2407 0.11 0.00 {�0.42, �0.42} {�0.42, �0.42} 12 Gender within collectivistic culture (female– male) Smoothing 0.10 0.22 4 2 2 410 0.05 0.00 {0.22, 0.22} {0.22, 0.22} 0 Withdrawing �0.17 �0.39 3 2 2 227 0.05 0.00 {�0.39, �0.39} {�0.39, �0.39} 3 Compromising 0.17 0.42 3 2 2 227 0.16 0.00 {0.42, 0.42} {0.42, 0.42} 3 Problem-solving 0.41 0.92 3 2 2 227 0.14 0.00 {0.92, 0.92} {0.92, 0.92} 11 Forcing �0.17 �0.39 3 2 2 227 0.07 0.00 {�0.39, �0.39} {�0.39, �0.39} 3 Gender (female– male) Smoothing 0.01 0.03 28 19 7 5050 0.21 0.00 {0.03, 0.03} {0.03, 0.03} 23 Withdrawing �0.01 �0.03 27 18 6 4638 0.22 0.00 {�0.03, �0.03} {�0.03, �0.03} 23 Compromising 0.25 0.64 29 19 6 4799 0.23 0.00 {0.64, 0.64} {0.64, 0.64} 63 Problem-solving 0.09 0.21 29 20 7 5203 0.29 0.00 {0.21, 0.21} {0.21, 0.21} 2 Forcing �0.13 �0.31 29 19 7 4869 0.22 0.00 {�0.31, �0.31} {�0.31, �0.31} 16 Note:Bolded figures are generalizable. d ¼ mean effect size; d ¼ corrected effect size; k ¼ number of effect sizes in mean effect size; k0 ¼ the number of studies from which the k effect sizes were derived; i ¼ number of Conflict Resolution instruments;PN ¼ sum of participants in each group; sd ¼ SD of mean effect size; sd ¼ SD of corrected effect size; CV ¼ range of Credibility Interval; CI ¼ range of Confidence Interval; Nf ¼ Failsafe N. J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 179 was reported as being used less with peers than with subordinates (d ¼ �1:17). The wide credibility intervals for compromising and forcing suggested the presence of a moderator, such as gender or culture; however, there were insufficient studies to complete this analysis. With regard to subordinates versus superiors, all of the styles appear to be generalizable. Smoothing (d ¼ �1:51) and withdrawing (d ¼ �1:89) were endorsed more with superiors than with subordinates. Compromising (d ¼ 1:51), problemsolving (d ¼ 1:08), and forcing (d ¼ 0:81) are endorsed more for use with subordinates than with superiors. All styles, with the exception of problem-solving, have credibility intervals spanning zero, suggesting that a moderator was operating. The styles used by respondents in conflict with superiors as compared to their peers reflected strong results, as shown in Table 3. Only forcing failed to reach the Failsafe N benchmark. Smoothing (d ¼ 2:84) and withdrawing (d ¼ 0:77) were reported to be used more with superiors than with peers. The relationship was in the opposite direction for compromising and problem-solving, as these styles were used less with superiors than with peers (ds ¼ �3:04 and �0.68, respectively). Compromising was the only conflict resolution style whose credibility interval suggested a moderator was present. As Fig. 3 demonstrates, respondents preferred smoothing and withdrawing styles when in conflict with superiors as opposed to ARTICLE IN PRESS 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 -2.00 Smoothing Withdrawing Compromising Problem Solving Forcing Corrected Effect Size Culture (Individualistic - Collectivistic) Gender within Individualistic Culture (Female - Male) Gender within Collectivistic Culture (Female - Male) Gender (Female - Male) True Difference Criterion .20 > d > -.20 Fig. 2. Corrected Effect sizes of conflict resolution styles for culture, gender within culture, and gender. For each style, positive effect sizes indicate the first named group chose the style more frequently; negative effect sizes indicate the second named group chose the style more frequently. 180 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 peers or subordinates. On the other hand, compromising was the style chosen for conflict with peers, when compared with superiors or subordinates. As noted previously, there were insufficient data to complete a meta-analysis dividing organizational information by gender or culture. There were adequate data to compare gender within referent organizational level for many of the conflict resolution styles. In other words, the meta-analysis answered the question: How do females and males differ regarding conflict resolution style with peers, superiors, and subordinates? It is important, however, to use caution when discussing the results because so few groups were used in the comparisons. Only one of the comparisons met all four criteria for true differences—forcing with superiors had an effect size of �0.70, meaning that males chose this style more than females. Putting aside the criteria for the number of comparisons, the same pattern emerged for forcing with peers (d ¼ �1:31) and subordinates (d ¼ �0:90) vis-a` -vis males. In addition, females chose smoothing more with subordinates (d ¼ 0:71) than males. Finally, females were also more likely to report the use of problemsolving with peers (d ¼ 0:63) and subordinates (d ¼ 1:37) than males (see Table 4 and Fig. 4). ARTICLE IN PRESS Table 3 Effect sizes for conflict resolution styles comparing organizational levels Conflict resolution style d d k k0 i PN sd sd CV CI Nf Organizational level (peers– subordinates) Smoothing �0.09 �0.20 15 4 1 1560 1.19 1.12 {�4.49, 4.09} {�0.39, 0.00} 0 Withdrawing 0.42 0.93 15 4 1 1560 0.50 0.00 {0.93, 0.93} {0.93, 0.93} 55 Compromising 0.51 1.32 11 3 1 948 1.81 2.35 {�7.71, 10.36} {0.88, 1.77} 62 Problem-solving 0.15 0.34 15 4 1 1560 0.84 0.00 {0.34, 0.34} {0.34, 0.34} 10 Forcing �0.51 �1.17 15 4 1 1560 1.28 1.08 {�5.31, 2.98} {�1.35, �0.98} 72 Organizational level (subordinates– superiors) Smoothing �0.68 �1.51 17 5 2 1978 1.77 2.18 {�9.89, 6.87} {�1.86, �1.16} 111 Withdrawing �0.84 �1.89 17 5 2 1978 1.37 1.11 {�6.16, 2.38} {�2.07, �1.71} 143 Compromising 0.59 1.51 4 3 2 682 0.90 1.30 {�3.50, 6.51} {1.10, 1.92} 26 Problem-solving 0.49 1.08 17 5 2 1978 0.66 0.00 {1.08, 1.08} {1.08, 1.08} 75 Forcing 0.35 0.81 17 5 2 1978 1.36 1.31 {�4.23, 5.84} {0.60, 1.02} 52 Organizational level (superiors– peers) Smoothing 1.28 2.84 20 5 1 2412 0.81 0.00 {2.84, 2.84} {2.84, 2.84} 264 Withdrawing 0.34 0.77 20 5 1 2412 0.62 0.00 {0.77, 0.77} {0.77, 0.77} 57 Compromising �1.18 �3.04 7 3 1 1116 2.46 3.51 {�16.52, 