,flfi \ / 90¢> 31/?(2t LIBRARIES MICHIGAN STATE UNIVERSITY EAST LANSING, MICI-I 48824-1048 This is to certify that the thesis entitled EXPLORATION OF THE RELATIONSHIP BETWEEN FRIENDSHIP AT WORK AND J OB SATISFACTION: AN APPLICATION OF BALANCE THEORY presented by Hye Eun Lee has been‘aocepted towards fulfillment of the requirements for the Master of Arts degree in Communication Major Professor’s Signature Date MSU is an Affirmative Action/Equal Opportunity Institution 'V—hfifi PLACE IN RETURN Box to remote this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE DEC I 0;; 20520689 _ 2/05 cJCIRC/DatoDuo.indd-p.15 EXPLORATION OF THE RELATIONSHIP BETWEEN FRIENDSHIP AT WORK AND JOB SATISFACTION: AN APPLICATION OF BALANCE THEORY By Hye Eun Lee A THESIS Submitted to Michigan State University In partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 2005 ABSTRACT EXPLORATION OF THE RELATIONSHIP BETWEEN FRIENDSHIP AT WORK AND JOB SATISFACTION: AN APPLICATION OF BALANCE THEORY By Hye Eun Lee The current study examined the relationship between workplace friendships and job satisfaction. Based on Balance Theory, it was predicted that if employees had more similar perceptions on organizational climate with their workplace friend(s), they would be more satisfied with their jobs. Also it was tested if similarity with one best friend was more strongly related to an employee’s job satisfaction than similarity with a group of friends. Eighty-one employees from two organizations completed questionnaires. The data were not consistent with the hypothesis predicting positive relationship between similarity in climate with fi’iends and job satisfaction. The finding was that the similarity in responsibility (i.e., one dimension of climate) with workplace friends was negatively related to employee’s job satisfaction. The data also showed that the similarity in ‘ responsibility with a group of friends was significantly related to employee’s job satisfaction, while the similarity in responsibility with one best fi'iend was not. Finally, implications and limitations of these findings were discussed. ACKNOWLEDGMENTS I wish to thank my adviser, Professor Hee Sun Park, for the guidance and friendship she has provided during my graduate student life in Michigan State University. I would never have made it this far without her helpfiIl insights and suggestions. Secondly I wish to thank my committee members, Professor Bradley Greenberg and Professor Timothy Levine for all the expertise they have contributed to my thesis. Finally I would like to express my deepest gratitude to my husband, Hyunjin, whose love and support made this possible. TABLE OF CONTENTS LIST OF TABLES .............................................................................. v INTRODUCTION .............................................................................. 1 CHAPTER 1 LITERATURE REVIEW ...................................................................... 3 Workplace Friends ........................................................................ 3 Balance Theory ........................................................................... 7 Workplace Friendships and Job Satisfaction ......................................... ll Hypothesis and Research Question .................................................... 18 CHAPTER 2 METHOD ........................................................................................ 24 Sample ..................................................................................... 24 Procedure .................................................................................. 25 Measures .................................................................................. 26 Operational Definitions .................................................................. 29 Procedure to Analyze .................................................................... 32 CHAPTER 3 RESULTS ....................................................................................... 33 Overview ................................................................... p ............... 33 Effects of Climate Similarities with Best Friend(s) on Job Satisfaction .......... 33 Effects of Climate Similarities with F riend(s) on Job Satisfaction ................ 34 Effects of Climate Similarities with Best Friend or with Friends on Job Satisfaction ................................................................................ 35 CHAPTER 4 DISCUSSION .................................................................................. 37 Implication for the Relation between Similarities in Climate with Friends and Job Satisfaction ........................................................................... 37 Implication for Similarities ............................................................. 40 Implication for Climate .................................................................. 41 Limitations and Directions for Future Research ...................................... 42 Conclusion ................................................................................ 43 FOOTNOTES .................................................................................. 44 APPENDICES ................................................................................. 46 REFERENCES .................................................................................. 60 LIST OF TABLES Table l. The Effects of Climate Similarities with the Best F riend(s) on Job Satisfaction ..................................................................................... 68 Table 2. The Effects of Climate Similarities with the Friends on Job Satisfaction ..................................................................................... 69 Table 3. The Effects of Climate Similarities with the Weighted Friendship on Job Satisfaction ..................................................................................... 70 INTRODUCTION Human beings cannot live alone. PeOple form various relationships with other people, and such relationships have various impacts on people’s attitudes and behaviors (Rogers & Kincaid, 1980). As one of many interpersonal relationship types, friendship is important because it afiects people’s cognitive or emotional states (Fleming & Baum, 1986; Solano, 1986). Furthermore, having balance in one’s relationship can be important because a balanced relationship encourages an individual’s stable and harmonious internal state (Heider, 1958). Considering that most people have one or more close fiiends and believe that their close friends are crucial in their lives, investigating the various aspects of friendship can provide insights into human social behaviors and attitudes. Most people spend a considerable portion of their lives at work. Consequently people often form friendships with their coworkers. Such friendship at work is called blended fiiendship, since friends in the workplace function simultaneously with both personal and role components (Bridge & Baxter, 1992). The blended friendship is associated with crucial work-related issues such as job involvement, job satisfaction, and organizational commitment (Riordan & Griffeth, 1995). When examining the relation between blended friendship and workplace issues, it may be necessary to consider whether one best fiiend or a group of close fi'iends have a bigger impact on an individual’s job satisfaction. Research has focused on one best friend’s influence rather than the cumulative influence of all friends at work (e.g., Boyd & Taylor, 1998; Lincoln & Miller, 1979; Wright, 1969). However, a group of friends as a clique also has the potential for influencing an individual worker’s decisions and actions (Rogers & Kincaid, 1980). Therefore, the goal of the current study is to compare the effect of the closest friend with the effect of a group of friends on an individual’s job satisfaction. Toward this goal, this study will first define fiiendship, blended friendship, and friendship network in the workplace. Second, a brief overview of the basic elements and assumptions of Balance Theory and its various empirical applications will be presented. Third, it will be discussed how this theory can be applied in understanding the relationship between workplace friendship and people’s job satisfaction. Finally, a rationale will be provided for a hypothesis about the relation between friends’ similar perceptions and their job satisfaction. A brief rationale will be offered for a research question about the importance of all friends or one best friend in terms of an individual ’5 job satisfaction. CHAPTER 1 LITERATURE REVIEW Workplace Friends Definitions and characteristics of friendship. Friendship has been defined as “voluntary interdependence between two persons over time that is intended to facilitate social-emotional goals of the participants, and may involve varying types and degrees of companionship, intimacy, affection and mutual assistance” (Hays, 1988, p.395). Based on this definition, friendship commonly includes four components: interdependence, the continuity of interaction over an extended time, voluntariness, and social-emotional goals. That is, friends allow themselves to be mutually influenced by each others’ behaviors, and the relationships endure over time for their Social and emotional needs. Friendships provide three values for a person: stimulation value, utility value, and ego support value (Wright, 1969). Stimulation value refers to the degree to which an individual sees another person as attractive, imaginative, and competent to introduce the individual to new ideas and activities for expanded and elaborated knowledge and outlook. Utility value refers to the extent to which the individual regards another as a cooperative and helpful person for the individual ’s own goals and needs. Finally, ego support value indicates the degree to which the subject considers another