10.44} {�3.87, �2.21} 99 Problem-solving �0.31 �0.68 20 5 1 2412 0.67 0.00 {�0.68, �0.68} {�0.68, �0.68} 48 Forcing �0.12 �0.27 20 5 1 2412 0.61 0.00 {�0.27, �0.27} {�0.27, �0.27} 7 Note:Bolded figures are generalizable. d ¼ mean effect size; d ¼ corrected effect size; k ¼ number of effect sizes in mean effect size; k0 ¼ the number of studies from which the k effect sizes were derived; i ¼ number of Conflict Resolution instruments;PN ¼ sum of participants in each group; sd ¼ SD of mean effect size; sd ¼ SD of corrected effect size; CV ¼ range of Credibility Interval; CI ¼ range of Confidence Interval; Nf ¼ Failsafe N. J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 181 4. Discussion Although this meta-analysis was conducted with a relatively small number of groups for comparisons from even fewer studies, it produced results that clarify and quantify the comparative use of conflict resolution styles by culture, gender within culture, gender, comparative organizational level, and gender within organizational level. Results for culture indicate individuals within a collectivistic orientation prefer withdrawing and compromising more than those with an individualistic cultural orientation. Second, the latter choose forcing more than the former. In all three findings, the group means differ by more than one standard deviation, strong evidence of true differences. The results regarding withdrawing corroborate prior research and discussion on the subject, given that individuals within collectivistic cultures prefer strategies that ‘‘save face’’ (Ting-Toomey, 1988). Similarly, compromising, in which one gives up some of one’s needs, can be understood within the collectivistic paradigm, given that the needs of the group supercede one’s own (Hofstede, 1980, 1983). The finding that forcing, high on concern for production, and low on concern for others, is endorsed by individuals from individualistic cultures by over one standard deviation, also backs up the theorized emphasis on ‘‘me’’ as opposed to ‘‘we’’ in such cultures (Hofstede, 1980, 1983). A final interesting result within culture concerns problem-solving. Contrary to original theory (Blake & Mouton, 1964), collectivistic cultures prefer problem-solving more than individualistic cultures by over half a standard deviation. This is ironic, ARTICLE IN PRESS Peers Peers Peers Peers Peers Subordinates Subordinates Subordinates Subordinates Subordinates Superiors Superiors 0.00 Superiors Superiors Superiors 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Smoothing Withdrawing Compromising Problem-Solving Forcing Conflict Resolution Style Relative Effect Size (Corrected) Fig. 3. Corrected effect sizes of conflict resolution styles for comparison of organizational level. 182 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 considering Blake and Mouton’s contention that ‘‘[m]anagers from South America and Japan identify themselves as the least likely to possess the 9, 9 [problem-solving] style’’ (Dena, 1994, p. 33). In fact, collectivistic cultures are more concerned with creating a ‘‘win-win’’ situation. Regarding gender comparisons, the results indicate differences in the self-reported use of compromising and forcing; however, compromising is the only style for which differences in gender exceed the study criteria for true differences. Females report using compromising more than males by a sizable margin (over half a standard deviation). Such results support societal notions in the United States concerning gender differences, where females may be more likely than males to give up part of their own needs in order to compromise (Kolb, 1993; Ting-Toomey, 1986). That there are no differences between females and males in reported use of smoothing and withdrawing, however, is contrary to popular notions that females are more willing to smooth over conflict or withdraw from it altogether. When gender and culture are analyzed together, results indicate, as in the results for gender alone, that compromising is used more frequently by females in both individualistic and collectivistic cultures, although the difference is greater for individualistic females (three-quarters of a standard deviation). In addition, both ARTICLE IN PRESS Table 4 Effect sizes for conflict resolution styles for gender within organizational level (females–males) Conflict resolution style d d k k0 i PN sd sd CV CI Nf Peers Smoothing 0.14 0.31 6 3 1 656 0.14 0.00 {0.31, 0.31} {0.31, 0.31} 3 Withdrawing 0.04 0.08 6 3 1 656 0.26 0.00 {0.08, 0.08} {0.08, 0.08} 4 Compromising 0.15 0.38 1 1 1 137 0.00 0.00 {0.38, 0.38} {0.38, 0.38} 1 Problem-Solving 0.29 0.63 2 2 1 290 0.35 0.37 {�0.80, 2.06} {0.45, 0.81} 4 Forcing �0.57 �1.31 2 2 1 290 0.02 0.00 {�1.31, �1.31} {�1.31, �1.31} 11 Subordinates Smoothing 0.32 0.71 2 2 1 249 0.16 0.00 {0.71, 0.71} {0.71, 0.71} 5 Withdrawing �0.23 �0.52 2 2 1 249 0.00 0.00 {�0.52, �0.52} {�0.52, �0.52} 3 Compromising �0.34 �0.87 1 1 1 96 0.00 0.00 {�0.87, �0.87} {�0.87, �0.87} 3 Problem-Solving 0.63 1.37 2 2 1 249 0.03 0.00 {1.37, 1.37} {1.37, 1.37} 12 Forcing �0.39 �0.90 2 2 1 249 0.17 0.00 {�0.90, �0.90} {�0.90, �0.90} 7 Superiors Smoothing 0.10 0.22 7 4 1 1437 0.27 0.00 {0.22, 0.22} {0.22, 0.22} 1 Withdrawing �0.05 �0.11 7 4 1 1437 0.22 0.00 {�0.11, �0.11} {�0.11, �0.11} 3 Compromising 0.07 0.18 6 3 1 1284 0.17 0.00 {0.18, 0.18} {0.18, 0.18} 1 Problem-Solving 0.17 0.37 7 4 1 1437 0.34 0.00 {0.37, 0.37} {0.37, 0.37} 6 Forcing �0.31 �0.70 6 3 1 1095 0.24 0.00 {�0.70, �0.70} {�0.70, �0.70} 15 Note:Bolded figures are generalizable. d ¼ mean effect size; d ¼ corrected effect size; k ¼ number of effect sizes in mean effect size; k0 ¼ the number of studies from which the k effect sizes were derived; i ¼ number of Conflict resolution instruments; PN ¼ sum of participants in each group; sd ¼ SD of mean effect size; sd ¼ SD of corrected effect size; CV ¼ range of credibility interval; CI ¼ range of confidence interval; Nf ¼ failsafe N. J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 183 individualistic and collectivistic males choose forcing more than their female counterparts (nearly half a standard deviation for each). This reinforces the idea that males are more likely to use aggressive tactics in order to achieve their own ends (Kolb, 1993; Ting-Toomey, 1986). Interestingly, collectivistic males report preferring withdrawing and collectivistic females report preferring problem-solving. However, the results regarding collectivistic cultures are very tentative. With regard to organizational role, patterns of difference are stronger than for culture or gender. Employees endorse the use of smoothing as a conflict resolution style more with superiors than with peers by nearly three standard deviations. Such results corroborate current literature:The more power or status another worker has in a job situation as compared to the respondent, the more likely the respondent’s own goals will be sacrificed in order to preserve the relationship when conflict arises (Kahn et al., 1964; Phillips & Cheston, 1979; Rahim, 1986). Compromising shows an equally strong effect, in that the style is preferred more with one’s peers than one’s superiors (three standard deviations). This backs up organizational theory concerning the sharing of power, ‘‘whereby both parties give up something to make a mutually acceptable decision’’ (Rahim & Buntzman, 1989, p. 197). That is, peers would seem more likely to compromise and ‘‘share’’ with each other, than with superiors, who have greater power. Employees report using the withdrawing conflict style more with peers than with subordinates by almost a standard deviation. This also supports literature and ARTICLE IN PRESS Smoothing Withdrawing Compromising Problem Solving Forcing Corrected Effect Size Gender Peers (Females - Males) Gender Subordinates (Females - Males) Gender Superiors (Females - Males) Gender (Female - Male) True Difference Criterion .20 > d > -.20 2.00 1.50 1.00 0.50 0.00 -0.50 -1.00 -1.50 Fig. 4. Corrected effect sizes of conflict resolution styles for gender by organizational role of referent. 184 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 research; simply, workers in conflict with peers may be more concerned with negative outcomes than when in conflict with subordinates, given that the latter have less power and status (Kahn et al., 1964; Phillips & Cheston, 1979; Rahim, 1986; Rahim & Buntzman, 1989). Similarly, employees report preferring the forcing conflict style more with subordinates than with peers by over a standard deviation. As past research indicates, the lower the status of the other worker versus oneself, the less inclined an employee will be to use passive styles, and the greater the inclination to endorse the use of aggressive conflict styles (Rahim & Buntzman, 1989). Somewhat weaker results indicate forcing is used less with peers than with subordinates. This is again in keeping with organizational modes of behavior. As Lee (1990) states, employees in conflict ‘‘tend to use different influence tactics, depending upon the relative status of the agent and target. More often assertive [forcing, problemsolving] tacticsywere used to influence subordinates rather than peers or superiors’’ (p. 331). Regarding gender across organizational role, the pattern of females choosing problem-solving more than males and males choosing forcing more than females is maintained regardless of organizational role. Kolb (1993) notes:‘‘[T]here are significant differences in the ways men and women approach negotiation and the styles they use in searching for an agreement [in the workplace]’’ (pp. 138–139; see also Soto-Fulp, 1996). 4.1. Limitations First and foremost, a significant number of studies could not be included for analysis due to insufficient information. In fact, two-thirds of the studies using selfreport conflict resolution style instruments were not viable. The five-style paradigm, upon which the ‘‘old guard’’ of conflict resolution theory and subsequent instruments are based, achieved its zenith during the 1970s and early 1980s. Historically, this era preceded the advent of stringent statistical standards in measurement and publication of study results. Thus, many of the studies did not publish the means and/or standard deviations for style preferences, which are the necessary statistics for a meta-analytic calculation. Second, there are some indications that the five-style paradigm has become outdated, as several interesting, rigorous studies from the more recent past have employed newer evolutions of conflict resolution self-report instrumentation (Knapp, Putnam, & Davis, 1988; Sternberg & Dobson, 1987; Sternberg & Soriano, 1984; Thomas, 1988; Trubisky et al., 1991; van de Vliert & Hordijk, 1989; van de Vliert et al., 1995). As van de Vliert et al. (1995) indicate, it is entirely possible that the five styles are too parsimonious to realistically represent the continuum of conflict resolution strategies. In addition, van de Vliert et al. argue that conflict styles may operate together, rather than discretely: ‘‘Rather than a single behavior [resulting from one conflict style], handling conflict is a conglomeration of behavioral components’’ (p. 271; see also Elangovan, 1998; Musser, 1982; Volkema & Bergmann, 1995). Researchers have also critiqued the validity and distinctness of each style. Van de Vliert and Hordijk (1989) found that ‘‘the social-psychological consequences of compromising and problem-solving tend ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 185 to be the same, however different the behaviors may be’’ (p. 681). That is, there may not be a solid justification for viewing compromising as a strategy equally differentiated from the other four styles. Social desirability is important to mention when discussing self-report instruments, due to the invalid assumption ‘‘that cognition is [necessarily] associated with choice of [conflict] style’’ (Sorenson et al., 1999, p. 26; see also Cosier & Ruble, 1981; Drake, Zammuto, & Parasuraman, 1981; Elangovan, 1998; Kabanoff, 1987; Musser, 1982; Putnam & Wilson, 1982; Thomas & Kilmann, 1975). That