as encouraging, supportive for the individual ’s positive self-impression. These values provide difi‘erent types of rewards in a fiiendship. For example, if Jane forms a friendship with Ryan for stimulation value, Jane may acquire a broader worldview through interaction with Ryan. If Jane has a fiiend, John, with strong utility value, Jane may use John’s time and resources for her personal needs. If Jane has Mark as a friend of good ego support value, Mark cheers Jane up and makes Jane feel greater self-worth. People consciously and unconsciously evaluate their friendships with others for the values their friends provide for them (Wright, 1969). The more values their friends provide for them, the more satisfied people are with their friends (Wright, 1969). Because of these values of friendship, people without fiiends experience loneliness and can be emotionally and socially devastated (Solano, 1986). However, people with fiiends also face some negative efl‘ects from time to time. People become stressed if their friends’ social support does not meet their expectations (Fleming & Baum, 1986). Also, people may experience some tension and discomfort due to misunderstandings or incongruence of values and goals with friends (Winstead, Derlega, Montgomery, & Pilkington, 1995). In the worse situations, the tension leads people to break off the fiiendship. Compared to other types of personal relationships (e.g., rOmantic relationships), the boundary of friendship is less clear. In other words, under a normal situation, it is easier to determine whether a person is a romantic partner or not than whether a person is a friend or not. Friendship can be better understood by examining differences between friendship and other relationships. For example, compared to romantic relationships, fiiendship is less exclusive, less intense in emotional expression, less regulated by social norms, and requires less commitment (Wright, 1987). In addition, friendship is more stable and more forgiving even when friends fail to meet relational expectations (Davis, 1985). On the other hand, love in romantic relationships connotes greater fascination, greater sexual desire, more demands and more willingness to give the utmost (Davis, 1985). Because of these differences, it is more common and normal to have multiple fiiends than multiple romantic partners at one time. Various factors influence friendship formation. The greater amount of contacts people have, the more likely they will become friends (F estinger, Schachter, & Back, 1950). The more similar their attitudes are, the more likely people find each other socially attractive to interact with (Byme & Clore, 1970).Friends share more similar attitudes than strangers (Park & Boldman, 1998). Besides attitudes, similarities in age, gender, education, preferred activities, and personalities can be positive factors for friendship formation and maintenance (e.g., Johnson, 1989; Werner & Parmelee, 1979). Blended friendship. Workplaces can be breeding grounds for friendship. It is very likely that a worker forms friendships with other workers with whom he(she) fi'equently interacts and share similarities in attitudes, age, education, and so on. When friendship is formed in a workplace, however, such friendship takes the form of a blended fiiendship (Bridge & Baxter, 1992). That is, while friendship is usually conceptualized as a personal relationship, relationships in the workplace are characterized mostly as role-based. In other words, fiiendship in the workplace has another dimension of a role relationship in addition to being a personal relationship. A role relationship is characterized by formalized tasks and jobs for organizational goals (Bridge & Baxter, 1992). So if two employees at work accomplish a task together, the relationship is defined as a role relationship. A relationship between a superior and a subordinate is also a role relationship because a superior is expected to perform an organizational supervisory role such as assigning tasks to subordinates, evaluating subordinates’ performance, and providing feedback to subordinates. In contrast, a personal relationship is illustrated by companionship, intimacy, affection and mutual assistance for social-emotional goals (Hays, 1988). If two employees share non-work-related activities outside of normal working hours,the relationship is characterized as a personal relationship. Thus, having a blended friend is to have a coworker as a fiiend. Blended friendship has implications for organizational as well as individual outcomes. Employees who have friendly relationships with each other can work together more productively (Duck, 1983). The quality of friendships at work is positively related to job satisfaction (Markiewicz, Devine, & Kausilas, 2000; Winstead et al., 1995). Furthermore, attitudes and behaviors of employees’ blended friends positively influence employees’ job satisfaction (Salancik & Pfeffer, 1978) and organizational commitment (Krackhardt & Porter, 1985). Blended friendships, however, can also be a source of tension, especially when a friend-role and a work-role clash (Bridge & Baxter, 1992). A network of blended friends forms an informal communication network within an organization. Communication networks are defined as “the patterns of contact that are created by the flow of messages among communicators through time and space” (Monge & Contractor, 2003, p.3). As friendships are based on understanding, shared experiences, similar values, and mutual trust, friendship networks enable employees to acquire information that they may not be able to obtain from the formal communication network (Graber, 2003). Moreover, Lincoln and Miller (1979) report that “friendship networks in organizations are not merely sets of liked friends. They are systems for making decisions, mobilizing resources, concealing or transforming information, and performing other functions closely allied with work behavior and interaction” (p.196). Balance Theory Balance Theory can provide an explanation of how a network of workplace fiiendships is related to an individual worker’s job satisfaction. Balance Theory posits relations among three types of elements (Heider, 195 8). The three elements include a focal person (P), another person (0), and an event (X) perceived by the two people. Balance Theory focuses on P’s three kinds of interpretations: one interpretation about his (her) relationship with 0, another one about his (her) perception of X, and the last one about 0’8 perception of X. Although Balance Theory involves relations among the three elements, the relations between only two of the elements can be considered at a time. In other words, a dyadic relationship can exist between P and 0 (or between P and X), when P considers only one relation with 0 (or X). On the other hand, a triadic relationship exists among P, 0, and X, when P considers three relations between P and 0, 0 and X, and P and X simultaneously. Some more specific relationships among these three elements can be characterized by sentiment and unit formation (Heider, 1958). A sentiment refers to the way P feels or evaluates 0 or X. Although a sentiment can take various types and forms, Heider originally classified two types of sentiments: “liking” and “disliking.” Later some researchers broadened the boundary of the sentiment to more complicated types of ‘ evaluations such as “approving/disapproving” and “agreeing/disagreeing” (e.g., Curry & Emerson, 1970; Insko, 1981). Unit formation occurs when P perceives himself (herself) to belong together with 0, when P perceives himself (herself) to belong together with X, or when P perceives 0 to belong with X. There are a number of factors for unit formation. For instance, if two people share similarity, proximity, or interaction, they may form a unit. Or if a person owns an entity, a unit is made up of the person and the entity. Although Heider (1958) suggested that unit relations can roughly be divided into two types of “belongs” and “does not belong” like the sentiment relations, Insko (1981) pointed out that “does not belong” is not a negative relation against “belongs.” For example, it is clear that “P likes X’ is a negative relation with “P dislikes X” in the sentiment relation. In terms of the unit relation, however, it is unclear which one is a negative relation against “P is married with 0” between “P is divorced from 0” and “P is not married with 0.” This is partly the reason why most research after Heider explored only positive unit formation while both positive and negative sentiments have been topics together (e.g., Eagly & Chaiken, 1993; Insko, 1981; Petty & Cacioppo, 1981). Balance Theory explains P ’s cognition of a balanced state or an imbalanced state with the relationships of these sentiment and unit formation. If P and 0 make a unit and have a similar attitude toward X of liking or disliking, or if P (or 0) owns X and P and 0 have a similar interpretation about X, P has a balanced state. Otherwise, P’s cognitive state is imbalanced. Namely, a balanced state refers to a stable situation because a perceived unit and sentiments coexist without any stress in P’s cognitive organization. One main assumption of Balance Theory is that human beings prefer harmonious states over imbalance states (Heider, 1958). As balance increases, the person’s pleasure increases. So people try to increase balance, but decrease imbalance. In addition, even when a person gets the balance state, generally positive sentiments create more pleasantness than negative sentiments because similarity in liking between P and 0 toward X creates attraction effects along with balance effects (Jordan, 1953; Zajonc, 1968). If people experience imbalanced states, they try to resolve these disharrnonious states. For example, P may deny the unit or change his (her) attitudes toward 0(or X). From time to time, however, P does