is, what a respondent reports for preference regarding conflict resolution style may be far different from the style utilized in real-life situations. With regard to another selfreport instrument, the Argumentativeness Scale, Nicotera (1996) states, [S]ince the act of argument is sanctioned more for men than women, women answering the scale may be reluctant to rate themselves as argumentative. Such reluctance may or may not reflect their actual level of trait argumentativeness. (p. 25) Similarly, females filling out a self-report conflict resolution instrument may be less inclined than males to report using conflict styles considered ‘‘unfeminine,’’ such as forcing. Instead, females may self-report preferring styles that emphasize a relational perspective, such as smoothing and compromising (Ting-Toomey, 1986). However, their actual behavior may not mirror such societal mores. Some researchers in conflict resolution eschew the implicit assumption that conflict can and/or should be resolved, arguing that in many cases, conflict ‘‘resolution’’ may only be a temporary solution, leaving deeper issues unresolved. Thus the idea of full agreement may not be realistic and in fact, in many circumstances, conflict resolution engagement may not produce long term successful results or generate cooperative work relationships (Rubin, 1993). 4.2. Conclusion In conclusion, within conflict resolution research, the five-style paradigm as created by Blake and Mouton (1964) and developed into self-report instruments by a number of researchers, appears to yield significant, if limited, results regarding differences among cultures, between genders, and among organizational roles, when tested via meta-analytic techniques. In a very practical sense, understanding cultural background and how this affects preferences regarding conflict resolution style may create greater understanding and less conflict in the workplace, as well as in communities at large. For example, a client of the first author who supervises nursing staff in a large, diverse hospital setting received several complaints from other staff members about a Chinese-born nurse. This nurse would react with disapproval and withdrawal when emergency situations necessitated the use of conflictual barked orders. A deeper grasp of cultural differences enabled the supervisor and all of the health workers to better understand each other’s behavior; this in turn allowed for greater respect and a more cohesive work environment. ARTICLE IN PRESS 186 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 Future research may wish to concern itself with the creation of a statistically rigorous conflict resolution instrument that will measure conflict behavior, rather than simply self-reported conflict style preferences. In addition, research contrasting respondents’ own interpretation of conflict style versus perception by others of their conflict style is important. Often, an individual’s perception of their style of resolving conflict may be far different from the receiver’s perception. For example, consider the following conversation within a peer mediator focus group: Peer Mediator 1:‘‘[We] were accused of not understanding her [an African American woman in mediation]. The other person was white, the rest of us were white, how could we understand her, [yet] part of how I saw her, my perception, was that she was a very powerful woman. And she was—even and including her body language. And she intimidated this white woman.’’ Peer Mediator 2:‘‘We gave her feedback that [her behavior] was aggressive and she told us, ‘No, in fact, I’m passive.’’’ Peer Mediator 1:‘‘In her culture she is [passive], but in our culture she is a very powerful woman.’’ (Holt, 2000, p. 64) Research specifically concerning conflict resolution styles may thus wish to include an emphasis on perceptions and actual behavior, as a means of updating the five style paradigm. Such research is vital, given its potential for creating greater understanding among cultures, between genders and within organizations. Acknowledgements The authors wish to thank Dr. Deniz Ones of the University of Minnesota Department of Psychology for her guidance on analyzing data and David Saez for help with initial coding of the studies. Appendix. Citation, number of groups contributed, and comparisons made for studies included in meta-analysis Nbr of groups Comparisons Citation 15 a,b,c,d Algert, N.E. (1998). Peer conflict:Simi larities and differences among three adolescent groups. (Doctoral dissertation, Texas A&M University, 1998/1999). Dissertation Abstracts International, 59(12), 4356A. 2 a Camp, O.H. (1984). The conflict management styles and selected characteristics of directors in a national network of volunteer staffed crisis intervention centers. (Doctoral dissertation, ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 187 University of Pittsburgh, 1984/1985). Dissertation Abstracts International, 45(11), 3405A. 3 a,c,i,l Campbell, G.M. (1993). Secondary school principals and conflict-handling styles. (Doctoral dissertation, University of Houston, 1993/1994). Dissertation Abstracts International, 54(8), 2813A. 9 e,f,g Cardona, F. (1995). A comparative study of the styles of handling interpersonal conflict among students, faculty, and administrators. (Doctoral dissertation, Michigan State University, 1995/1996). Dissertation Abstracts International, 57(2), 521A. 2 a Chanin, M.N., & Schneer, J.A. (1984). A study of the relationship between Jungian personality dimensions and conflict-handling behavior. Human Relations, 37(10), 863–879. 2 a Content, S.H. (1986). Conflict management styles of principals in elementary and secondary schools. (Doctoral dissertation, University of California, Los Angeles, 1986). Dissertation Abstracts International, 47(6), 1942A. 4 a Corcoran, K.O. (1997). Deficits in conflict style, attachments, social self-efficacy, and perspective taking of parenting skills training. (Doctoral dissertation, University of Oregon, 1997/1998). Dissertation Abstracts International, 58(9), 5188B. 2 a Danes, S.M., Leichtentritt, R.D., Metz, M.E., & Huddleston-Casas, C. (2000). Effects of conflict styles and conflict severity on quality of life of men and women in family business. Journal of Family and Economic Issues, 21(3), 259–286. 