not resolve the disharmony, even when he (she) is aware of the conflicting situation. In this case, P experiences tension and stress to change, and the greater imbalance a person faces, the more stressful and uncomfortable the person is (Insko, 1981). Heider (1 95 8) also assumed that people have reciprocity in their sentiments. That is to say, if P likes 0, P automatically assumes that 0 likes P. A tendency toward reciprocated liking is considered a very common assumption. However when P dislikes 0, P is unlikely to assume that 0 dislikes P. A tendency toward reciprocated disliking is somewhat problematic (Insko, 1981). Thus, the assumption of reciprocity is usually applied to positive sentiments. Although Balance Theory was originally formulated to explain individuals’ psychology related to the context of interpersonal relations (Heider, 195 8), the theory has been broadly used for explaining and predicting attitudes, persuasion, management, social networks and so on. For example, Curry and Emerson (1970) found that people tended to perceive another person’s attraction toward the third person as similar to their own attraction to the third person. Aronson and Cope (1968) supported Balance Theory with the finding that people like their fiiend’s friend and their enemy’s enemy, and dislike their fiiend’s enemy and their enemy’s friend. Also Woodside and Chebat (2001) argued i that consumers’ behavior could be explained by Balance Theory. That is, purchasing behavior of consumers can be explained by a balanced triadic relationship among a consumer, a quality of product, and a producer. Finally, social scientists have attempted to expand the application of Balance Theory to the study of social networks (e. g. Markiewicz, Devine, & Klausilas, 2000). Because the Balance Theory focuses on individual ’s strong tendency for the cognitive consistency based on his (her) subjective interpretation, the theory is considered as one type of consistency theories (Eagly & Chaiken, 1993; Insko, 1981; Monge & Contractor, 2001 , 2003; Petty & Cacioppo, 1981). Symmetry Theory, Congruity Theory and Cognitive Dissonance Theory as well as the Balance Theory are well known as the major consistency theories (Petty & Cacioppo, 1981). These four theories are similar in that individual ’s cognitive consistency is determined by his (her) own subjective interpretation regardless of objective state of affairs, although the Balance Theory, the Symmetry Theory, the Congruity Theory, and the Cognitive Dissonance Theory call the consistency differently such as balance, symmetry, congruity, and consonance, respectively. In addition, these theories commonly assume that imbalance leads people to be motivated to restore balance and to get rid of cognitive tension. These similarities aside, four theories have their unique parts. The Balance Theory (Heider, 1958) focuses on interpersonal relations between two people (P and 0). The Balance Theory considers mainly how these two people (P and 0)’s attitudes toward X influence the P’s cognitive balance and how P reacts toward cognitive imbalance. Newcomb’s Symmetry theory (N ewcomb, 1953) centers on interpersonal attraction rather than P’s attitude change. The Balance Theory does not consider the extent of how much liking (disliking) exists or how strong the unit is, while the Congruity Theory (Osgood & Tannenbaum, 1955) provides degrees of liking or belongingness between two elements, so produces quantitative predictions about the effects of incongruity. Finally, the Cognitive 10 Dissonance Theory (F estinger, 195 7) emphasizes more on how perople reduce cognitive dissonance whereas the Balance Theory and the Congruity Theory underscore what makes balance states. Four theories have been actively applied to various fields similarly. Especially F estinger ’s cognitive dissonance theory has been widely adopted and productive (Eagly & Chaiken, 1993; O’Keefe, 2002). Since Balance Theory was originally formulated to explain individuals’ psychology related to the context of interpersonal relations (Heider, 1958), however the current study used the Balance Theory. Worlmlace Friendships and Job Satisfaction Balance Theory can be applied to understanding the relationship between workplace friendships and job satisfaction. As stated before, when P, as an employee, interacts with 0, their workplace friend, disagreements on various issues between the two parties (P and 0) can occur. Such disagreements can lead to P’s cognitive imbalance. As a consequence, P will try to resolve these imbalanced situations because people prefer a balanced state to an imbalanced state (Heider, 1958). People can show four reactions to imbalanced states. First, P may break off the relationship with the fiiend. It means that P denies the unit formation. Then the friend will not be a significant element to P anymore. The second way is that P changes his (her) attitude toward X to be compatible with 0’s attitude toward X. Like the first one, it makes p to eliminate the cause of imbalance. The third way is that P persuades 0 to change 0’s attitude toward X to be compatible with P’s. However, it is not easy for people to break up with their friends, adjust their attitudes to be harmonious with others, or persuade others to change their attitudes. Thus, people may choose the fourth way of dealing with an imbalanced state; enduring the imbalanced state. If the fiiend, 0, is very close to P and X is not critical enough to destroy the fiiendship, P 11 will bear the inconsistent state. However, enduring the inconsistent state may cause unpleasantness, tension, or stress on the friendship. Since friendships are formed and maintained within the organizational setting, these negative feelings may become relevant to an individual’s job satisfaction. The current study focuses on this last case. Job satisfaction is defined as an overall emotional reaction to a job that results from employees’ comparison of actual outcomes with expected ones (Cranny, Smith, & Stone, 1992). According to this definition, job satisfaction consists of three components: affection, outcomes related to a job, and comparison processes. Employees with greater job satisfaction perceive their jobs to be more meaningful and have greater motivation to do their jobs better. According to these effects of j ob satisfaction, job satisfaction appears very similar to job involvement because job involvement refers to the degree to which work is an important part of individuals’ life and identity (Cheloha & Farr, 1980). However, there is a major difference between job satisfaction and job involvement, which is the emotional dimension of job satisfaction. For instance, intention to work well is associated with both job satisfaction and job involvement. But employees’ liking or affection toward their job is explained only by their job satisfaction. Job involvement is a cognitive dimension, whereas job satisfaction emphasizes employees’ affective as well as cognitive dimensions (Locke, 1976; Porter, Lawler, & Hackrnan, 1975). Job satisfaction has received great attention because job satisfaction significantly influences positive organizational outcomes (Cranny et al., 1992). In the short term, if employees have higher levels of job satisfaction, their productivity increases and their absences and intention to turnover decrease (e.g., Katzell, Thompson, & Guzzo, 1992; Smith, 1992). In the longterrn, higher levels of job satisfaction are positively associated 12 with employees’ active attitude toward adapting to changed environment such as downsizing, cooperative attitudes with coworkers and positive contribution toward organizational culture and climate (Smith, 1992). Lambert (1991) also contended that job satisfaction positively influences employees’ motivation to do their job well. Lastly, higher levels of job satisfaction are related to less stress regarding their job (Ironson, 1 992). Because of these positive consequences, social scientists have explored the factors that can affect job satisfaction. In general, four kinds of predictors have been investigated. The first type of predictors is related to the job itself. Hackrnan and Oldham (1975, 1976) presented Job Characteristics Theory emphasizing that job characteristics are strongly related to employees’ job satisfaction. The theory originally followed the assumption of Needs Satisfaction Models (e.g., Argyris, 1957; Hackrnan & Oldham, 1976; Maslow, 1943). The assumption is that if jobs possess certain characteristics that fulfill employees’ needs, the employees’ satisfaction level increase. Based on the assumption, Job Characteristic Theory suggests that certain job characteristics such as skill variety, task significance, autonomy, role conflict, and feedback should be designed to satisfy employees’ needs. This proposition has been empirically supported in various studies (e.g., Bedian & Armenakis, 1987; Hackman & Oldham, 1980; Pollock, Whitbred, & Contractor, 2000). Additionally, if a job is characterized as possessing higher role clarity, pay, job security, and safety, employees report higher job satisfaction (e.g., Abdel- Halim, 1981; Rizzo, House, & Litzman, 1970; Wright, King, Berg, & Creecy, 1987). Job characteristics have received the most empirical attention and support as predictors of j ob satisfaction (Glisson & Durick, 1988). 13 The second category of predictors in terms of j ob satisfaction is management styles of organizations. Workplace justice, including the distributive and procedural justice, is regarded as an important factor for employees’ job satisfaction (Fryxell & Gordon, 1989). Procedural justice concems how decisions in the organization are made, while distributive justice is about content and consequences of the decisions (Folger & Greenberg, 1985). In other words, procedural justice regards how resources are allocated or how disputes in the organization are resolved. Distributive justice concerns the overall outcome fairness of the organization or authorities. Research shows that the manner in which management deals with grievances is an important correlate of employees’ satisfaction (F ryxell & Gordon, 1989). Much variance in employees’ job satisfaction is also explained by employees’ participation in management and decision process (Glisson & Durick, 1988). The more employees are involved in major decision processes, the more satisfied they are with their jobs. The third type of predictors of job satisfaction is found in employees’ characteristics as workers. For example, Watson and Slack (1993) showed that employees’ positive affective temperament is positively correlated with their job satisfaction. Also, Weiss, Nicholas and Daus (1999) found that individuals’ affect intensity and dispositional happiness positively influenced their job satisfaction. Employees whose mood was pleasant over time at work were more satisfied with their jobs than employees whose mood was not pleasant over time (Weiss et al., 1999). Lastly, stress reduction ability (Ratiu, 1983) and tolerance for ambiguity (Hammer, Gundykunst & Wiseman, 1978) are also regarded as individual characteristics that positively influence job satisfaction. 14 The final type of predictors is characterized as employees’ interpersonal relationships in the workplace. Although many studies have focused mainly on job characteristics, management styles, and employees’ characteristics as predictors of job satisfaction (Glisson & Durick, 1988; Neumann, 1993), an increasing attention has been paid to employees’ workplace relationships with superiors, subordinates, or coworkers as new predictors of job satisfaction. For example, the Leader Member Exchange model (LMX) shows how the quality of the relationship that employees have with their superior influences their job satisfaction. According to LMX, superiors do not use the same style in dealing with all subordinates but rather develop a difl‘erent type of relationship or exchange with each subordinate. These relationships range from those that are based strictly on employment contracts (e.g., low quality LMX) to those that are characterized by mutual trust, respect, liking and reciprocal influence (e.g., high quality LMX) (Graen, Novak, & Sommerkamp, 1982). Workers who perceive his (her) relationship with his (her) superior as having a higher quality of interaction reported higher level of job satisfaction (Epitropaki & Martin, 1999; Graen et al., 1982). The reason is that employees with higher quality LMX have less difficulty in maintaining the relationship (Lee, 1998; Lee & Jablin, 1995), greater satisfaction with their supervisors (Duchon, Green, & Taber, 1986), a higher level of satisfaction in communicating with their superiors (Lee, 1999; Mueller & Lee, 2002), and more social support from superiors (Wayne, Shore & Liden, 1997) Workplace friends also influence employees’ job satisfaction. For example, the quality of friendships at work is predictive of job satisfaction (Markiewicz et al., 2000; Winstead et al., 1995). Also, Social Information Processing Theory (SIP) (Salancik & 15 Pfeffer, 1978) accounts for the effects of social information on employees’ attitudes. Much social information comes from what their workplace fi'iends know. Based on SIP, Pollock, Whitbred, and Contractor (2000) found that employees’ satisfaction is explained not only by their job characteristics or their individual dispositions, but also by their workplace friends’ job satisfaction. Lastly, Krackhardt and Porter (1985) noticed that when unhappy employees left an organization, their friends who stayed at the organization experienced higher job satisfaction and higher commitment to their organization. Because stayers would not be exposed to their friends’ unhappiness anymore, they would restore their balance on interpretation about the organization. Taken together, these examples suggest that workplace fiiendships afl‘ect employees’ job satisfaction in various ways. A theoretical explanation is needed for a better understanding of how and why friends at work can significantly influence employees’ job satisfaction. Accordingly, the present study uses Balance Theory, because it provides some explanations for contradictory findings about the extent to which individuals’ involvement in friendship networks is related to individuals’ job attitudes and behaviors. For example, Roberts and O’Reilly (1979) found that the more involved an employee is in communication networks, the more satisfied the employee is with his (her) job. In contrast, Brass (1981) found that an employee’s involvement in communication networks was not directly related to their job satisfaction, but job characteristics moderated the relationship between communication flow and satisfaction. Finally, Kilduff and Krackhardt (1994) failed to find a relationship between actual friendship networks and job performance. Balance Theory can shed a light on these inconsistent findings. An individual’s perception about 16 his (her) friendship network plays a more important role than the sheer numbers of fiiends when it comes to influencing the job attitudes and behaviors (Monge & Contractor, 2003). Therefore, it is possible that an employee who has a balanced state with a fiiend has greater job satisfaction than an employee who has unbalanced states with multiple friends. Moreover Balance theory presents a more elaborate rationale about the empirical findings than other theories. Markiewicz, Devine, and Kausilas (2000) and Winstead et al. (1995) empirically showed that the quality of fiiendships at work significantly affects people’s job satisfaction. Especially, as employees face greater difficulties maintaining the relationship with their close fiiends in the workplace, they are more dissatisfied with their jobs. This finding can be explained by Balance Theory. Difficulties with maintaining a relationship indicate the difficulties with regard to misunderstandings, incompatibility of goals and values, and disagreements between two fiiends (Wright, 1987). These relationship maintenance difficulties may represent cognitive imbalance. If employees try to maintain the fiiendships in spite of these difficulties, they inevitably experience some tensions and stresses caused by the cognitive imbalance. Therefore, coping with these relationship maintenance difliculties can be associated with decreased job satisfaction. Taken together, Balance Theory provides clearer explanations about inconsistent findings and theoretically unexplained findings. In sum, when applying Balance Theory to blended friendships, it can be argued that disagreements between two friends at work can lead to unpleasantness and stress. Workplace friendships have been considered as significant predictors of employees’ job satisfaction. The question is what dimensions are important for individuals’ job 17 satisfaction. As an important dimension of workplace friendships, the next section will provide a discussion on organizational climate, followed by a hypothesis and a research question. Hypothesis and Research Question Despite the usefulness of Balance Theory, one challenge for the theory is that Heider (1 95 8) did not specify what constitutes X in the triads of P-O-X (Curry & Emerson, 1970; Insko, 1981). Thus, it is necessary to examine a wide variety of X to see which type of X is important for the balance among a triad. First, a boundary of X can be designated (Curry & Emerson, 1970; Eagly & Chaiken, 1993; Zajonc & Burstein, 1965). One prime boundary is that X should be relevant to both P and 0. Suppose that imbalance occurs in a triad of P-O-X. If X is relevant to both P and 0, P will experience unpleasantness. However, if X is not relevant to P and 0, P does not need to care about the imbalance. As a result, P does not experience stress from the imbalance. One other criterion for X is that X should be a critical object, issue, or person for both P and 0. Suppose again that imbalance happens in a triad of P-O-X. If X is not critical to either P or X, P will endure the imbalance with no stress or trivial unpleasantness. Otherwise, P will face significant imbalance leading to decreased job satisfaction. Considering that the current study focuses on fiiendship networks in workplaces, it is necessary to examine X as a work-related issue. Specifically, the current study focuses on psychological climate as a work-related issue. There is a primary distinction between psychological climate and organizational climate (James & Jones, 1974). Whereas organizational climate is a shared perception that people attach to particular characteristics of their organization, psychological climate refers to individuals’ 18 perceptions with respect to various aspects of their working environment. As a result, psychological climate can be idiosyncratic unlike organizational climate. Psychological climate has been considered significant for employees’ job satisfaction and performance (e.g., Hellrifgel & Slocum, 1974, Muchinsky, 1977; Redding, 1972). Specifically, Hofrnann and Stetzer (1996) found that a climate promoting safety was negatively associated with employees’ unsafe behaviors. Kozlowski and Hults (1987) showed that a climate that promoted innovation and updating influenced engineers’ actual performance related to innovation and updating. Therefore, employees’ psychological climate is important for their job attitudes as well as their job performance. The psychological climate is defined as sets of attributes that individuals perceive about relevant organizational contexts including subsystems, features, events, and processes (James & Jones, 1974). This definition shows that the climate is individuals’ cognitive perception or interpretation of the organizational environment (James & Jones, 1974). In other words, each individual filters, structures, and