4 a,b,c,d,j,m Dena, Z.M. (1994). A comparison of conflict management styles of Hispanic and Anglo administrators in elementary and secondary schools. (Doctoral dissertation, University of La Verne, Pennsylvania, 1994/1995). Dissertation Abstracts International, 56(3), 773A. 2 b Elsayed-Ekhouly, S.M., & Buda, R. (1996). Organizational conflict:A comparative ARTICLE IN PRESS 188 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 analysis of conflict styles across cultures. The International Journal of Conflict Management, 7(1), 71–80. 2 a,j,m Fenlon, M.J. (1997). The impact of management practices, climate, and conflict resolution styles on individual adjustment and expressions of motivation. (Doctoral dissertation, Columbia University, 1997). Dissertation Abstracts International, 58(4), 2164B. 8 a,e,f,g,h,i,j,k,l,m Frederickson, J.D. (1997). Assessing the validity of the Rahim Organizational Conflict Inventory-II (ROCI-II). (Doctoral dissertation, University of Minnesota, 1997/ 1998). Dissertation Abstracts International, 58(7), 3961B. 2 b Gabrielidis, C., Stephan, W.G., Ybarra, O., Pearson, V.M.D.S., & Villareal, L. (1997). Preferred styles of conflict resolution:Mexico and the United States. Journal of Cross- Cultural Psychology, 28(6), 661–677. 2 a Johnson, A.K. (1997). Conflict-handling intentions and the MBTI:A construct validity study. Journal of Psychological Type, 43, 29–39. 2 a,c Kearns, W.P. (1980). The development of a marriage enrichment program on conflict management for recently married couples. (Doctoral dissertation, Eastern Baptist Theological Seminary, 1980). Dissertation Abstracts International, 41(3), 1087A. 2 b Kim, M.-S., & Kitani, K. (1998). Conflict management styles of Asian- and Caucasian- Americans in romantic relationships in Hawaii. Journal of Asian Pacific Communication, 8(1), 51–68. 3 e,f,g Lee, C.-W. (1990). Relative status of employees and styles of handling interpersonal conflict:An experimental study with Korean managers . The International Journal of Conflict Management, 1(4), 327–340. 1 a Mills, J. & Chusmir, L.H. (1988). Managerial conflict resolution styles:Wo rk and home differences. In E. Goldsmith (Ed.), Work and ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 189 family: Theory, research, and applications (special issue). Journal of Social Behavior and Personality, 3(4), 303–316. 2 b Miyazaki, C., Moroi, E., & Stephan, W.G. (in press). Preferences for styles of conflict resolution:A comparison of Japan and the US Manuscript submitted for publication. 2 a,c Nelson, D.T. & Lubin, B. (1991). Performance of state legislators on the Conflict MODE Instrument. Organization Development Journal, 9(1), 79–80. 2 b Oetzel, J. (1998). The effects of self-construals and ethnicity on self-reported conflict styles. Communication Reports 11(2), 133–144. 2 b Ozcelik, S. (2001, February). The relationship between culture and the conflict resolution Styles: A Survey Method at ICAR (Institute for Conflict Analysis and Resolution), Paper presented at the Old Dominion University Graduate Research Conference ‘‘Global Changes in the 21st Century,’’ Norfolk, VA. Available: http://www.geocities.com/ tatarkirim/culture1.html 4 g,h,j,k,m Persico, J. (1986). Levels of conflict, worker performance, individual conflict styles, type of work, organizational characteristics and the external environment of the organization. (Doctoral dissertation, University of Minnesota, 1986/1987). Dissertation Abstracts International, 47(12), 4371A. 2 a,c Pritchard, B. (1985). The relationship between managerial experience and conflict management styles of men and of women in community college administration. (Doctoral dissertation, Western Michigan University, Kalamazoo, 1985/1986). Dissertation Abstracts International, 46(9), 2510A. 2 a,c Rahim, M.A. (1992). Table C.4, Collegiate reference group norms of five styles of handling interpersonal conflict with superiors, subordinates, and peers. In:M.A. Rahim (Ed.), Managing conflict in organizations. 2nd ed. (p. 191) 6 a,c Renwick, P.A. (1977). The effects of sex differences on the perception and ARTICLE IN PRESS 190 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 management of superior-subordinate conflict:An exploratory study. Organizational Behavior and Human Performance 19, 403–415. 2 a Shockley-Zalabak, P.S. (1981). The effects of sex differences on the preference for utilization of conflict styles of managers in a work setting:An exploratory study. Public Personnel Management Journal, 10, 289–295. 4 a Shockley-Zalabak, P., & Morley, D.D. (1984). Sex differences in conflict style preferences. Communication Research Reports, 1(1), 28–32. 4 e,f,g Sorenson, R.L., Morse, E.A., & Savage, G.T. (1999). A test of the motivations underlying choice of conflict strategies in the dualconcern model. The International Journal of Conflict Management, 10(1), 25–44. 4 h,k Soto-Fulp, S. (1996). Conflict resolution styles:Eth nic and gender differences. (Doctoral dissertation, New Mexico State University, 1996/1997). Dissertation Abstracts International, 57(7), 4787B. 5 b Ting-Toomey, S., Gao, G., Tribusky, P., Yang, Z., Kim, H.S., Lin, S.-L., & Nishida, T. (1991). Culture, face maintenance and styles of handling interpersonal conflict:A study in five cultures. The International Journal of Conflict Management, 2(4), 27 4 b Ting-Toomey, S., Yee-Jung, K.K., Shapiro, R.B., Garcia, W., Wright, T.J., & Oetzel, J.G. (2000). Ethnic/cultural identity salience and conflict styles in four US ethnic groups. International Journal of Intercultural Relations, 24, 47–81. 2 b Trubisky, P., Ting-Toomey, S. & Lin, S.-L. (1991). Influence of individualismcollectivism and self-monitoring on conflict styles. International Journal of Intercultural Relations, 15, 65–84 4 f Utley, M.E., Richardson, D.R., & Pilkington, C.J. (1989). Personality and interpersonal conflict management. Personality and Individual Differences, 10(3), 287–293. ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 191 2 f van de Vliert, E., Euwema, M. C., & Huismans, S. E. (1995). Managing conflict with a subordinate or a superior: Effectiveness of conglomerated behavior. Journal of Applied Psychology, 80(2), 271–281. 3 a,c Zammuto, R. F., London, M. & Rowland, K. M. (1979). Effects of sex on commitment and conflict resolution. Journal of Applied Psychology, 80(2), 250–252. Note. Comparisons:a: Gender (Female–Male); b:Culture (Individualistic–Collectivistic); c:Gender within Individualistic Culture (Female–Male); d:Gender within Collectivistic Culture (Female–Male); e: Organizational Level (Peers–Subordinates); f:Organizat ional Level (Subordinates–Superiors); g: Organizational Level (Superiors–Peers); h:Gender Peers; i:Gender Subordinates; j:Gender Superiors; k:Gender Peers Individualistic; l:Gender Subordinates Individualistic; m:Gender Superiors Individualistic. References Agee, M. L., & Kabasakal, H. E. (1993). Exploring conflict resolution styles:A study of Turkish and American university business students. International Journal of Social Economics, 20, 3–14. Algert, N. E. (1998). Peer conflict:Similar ities and differences among three adolescent groups. (Doctoral dissertation, Texas A&M University, 1998/1999). Dissertation Abstracts International, 59(12), 4356A. Blake, R. R., & Mouton, J. S. (1964). The managerial grid. Houston, TX:Gulf Publishing. Blake, R. R., & Mouton, J. S. (1970). The fifth achievement. Journal of Applied Behavioral Science, 6, 413–426. Blumberg, H. H. (1998). Peace psychology after the cold war:A selective review. Genetic, Social, and General Psychology Monographs, 124(1), 5–37. Cardona, F. (1995). A comparative study of the styles of handling interpersonal conflict among students, faculty, and administrators. (Doctoral dissertation, Michigan State University, 1995/1996). Dissertation Abstracts International, 57(2), 521A. Chua, E., & Gudykunst, W. B. (1987). Conflict resolution style in low- and high-context cultures. Communication Research Reports, 4(1), 32–37. Conrad, C. (1985). Strategic organizational communication: Cultures, situations, and adaptation. New York:CBS College Publishing. Content, S. H. (1986). Conflict management styles of principals in elementary and secondary schools. (Doctoral dissertation, University of California, Los Angeles, 1986). Dissertation Abstracts International, 47(6), 1942A. Cosier, R. A., & Ruble, T. L. (1981). Research on conflict-handling behavior:An experimental approach. Academy of Management Journal, 24, 816–831. Danes, S. M., Leichtentritt, R. D., Metz, M. E., & Huddleston-Casas, C. (2000). Effects of conflict styles and conflict severity on quality of life of men and women in family business. Journal of Family and Economic Issues, 21(3), 259–286. Dena, Z. M. (1994). A comparison of conflict management styles of Hispanic and Anglo administrators in elementary and secondary schools. (Doctoral dissertation, University of La Verne, Pennsylvania, 1994/ 1995). Dissertation Abstracts International, 56(3), 773A. Drake, B. H., Zammuto, R. F., & Parasuraman, S. (1981). Five conflict styles? Factor structure and social desirability issues underlying the Conflict MODE Instrument. Paper presented at the ARTICLE IN PRESS 192 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 13th Annual Meeting of the American Institute for Decision Sciences, Boston, MS, November, pp. 406–408. Dubinskas, F. A. (1992). Culture and conflict:The cultural roots of discord. In D. M. Kolb, & J. M. Bartunke (Eds.), Hidden conflict in organizations: Uncovering behind-the-scenes disputes (pp. 187–208). London:Sage. Duryea, M. L. (1992). Conflict and culture: A literature review and bibliography. Victoria, BC, Canada: University of Victoria Institute for Dispute Resolution. D’Silva, M. U., & Whyte, L. O. (1998). Cultural differences in conflict styles:Vietnam ese refugees and established residents. Communication Abstracts, 21(3). Elangovan, A. R. (1998). Managerial intervention in organizational disputes:Testing a prescriptive model of strategy selection. The International Journal of Conflict Management, 9(4), 301–335. Elsayed-Ekhouly, S. M., & Buda, R. (1996). Organizational conflict:A comparative analysis of conflict styles across cultures. The International Journal of Conflict Management, 7(1), 71–81. Firebaugh, G., & Davis, K. (1988). Trends in anti-black prejudice, 1972–1984:Region and cohort effects. American Journal of Sociology, 94, 251–272. Fisher, R., & Ury, W. (1981). Getting to yes. New York:Penguin. Gabrielidis, C. A., Stephan, W. G., Ybarra, O., Pearson, V. M. D. S., & Villareal, L. (1997). Preferred styles of conflict resolution:Mexico and the United States. Journal of Cross-Cultural Psychology, 28(6), 661–677. Gilligan, C. (1977). In a different voice:Women’s conceptions of self and of morality. Harvard Educational Review, 47(4), 481–517. Goodall, H. L., Jr. (1996). Divine signs: Connecting spirit to community. Carbondale and Edwardsville, IL: Southern Illinois University Press. Gudykunst, W. B. (1998). Bridging differences: Effective intergroup communication (3rd ed.). Thousand Oaks, CA:Sage. Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K. S., & Heyman, S. (1996). The influence of cultural individualism-collectivism, self construals, and individual values on communication styles across cultures. Human Communication Research, 22, 510–543. Hall, J. (1969). Conflict management survey: A survey of one’s characteristic reaction to and handling conflict between himself and others. Canoe, TX:Teleometrics International. Harris, W. H. (1988). Self-actualization levels and conflict management styles of middle and junior high school principals in Connecticut. (Doctoral dissertation, University of Bridgeport, Connecticut, 1988/ 1989). Dissertation Abstracts International, 49(10), 2874A. Haslam, S. A., Turner, J. C., Oakes, P. J., McGarty, C., & Hayes, B. K. (1992). Context-dependent variation in social stereotyping:I. The effects of intergroup relations as mediated by social change and frame of reference. European Journal of Social Psychology, 22(1), 3–20. Haslam, S. A., Turner, J. C., Oakes, P. J., Reynolds, K. J., Eggins, R. A., Nolan, M., & Tweedie, J. (1998). When do stereotypes become really consensual? Investigating the group-based dynamics of the consensualization process. European Journal of Social Psychology, 28, 755–776. Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. Orlando, FL:Academic Press. Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Beverly Hills: Sage. Hofstede, G. (1983). National cultures in four dimensions:A research-based theory of cultural differences among nations. International Studies of Management and Organization, XIII(1–2), 46–74. Holt, J. L. (2000). Impact of gender, organizational role, and multicultural status on conflict resolution style preference. (Doctoral dissertation, University of Minnesota, 2000/2001). Dissertation Abstracts International, 61(8), 4457B. Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in research findings. Newbury Park, CA:Sage. Kabanoff, B. (1987). Predictive validity of the MODE Conflict Instrument. Journal of Applied Psychology, 72(1), 160–163. Kagan, S., Knight, G., & Martinez-Romero, S. (1982). Culture and the development of conflict resolution style. Journal of Cross-Cultural Psychology, 13, 43–59. ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 193 Kahn, R. L., Wolfe, D. M., Quinn, P. R., Snoek, J. D., & Rosenthal, R. A. (1964). Organizational stress: Studies in role conflict and ambiguity. New York:Wiley. Kilmann, R. H., & Thomas, K. W. (1978). Four perspectives on conflict management:An attributional framework for organizing descriptive and normative theory. Academy of Management Review, 3, 59–68. Kim, M.-S., & Kitani, K. (1998). Conflict management styles of Asian- and Caucasian-Americans in romantic relationships in Hawaii. Journal of Asian Pacific Communication, 8(1), 51–68. Kliman, J., & Llerena-Quinn, R. (2002). Dehumanizing and rehumanizing responses to September 11 (Abstract). Journal of Systemic Therapies, 21(3), 8–18. Knapp, M. L., Putnam, L. L., & Davis, L. J. (1988). Measuring interpersonal conflict in organizations: Where do we go from here? Management Communication Quarterly, 1(3), 414–429. Kolb, D. M. (1993). Her place at the table:Gender and negotiation. In L. Hall (Ed.), Negotiation: Strategies for mutual gain (pp. 138–150). London:Sage. Kozan, M. K. (1990). Relationships of hierarchy and topics of conflict management styles:A comparative study. In M. A. Rahim (Ed.), Theory and research in conflict management (pp. 174–187). New York: Praeger Publishers. Lagao, M. L. (1996). Cultural differences of managers’ conflict resolution styles. (Master’s thesis, California State University, Long Beach, California, 1996/1997), Masters Abstracts International, 35(3), 903. Lee, C. W. (1990). Relative status of employees and styles of handling interpersonal conflict:An experimental study with Korean managers. The International Journal of Conflict Management, 1(4), 327–340. Lee, H. O., & Rogan, R. G. (1991). A cross-cultural comparison of organizational conflict management behaviors. The International Journal of Conflict Management, 2(3), 181–199. Lewicki, R. J., Weiss, S. E., & Lewin, D. (1992). Models of conflict, negotiation and third party intervention:A review and synthesis. Journal of Organizational Behavior, 13, 209–252. Mills, J., & Chusmir, L. H. (1988). Managerial conflict resolution styles:Work and home differences. In: Goldsmith, E. (Ed.), Work and family: Theory, research, and applications (special issue). Journal of Social Behavior and Personality, 3(4), 303–316. Miyazaki, C., Moroi, E., & Stephan, W.G. (in press). Preferences for styles of conflict resolution:A comparison of Japan and the US Manuscript submitted for publication. Muir, B. O. (1991). Managerial conflict resolution as a function of role status and gender. (Master’s thesis, California State University, Long Beach, California, 1991/1992). Masters Abstracts International 30(1), 149. Musser, S. J. (1982). A model for predicting the choice of conflict management strategies by subordinates in high-stakes conflicts. Organizational Behavior and Human Performance, 29, 257–269. Nelson, D. T., & Lubin, B. (1991). Performance of state legislators on the conflict MODE instrument. Organization Development Journal, 9(1), 79–80. Nicotera, A. M. (1996). An assessment of the argumentativeness scale for social desirability bias. Communication Reports, 9(1), 23–35. Oetzel, J. G. (1998). The effects of self-construals and ethnicity on self-reported conflict styles. Communication Reports, 11(2), 133–144. Oetzel, J. G. (1999). The influence of situational features on perceived conflict styles and self-construals in work groups. International Journal of Intercultural Relations, 23(4), 679–695. Paulson, A. D. (1986). Organizational space as a determinant of conflict resolution modes used by middle managers. (Master’s thesis, University of California, Los Angeles, 1986). Masters Abstracts International, 24(4), 423. Pearson, V. M., & Stephan, W. G. (1998). Preferences for styles of negotiation:A comparison of Brazil and the US. International Journal of Intercultural Relations, 22(1), 67–83. Phillips, E., & Cheston, R. (1979). Conflict resolution:What works? California Management Review, 21(4), 76–83. Putnam, L. L., & Wilson, C. E. (1982). Communicative strategies in organizational conflicts:Reliabil ity and validity of a measurement scale. In M. Burgoon, & N. E. Doran (Eds.), Communication yearbook, vol. 6 (pp. 629–652). Beverly Hills:Sage. ARTICLE IN PRESS 194 J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 Rahim, M. A. (1983). Rahim organizational conflict inventories: Professional manual. New York: Consulting Psychologists Press. Rahim, M. A. (1986). Referent role and styles of handling interpersonal conflict. The Journal of Social Psychology, 126(1), 79–86. Rahim, M. A. (1992). Managing