describes numerous type of context from the organization, and makes one summary perception about the organization’s work environment. Accordingly, psychological climate is descriptive rather than evaluative in nature (James & Jones, 1974; Schneider, 1975). There are important characteristics of psychological climate. First, a climate is a function of the way employees perceive organizational processes. At the same time this perceived climate influences individuals’ organizational behaviors (Cohen, 1995; Jones & James, 1979). Second, different perceptions of a climate can exist within the same organization because individuals can interpret the same environment dissimilarly based on their own personalities and needs (Cohen, 1995; Downey, Hellriegel, & Slocum, 19 1975). Because of the possibility that different individuals may form different perceptions about the same thing, a shared organizational climate may not be same as each individual’s psychological climate (Joyce & Slocum, 1984; Patterson, Payne, & West, 1996). Last, aclimate consists of multiple dimensions (Kopehnan, Brief, & Guzzo, 1990). In other words, a climate includes prevailing norms related to organizational goals, means to attain the goals, reward systems, allocation of resources, and social-emotional support from others (Muchinsky, 197 7). Various factors contribute to climate. Key factors internal to organizations include organizational structures, power distribution, supervisory practices, informal group culture and individual characteristics. The main external factors include social, economic, and legal conditions, and environmental turbulence (e.g., Cohen, 1995; Joyce & Slocum, 1984; Reichers & Schneider, 1990). As these various factors influence organizational climate simultaneously, employees within the same firm may differ substantially in their interpretation of their work setting. Because a climate is described, but not evaluated by each employee (Cohen, 1995), there is neither a good nor had climate. Climate just functions as a moderator, a mediator, or an independent variable. For instance, if individuals’ values and traits are congruent with their perceived climate of their workplace, they perform better, and express greater job satisfaction (Downey, Hellriegel, & Slocm, 1975). Kozlowski and Doherty (1989) suggested that individuals with similar attitudes and traits experience similar socialization processes, are exposed to similar aspects within enviromnents, and finally share their interpretations with others in the organization. Thus, generally, similar people form similar perceptions of a climate. Finally, Muchinsky (1977) found that 20 organizational communication style is highly related to people’s psychological climate and job satisfaction. Since employees share their interpretations with their coworkers, superior and subordinates, people who interact with each other have similar perceptions about their organization. The process of perceiving organizational climate is interactive and reciprocal (Kozlowski & Doherty, 1989). While interacting with each other, friends at work share their interpretations with each another. If their interpretations are not similar, they may experience discomfort caused by the disagreement. Further they may face unanticipated organizational conflicts caused by the disagreement. Disagreements with superiors are also negatively related to job satisfaction. Wexley, Alexander, Greenawalt, and Couch (1980) reported that if a subordinate is congruent with his (her) supervisor in terms of attitudinal similarity, the subordinate is more satisfied with his (her) job. Although the relationship with superiors is mainly considered as formal relationships, this rationale is expanded to blended friends network as an informal relationship at workplace. These feelings of discomfort feelings and conflicts with their fiiends affect their friendships quality negatively. Considering that the quality of fiiendships at work is crucial for an employee’s job satisfaction (Winstead et al., 1995), stress, tensions, and conflicts in the fiiendships quality can be negatively related to job satisfaction. Disagreements can decrease employees’ job satisfaction. Therefore, the following hypothesis is advanced: H1: The more similar employees are with their workplace friends in their perceptions about psychological climate, the more satisfied they will be with their job. Many network researchers have used pairs of individuals who share membership 21 in an organization as their unit of analysis (e. g. Boyd & Taylor, 1998; Lincoln & Miller, 1979; Sias & Cahill, 1998; Winstead et al., 1995). Although most peOple have more than one friend in their workplace, researchers have focused only on each employee’s one best fiiend. This research practice is congruent with Weick’s (1969) argument: Even though most networks contain more than two people and more than a single relationship, in actual functioning only one dyad and one relationship are activated at any moment in time. The basic unit in the network remains a dyad, the members of which interlock their behaviors relative to the particular components of the task that each possesses. (p.98) This argument suggests that an agreement with one best friend will be more strongly related to employees’ job satisfaction rather than the aggregated agreement with all of their fiiends at work will be. Although one best friend can be very influential, a group of friends as a clique also has the potential for influencing an individual worker’s decisions and actions (Rogers & Kincaid, 1980). According to a theory of social comparison processes (F estinger, 1957), people consider others’ opinions in evaluating their own opinions and abilities. In many cases of comparison processes, people think about opinions and ideas of intimate people as social information in order to evaluate their ideas. Friends are one prime example of intimate people. A theory of social comparison processes suggests that if a person, P finds the difference between his (her) opinion and other specific person, 0’s, P would try to consider some other intimate person’s opinions for precise comparisons. So if P perceives that the closest fi'iend, 0’s perception is very divergent from P’s with respect to organizational climate, P would consider other friends’ interpretations. Also a 22 theory of social comparison processes indicates that people tend to pursue stable evaluations over unstable ones. Even when people’s opinions are compatible with one other person, multiple comparisons provide more stable information. Accordingly, if P perceives that his (her) closest friend, 0’s perception is dissimilar with P’s in terms of psychological climate, P would seek to check out other friends’ interpretations. In fact, Krackhardt and Porter (1985) showed that all fiiends contribute to stayers’ attitudes, when they examined how blended friends’ departure affect stayers. The argument on the superiority of one best friend over a group of friends needs an empirical examination. Therefore the current study will examine the difference between the effect of the closest fiiend and the collective effect of all friends in an organization on employees’ job satisfaction. The following research question is advanced. RQ: Is it similarity in employee’s perceptions about psychological climate with one . best fiiend or a group of fiiends which is more strongly related to their job satisfaction? 23 CHAPTER 2 METHOD Sample Employees in two organizations (hereafterA and B) located in Kwangju and Seoul in South Korea participated in the study. Organization A is a non-profit local health center run by the Korean government, and organization B is headquarters of one of the major insurance companies in Korea. Several organizations were solicited by the researcher. Since a desirable number of employees in the organizations A and B voluntarily participated in the study, these organizations were selected for the current study. All 30 (10 males and 20 females) employees in organization A voluntarily agreed to participate in the study. Fifty one (22 males and 29 females) employees out of 200 employees in organization B also voluntarily agreed to participate in the study. In order to obtain a complete network of friendship, it is desirable that all employees within each organization participate. Unlike organization A, only 51 out of 200 employees of organization B participated in the study, but these 51 employees work on the same floor. Except for three managers, 48 employees hardly interact with other employees working in different floors. Therefore, the 51 employees in organization B were used as participants in the current study. All of the participants were ethnically as well as culturally Koreans. The average age of the employees was 42 years old (SD = 9.31, Median = 41, Mode = 49) ranging fiom 23 to 57 in organization A and 29 years old (SD = 5.28, Median 24. = 28, Mode = 27) ranging from 19 to 47 in organization B. The employees worked in organization A for 74 months on average (SD = 96.89, Median = 25, Mode '4' 120) ranging from 1 to 320, and the employees worked in organization B for 64 months on average (SD = 48.57, Median = 50, Mode = 50) ranging from 7 to 260. Employees in organization A work on tasks related to health management, immunization, sanitization, primary care of residents in the local area and so on. Employees in organization B perform tasks associated with developing insurance products, long-terrn insurance management, marketing, public relations, information technology and so on. For education level, 66 (82 %) of the participants had 4-year college degrees or higher level degrees, 9 (11 %) had high school degrees, and 6 (7 %) did not indicate their education level. Procedure Participants were asked to identify their fiiends in the list of all the employees in the questionnaire.1 Accordingly, pre-approvals from all of the employees were obtained before the questionnaire was developed. Participants were told that the purpose of the study was to investigate employees’ job satisfaction related to workplace fiiends’ interpretatiOn about the organization climate. To protect the participants’ confidentiality, all participants were instructed to directly mail their completed questionnaire to a designated person in Korea unrelated to the organizations and participants. Then, the designated person changed the participants’ names into numbers (ex. 501, 502 etc.). The person was designated by the researcher before the questionnaire was distributed to the employees. In addition to the questionnaire, participants received envelopes with a return address and postage. 25 Measures The questionnaire consists of three parts: a fiiendship network measure, an organizational climate scale, and a job satisfaction scale. All of the items in the questionnaire were written in Korean. Friendship network measure. Closeness of friendship was assessed by asking participants to indicate the degree of closeness to all other employees within the organization in the list. The response format for these measures was a 5-point scale (1 = just a coworker, 5 = a very close fiiend). Additionally the participants were asked to indicate an average communication time with each of them at workplace and outside of the work (see Appendix A). After identifying the closeness and daily average communication time at workplace and outside of work, they were asked to write the name of one best friend in the workplace in the blank space. Psychological climate. Psychological climate was measured with Muchinsky’s (1977) modified scale derived from the Litwin and Stringer’s (1968) original one. The scale had 47 items for six factors, which are interpersonal milieu, standards, affective tone toward management/organization, organizational structure and procedures, responsibility, and organizational identification. Interpersonal milieu indicates the interpersonal relations environment that prevails in the organization. The dimension of standards describes the feeling that the organization has established the standards of performance. General affective tone toward management/organization identifies the perceived image on management. Organizational structure and procedure identifies the feelings people have related to processes and things done in the organization. Responsibility indicates the feelings concerning who has the responsibility for getting a 26 job done. Finally organizational identification describes the feelings being a part of the organization (see Appendix B). The response format for these measures was a 5-point scale (1 = strongly disagree, 5 = strongly agree). An example item was “The jobs in this organization are clearly defined and logically structured.” Items 2, 4, 6, 7, 9, 14, 15, 16, 19, 20, 23, 25, 32, 40, and 46 were recoded since these items were reverse items. When using EQS, CFA results for multidimensionality of six factors were not acceptable (NF I = .437, CF I =.577, GFI = .672, AGF I = .614).2 These results were also consistent with Hunter’s Confirmatory Factor Analysis (CFA). For this CFA method, the predicted correlation for each combination of items based on factor loadings is computed first, and then the differences between predicted and observed correlations are calculated. If tests for internal consistency and parallelism include a lot of large deviations, the multidimensional factors for climate are not accepted. Six factors for organizational climate actually could not be accepted because of many large deviations. After removing items with large deviations, CFAs were repeatedly conducted to find items and factors consistent with multidimensionality and parallelism. This procedure is part of model generating applications (J dreskog, 1993). The model generating applications are reasonable when the initial model does not fit the data and is modified by the researcher based on theoretical sense and reasonable statistical correspondence to the data. As a result, two new factors emerged from both EQS and Hunter’s CFA results. These were affective tone as the first factor and responsibility as the second factor. The first factor included 7 items numbered 24, 29, 31, 33, 35, 38 and 44. The example items were “Our management is willing to take a change on a good idea,” and “People are proud of belonging to this organization.” The second factor included 5 items numbered 11, 12, 13, 27 40, and 46. The example items were “Supervision in this organization is mainly a matter of setting guidelines for your subordinates: you let them take responsibility for the job,” and “You won’t get ahead in this organization unless you stick your neck out and try things on your own sometimes.” EQS results showed considerably stable multidimensionality (NFI = .724, CFI = .812, GFI = .823, AGFI = .740). In addition, Hunter’s CFA results supported internal consistency within the factors and parallelism between factors (See Appendix D). Reliabilities (Cronbach’s a) of these two factors were .84 and .76 respectively. Job satisfaction. The Minnesota Satisfaction Questionnaire (MSQ) short form (Weiss, Dawis, England, & Lofquist, 1967) has been widely used to measure job satisfaction. Many studies show the scale is highly reliable. To name a few, Corbett, Martin, Wall, and Clegg (1989) reported .88, Naumann (1993) reported .90, and Watson and Slack (1993) reported .91. However, MSQ has been criticized for having little affective content in spite of the fact that job satisfaction is defined by its affective dimension (e. g. Brief & Roberson, 1989). Another problem related to the measurement of job satisfaction is that most employees tend to report at least moderate level of j ob satisfaction (Hamilton & Wright, 1986). Asian employees are especially reluctant to report their dissatisfaction with promotion and pay (Money & Graham, 1999). In Money and Graham’s (1999) study, items for pay and promotion were reported to have low level of reliabilities such as .66 and .65 (Money & Graham, 1999). Considering that data were obtained from Korea, it would therefore be better to exclude these items. Thus, job satisfaction measured in this study included a few affective items from Money and Graham’s (1999) study in addition to MSQ short form and excluded items related to pay 28 and promotion. Finally, 23 items were used in the study (see Appendix C). The response format for these measures was a 5-point scale (1 = strongly disagree, 5 = strongly agree). An example item was “I am satisfied with being able to keep busy all the time.” Items 2, 6, 9, 13, 17, 19, and 23 were recoded because these items were reverse items. Hunter’s Confirmatory Factor Analysis (CFA) program was used to perform a test for internal consistency. There were many significant deviations, which indicated data inconsistency with a unidimensional factor. As a result, 11 items were removed from the analysis. The remaining 12 items were items numbered 1, 4, 5, 9, 12, 14, 15, 17, l8, 19, 20, and 21. CFA result produced a unidimensional model with these 12 items. Therefore these 12 items were used for further analyses. The reliability (Cronbach’s a) of the job satisfaction scale in the current study was .81. Operational Definitions Before testing the hypothesis and research question, several issues with respect to the operational definitions should be addressed. These issues address various ways to compute the similarity and ways to define one best friend and friends operationally. Operational definition of similarity. Mainly there are two ways to compute the similarity of climate between friends. The factors (hereafter Affective tone and Responsibility) of climate were measured with seven items and five items, respectively. One way is to compute similarity from the values of factors. After calculating the mean of each factor from the values of items, absolute differences between means of fi'iends were obtained. Since the possible maximum value of difference is 4 (the subtraction between 5 [strongly agree] and 1 [strongly disagree]), the difference between 4 and obtained difference is the similarity score. The other way is to calculate similarity from the values of items. The absolute differences obtained from each item for the factors were averaged and then subtracted from 4. For exarrrple, person A and person B answered 1, 3, 5 and 3, 3, 3 for one factor with 3 items. Four is obtained from the first method while 2.67 is obtained from the second method. Among the pairs of (1, 3), (3, 3) and (5, 3), however, only one pair is similar. If the means of each factor are compared, the dissimilarity of the first pair is canceled out with the dissimilarity of the third pair. The second method covers broader disparities than the first method. Therefore the data were analyzed with the second method in the study.3 Finally, the reliabilities (Cronbach’s a) of the similarities in affective tone and responsibility with the best friend in the current study were .96 and .78, respectively. The reliabilities (Cronbach’s a) of the similarities in affective tone and responsibility with a group of friends were .95 and .96, respectively. Operational definition of the best friend. The other issue involves how to define friends and the best friend operationally. Since participants wrote down one best fiiend on the questionnaire directly, identifying the best friend is very clear. There are two types of best friends, however. One-way best fi'iend is the person whom a respondent indicated as the best fiiend in the questionnaire. Two-way best fiiend is the person whom a respondent indicated as the best friend who indicated the respondent as the best friend. For example, when a person A indicated a person B as the best friend and the person B also indicates the person A as the best friend, the