conflict in organizations (2nd ed.). Westport, CT:Praeger. Rahim, M. A., & Blum, A. A. (Eds.). (1994). Global perspectives on organizational conflict. Westport, CT: Praeger. Rahim, M. A., & Buntzman, G. F. (1989). Supervisory power bases, styles of handling conflict with subordinates, and subordinate compliance and satisfaction. The Journal of Psychology, 123(2), 195–210. Renwick, P. A. (1975). Perception and management of superior-subordinate conflict. Organizational Behavior and Human Performance, 13, 444–456. Renwick, P. A. (1977). The effects of sex differences on the perception and management of superiorsubordinate conflict:An exploratory study. Organizational Behavior and Human Performance, 19, 403–415. Rubin, J. Z. (1993). Some wise and mistaken assumptions about conflict and negotiation. In J. W. Breslin, & J. Z. Runbin (Eds.), Negotiation theory and practice, (2nd ed.) (pp. 3–12). Cambridge, MA:Program on Negotiation at Harvard Law School. Schuman, E. (2002). Combatting terrorism:An immodest proposal (Abstract). Psychology & Education: An Interdisciplinary Journal, 39(2), 43–45. Shockley-Zalabak, P. (1981). The effects of sex differences on the preference for utilization of conflict styles of managers in a work setting:An exploratory study. Public Personnel Management Journal, 10, 289–295. Smith, L., & Haar, J. (1990). An assessment of conflict-handling and personality characteristics of project management personnel in the People’s Republic of China. Journal of Social Behavior and Personality, 5(3), 61–76. Sorenson, R. L., Morse, E. A., & Savage, G. T. (1999). A test of the motivations underlying choice of conflict strategies in the dual-concern model. The International Journal of Conflict Management, 10(1), 25–44. Soto-Fulp, S. (1996). Conflict resolution styles:Ethnic and gender differences. (Doctoral dissertation, New Mexico State University, 1996/1997). Dissertation Abstracts International, 57(7), 4787B. Sternberg, R. J., & Dobson, D. M. (1987). Resolving interpersonal conflicts:An analysis of stylistic consistency. Journal of Personality and Social Psychology, 52(4), 794–812. Sternberg, R. J., & Soriano, L. J. (1984). Styles of conflict resolution. Journal of Personality and Social Psychology, 47(1), 115–126. Stockard, J., & Lach, D. (1989). Conflict resolution:Sex and gender roles. In J. B. Gittler (Ed.), The annual review of conflict knowledge and conflict resolution, Vol. 1 (pp. 69–99). New York:Garland Publishing Inc. Sweeney, B., & Carruthers, W. L. (1996). Conflict resolution:History, philosophy, theory, and educational applications. School Counselor, 43(5), 326–344. Thomas, K. W. (1976). Conflict and conflict management. In M. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp. 889–935). Chicago:Rand-Mc Nally. Thomas, K. W. (1988). The conflict-handling MODES:Toward more precise theory. Management Communication Quarterly, 1, 430–437. Thomas, K. W., & Kilmann, R. H. (1974). Thomas-Kilmann conflict mode instrument. Tuxedo, NY: Xicom, Inc. Thomas, K. W., & Kilmann, R. H. (1975). The social desirability variable in organizational research. An alternative explanation for reported findings. Academy of Management Journal, 18, 741–752. Ting-Toomey, S. (1986). Conflict communication styles in black and white subjective cultures. In Y. Y. Kim (Ed.), Interethnic Communication: Current research (pp. 75–88). Beverly Hills, CA:Sage. Ting-Toomey, S. (1988). Intercultural conflict styles:A face negotiation theory. In Y. Y. Kim, & W. B. Gudykunst (Eds.), Theories in intercultural communication (pp. 213–235). Newbury Park, CA:Sage. ARTICLE IN PRESS J.L. Holt, C.J. DeVore / International Journal of Intercultural Relations 29 (2005) 165–196 195 Ting-Toomey, S., Gao, G., Trubisky, P., Yang, Z., Kim, H. S., Lin, S.-L., & Nishida, T. (1991). Culture, face maintenance, and styles of handling interpersonal conflict:A study in five cultures. The International Journal of Conflict Management, 2(4), 275–296. Ting-Toomey, S., Yee-Jung, K. K., Shapiro, R. B., Garcia, W., Wright, T. J., & Oetzel, J. G. (2000). Ethnic/cultural identity salience and conflict styles in four US ethnic groups. International Journal of Intercultural Relations, 24, 47–81. Trubisky, P., Ting-Toomey, S., & Lin, S.-L. (1991). The influence of individualism-collectivism and selfmonitoring on conflict styles. International Journal of Intercultural Relations, 15, 65–84. van de Vliert, E., Euwema, M. C., & Huismans, S. E. (1995). Managing conflict with a subordinate or a superior:Effectiveness of conglomerated behavior. Journal of Applied Psychology, 80, 271–281. van de Vliert, E., & Hordijk, J. W. (1989). A theoretical position of compromising among other styles of conflict management. The Journal of Social Psychology, 129(5), 681–690. Volkema, R. J., & Bergmann, T. J. (1995). Conflict styles as indicators of behavioral patterns in interpersonal conflicts. The Journal of Social Psychology, 135(1), 5–15. Yinger, J. M. (1994). Ethnicity: Source of strength? Source of conflict?. Albany:State University of New York Press. Zammuto, R. F., London, M., & Rowland, K. M. (1979). Effects of sex on commitment and conflict resolution. Journal of Applied Psychology, 64(2), 227–231. Further reading Antonioni, D. (1998). Relationship between the big five personality factors and conflict management styles. The International Journal of Conflict Management, 9(4), 336–355. Brubaker, D., & Verdonk, T. (1999). Conflict transformation training in another culture:A case study from Angola. Mediation Quarterly, 16(3), 303–314. Gudykunst, W. B., & Ting-Toomey, S. (1988). Culture and interpersonal communication. Newbury Park, CA:Sage. Haslam, S. A., McGarty, C., Oakes, P. J., & Turner, J. C. (1992). Social comparative context and illusory correlation:Testing between in-group bias and social identity models of stereotype formation. Australian Journal of Social Psychology, 41, 97–102. Richardson, J. (1995). Avoidance as an active mode of conflict resolution. Team Performance Management: An International Journal, 1(4), 19–25. Ting-