person B is the two- way best friend to the person A. Whether the person B indicated the person A as the best fiiend or not, the person B is 30 one-way best friend, as long as the person A indicated the person B as the best fiiend. If respondents identify their best fiiends, they have at least one-way best friends. If two respondents consider each other as their best friends, the two respondents have two-way best friends. As a result, the respondents with two-way best fiiends were a subset of the respondents with one-way best friends. In the current study, only one-way best friend was considered.5 In organization A, 11 out of 30 employees did not report their workplace best friends while 19 employees reported their best friends on the list in the questionnaire. Four employees identified two best friends. In organization B, six out of 51 employees failed to report their best fiiends. Among 45 employees who indicated their best friends, four employees indicated two best friends. Operational definition of friends. Operational definition of friends has a different issue from the operational definition of the best friend because fiiendship with other employees was measured with continuous scale (1 = just a coworker, 5 = a very close friend). First, the dichotomous way was applied. That is, employees rated 4 (close friend) and 5 (very close friend) were considered as friends. This method excluded employees rated 1 (just coworker), 2 (coworker acquainted), and 3 (coworker having casual conversations). The second way is to consider similarity of all other employees and to weigh the friendship. If the person A rated 2, 3, and 4 for person B, C, and D in the closeness measure, only person D is considered as a friend in the first method. In the second method, similarities with person B, C and D are considered with the degree of closeness with each person. Finally, only one-way fiiends were considered in the current study like the best friend case by the dichotomous way.6 As a result, two different 31 methods were employed for friends. They are one-way friends, and weighted friends. When the dichotomous way was applied, four employees do not have their workplace friends while 24 employees have fiiends in the organization A.7 The average number of friends is 8.37 ranging from 0 to 29 (SD= 7.745, median=5.5, mode=0). In the organization B, two employees do not have their workplace friends while 48 employees have their workplace fi‘iends.8 The average number of fiiends is 8.1 ranging from 0 to 28 (SD= 6.152, median=7, mode=6). Procedure to analyze The similarities of affective tone and responsibility with one best fiiend and a group of friends were obtained using SPSS MATRIX. With the similarities, SPSS LINEAR REGRESSION was used to investigate the effects of the affective tone and responsibility similarities on job satisfaction. Before conducting the linear regression, the assumption of linearity was roughly confirmed through the scatter plots between similarities in affective tone and responsibility, and job satisfaction. When the similarities were regressed on job satisfaction, the type of the organization was also regressed on job satisfaction because two organizations are not equivalent in terms of employees’ demographical characteristics and organizational characteristics. To name a few, organization A is located in the region of Chunra County, which is the hometown of all employees. In contrast, employees of organization B came from various regions and work at Seoul, the capital of Korea. Organization A is a nonprofit organization whereas organization B is a for-profit one. As a result, tasks employees perform in the organizations are very different. Therefore, it is necessary to consider the organization type as a dummy variable (i.e. Organization A = 0; Organization B = 1). 32 CHAPTER 3 RESULTS Overview One hypothesis and one research question were tested in the current study. The hypothesis predicted that if employees have more similar perceptions on organizational climate with their workplace friend(s), they would be more satisfied with their job. The 1 research question asked whether similarity with one best friend or with a group of friends is more strongly related to an employee’s job satisfaction. For the hypothesis and the research question, three moderated regression analyses were conducted for the dependent variable (job satisfaction) as shown Table 1, 2 and 3. The independent variables (similarity in affective tone and the similarity in responsibility) were centered before entering in the equation (cf., Cohen, Cohen, West, & Aiken, 2003). Eflects of Climate Similarities with Best F riend(s) on Job Satisfaction Main eflects. The regression analysis showed non-significance, R2=.047, F (3, 59)=0.968, ns (See Tablel). Similarity in affective tone with the best friend at workplace did not predict job satisfaction, fi=.052, t(62)= 0.387, us. Also, similarity in respOnsibility with the best fiiend did not predict job satisfaction, fl=—.215, t(62)= -1.618, us. The type of organization did not make any difference in job satisfaction, fi=—.043, t(62)= -0.327, ”S. 33 Interaction effects. The dependent variable was regressed onto the product terms of the independent variables (i.e. similarity in affective tone with the best friend x the type of the organization; similarity in responsibility with the best fiiend x the type of the organization). As shown in table 1, the interaction between similarity in affective tone with the best fiiend and the type of the organization was not statistically significant, ,B=.152, t(62)= 0.694, ns. The interaction between similarity in responsibility with the best friend and the type of the organization was not statistically significant, ,B=.03 8, t(62)= 0.205, ns. Eflects of Climate Similarities with F riend(s) on Job Satisfaction Main eflects. As mentioned in the overview of the results, the effects of similarity with the group of friends were measured two ways, one-way friends, and weighted friendship. In the one-way friends, the regression analysis results showed significance (See Table2). Similarity in affective tone with the friends at workplace did not predict job satisfaction, ,6=-.045, t(71)= -0.352, ns. In contrast, similarity in responsibility with the friends predicted job satisfaction, ,B=-.305, t(71)= -2.413, p<.05. The type of organization did not make any difference in job satisfaction, fl=.097, t(71)= 0.838, ns. These three predictors explained a significant proportion of variance in job satisfaction, R2=.113, F (3, 68)=2.878, p<.05. However, the standardized coefficient ,6 for the sirrrilarity in responsibility (-2.413) is the opposite direction with the hypothesis. While the hypothesis predicted a positive relationship between similarity and job satisfaction, the data showed a negative relationship between the two variables. In the weighted friendship, the regression analysis results showed that these three predictors did not explain significant proportion of variance in job satisfaction, R2=.013, F(3, 74)=0.335,ns (See Table 3). Similarity in affective tone did not predict job satisfaction, fl=.065, t(77)= 0.467, ns. Also, similarity in responsibility did not predict job satisfaction, fl=—. l 32, t(77)= -1.002, ns. The type of organization did not make any difference in job satisfaction, ,B=—.005, t(77)= 0038, ns. Interaction eflects. In one-way friends, the dependent variable was regressed onto the product terms of the independent variables (i.e. similarity in affective tone with friends x the type of the organization; similarity in responsibility with fiiends x the type of the organization). As shown in table 2, the interaction between similarity in affective tone with friends and the type of the organization was not statistically significant, fi=.084, t(71)= 0.416, ns. The interaction between similarity in responsibility with fiiends and the type of the organization was not statistically significant, fl=.207, t(71)= 0.915, ns. In the weighted friendships, the dependent variable was regressed onto the product terms of the independent variables (i.e. similarity in affective tone with weighted friendships x the type of the organization; similarity in responsibility with the weighted friendships x the type of the organization). As shown in table 3, the interaction between Similarity in affective tone and the type of the organization was not statistically significant, fl=—.13l, t(77)= 0664, ns. The interaction between similarity in responsibility and the type of the organization was not statistically significant, fi=—.085, t(62)= -0.478, ns. Efi’ects of Climate Similarities with Best Friend or with Friends on Job Satisfaction The research question asked whether variance in an employee’s job satisfaction 35 would be accounted for more by similarity with one best friend or similarities with a group of friends. The proportion explained by similarity with best friend is not statistically significant, R2=.047, F (3, 59)=0.968, ns. In contrast, the proportion explained by similarities with friends is statistically significant, R2=.113, F (3, 68)=2.878, p<.05. 36 CHAPTER 4 DISCUSSION The data were not consistent with the hypothesis predicting positive relationship between similarity in climate with friends and job satisfaction. The finding is that the similarities in responsibility with workplace friends are negatively related to employee’s job satisfaction. The data also showed that the similarities in responsibility with a group of friends explained employee’s job satisfaction better than the similarity in responsibility with one best fiiend. Implication for the Relation between Similarities in Climate with Friends and Job Satisfaction The current study predicted that the more similar perceptions about climate that employees share with their workplace friends, the more satisfied employees are with their jobs. The finding showed, however, that the more similar employees are with fiiends in terms of climate, the less satisfied they are with their jobs. Five explanations may explain this inconsistent finding. The first possibility is the difference between the actual sirrrilarity and the perceived similarity. Each person’s perception on the climate and their workplace friendship were directly measured and the similarities with fiiends were computed by the researcher in the current study. If people believe that their friends have similar perceptions with them, friends’ actual perceptions do not necessarily influence 37 their cognitive balance. Although employees communicate with their workplace fiiend, they might not obtain any specific information about their fiiends’ perceptions on climate. If so, people are more likely to assume that their in-group members have similar attitudes and beliefs with them. Muraskin and Iverson (195 8) found that people expected that the “ideal” American has a similar pattern of social distance ratings with their own. Fang and Kenrick (2002) empirically showed that people assumed in-group members’ attitudes to be more similar with their own attitudes than ones of out-group members. Therefore, the disparity between the real similarity and the perceived similarity might have led to the result inconsistent with the hypothesis. Second, some extraneous variables had effects upon job satisfaction. As mentioned before, various factors influence employees’ job satisfaction such as job characteristics, management styles, worker’s characteristics and interpersonal relationships in the organization (e. g. Fryxell & Gordon, 1989; Graen et al., 1982; Hackrnan & Oldham, 1975, 1976; Watson & slack, 1993). When participants’ demographical information such as age, the length of the employment, hometown, gender, and educational level were examined as the predictors of job satisfaction prior to the main analyses, no variable had any significant effects on job satisfaction. Nevertheless, other potential effects such as workers’ personalities, job characteristics, and relationships with coworkers and superiors were not examined in the current study. Unidentified effects might have changed the direction of the relationship between the similarities with fiiends and job satisfaction. Third, the inconsistent finding against the hypothesis might be due to the relationship between cognitive balance and culture. Most of the evidence supporting the 38 Balance Theory comes fi'om North America, and the hypothesis of the current study was develOped based on North American literature. If the human nature inherently has motive to keep cognitive balance, it should be fairly pervasive across cultures. Because imbalance is assumed as a cognitively uncomfortable state (Heider, 1958), the Balance Theory emphasizes cognitive consistency. However, some empirical studies suggested that Asians are more tolerant toward cognitive inconsistency than North Americans. For example, Suh’s (2002) study showed that the relationship between consistency and psychological well-being was stronger in North America than in Korea. In Suh’s study, Korean and North American undergraduate students were asked to describe themselves in several relationships. Koreans’ self-descriptions were more flexible across situations than the North Americans’ ones were. Furthermore, this inconsistency was less strongly related to psychological well-being for the Koreans than for the North Americans. Cross, Gore and Morris (2003) also found that if the self was constructed to be interdependent with others, consistency was less important in mental health. Although these studies did not test the balance theory directly, considering that imbalance is similar with inconsistency, imbalance may not be a universally essential component of psychological well-being. If it is true, the imbalance caused by the dissimilarity with friends might not influence Korean employee’s job satisfaction. Fourth, the inconsistent direction of the relationship between the similarities with friends and job satisfaction ([3 = -.305) rrright be due to sampling error. Since its 95% confidence interval ranges from -.558 to -.053, this possibility is not high. In spite of this, the sample size of the current study is quite small, so this possibility cannot be ruled out. Finally, the inconsistent finding against the hypothesis might be because of the 39 false prediction. Employees may not care about similarities in climate with their workplace friends, so the similarities do not positively affect their cognitive balance. Instead, the dissimilarities with their friends in terms of climate can be interpreted as diverse climate in the organization. Ifthe organization has diverse climate, the employees have diverse perceptions on the climate and the organization will satisfy various employees’ needs. If this is true, the finding of the current study is consistent with this rationale. That is, the less similar with their fiiends, the more satisfied employees are with their job. If the first, second and third explanations are true, there will be no relationship between similarities in climate with workplace fiiends and job satisfaction cannot be expected. In contrast, if the fourth and fifth explanations are true, the negative relationship will exist between similarities in climate with workplace friends and job satisfaction like the finding in the current study. Implication for Similarities The similarity scores in affective tone and responsibility with friends were used as predictors in the multiple regressions. The similarity scores were computed from the absolute difference between participants’ own score and their fiiends’ score. This difference scores have been pervasively used in organizational behavior research (Edwards, 1994) and in personality research (Watson, Hubbard, & Vl/"rese, 2000). In spite of their widespread use, difference scores suffer fiom many methodological problems (Edwards, 1994). The current study also has these problems. One of the main problems is failure to compute the reliability of difference score. According to the well-known formula (Guilford, 1954), the reliability of a difference score can be articulated in terms of reliabilities of the prescores and postscores. When participants’ scores and their 40 friends’ scores are considered as the presocres and the postscores, the reliability of the prescores are very similar with the reliability of the postscores because participants’ friends also are participants in the current study. As a result, the reliability of the difference scores is not obtained with this method. Moreover, the absence of the reliability of the difference scores makes it difficult evaluating the measurement errors of the difference scores. When the difference scores are computed from participants’ scores and their friends’ scores, the assumption of independent observation is violated because participants’ fiiends are also participants in the current study. As a result, the significant tests may be biased and misleading. Finally if the difference scores are used as independent variable, range of the variable is restricted and subsequently reduces the explained variance (Edwards, 1994). Implication for Climate The study shows that Muchinsky’s (1977) scale has a problem of invalidity and unreliability. The original climate scale had six factors, but only two factors showed moderate reliability and parallelism. This problem is pervasive in other climate research. For example, Joyce and Slocum Jr. ’3 study (1984) reported to use unpublished Climate scale developed by Campbell and Pritchard (1969). The scale has 6 dimensions like Muchinsky’s (1977), but 5 factors were measured with only two or three items, and the reliabilities (Cronbach’s a) of four factors are .59, .56, .56, and .53, repectivley. Although the sample size (695) was big enough in Muchinsky’s (1977) study, low reliabilities in two factors were reported. If the measurement is not valid, firrther investigation is not possible. Therefore, valid scale development is necessary for future climate research. 41 Limitations and Directions for Future Research This research represents a theory-driven examination of how similarities in climate with fiiends relates to job satisfaction. For this, all fiiendship networks within organizations were acquired and an effect of a group of fiiends as well as an effect of the best friends was investigated. Usually cognitive inconsistency is a hypothetical construct, so it was measured indirectly by the strength of attempts to decrease it in consistency research (Oshikawa, 1970). But this study tried to measure imbalance directly through the differences among friends in climate. In spite of these merits, the study has several limitations. First, the sample size is very small. Vlfrth larger sample size, the error will decreases and the power will increases. When asking the best friend, only 63 participants out of 81 indicated their best friends on the list. 18 participants who failed to write their best friend’s name might be negligent to answer the question, or might not have one best fiiend, but a group of friends. Considering that 72 participants have their close friends in workplace and 78 participants reported their closeness, the second reason, not having one best fiiend, is more reasonable than the first reason, participants’ negligence. Since employees may not have single best fiiends, some missing values are expected when investigating participants’ best friends. Accordingly, bigger sample size is necessary for the expected missing values. Moreover, more samples fiom various organizations are needed for the generalization of the finding. Therefore, big samples are desirable for future study. Second, the study was not sure whether the disparity between the real similarity and the perceived similarity might have led to the result inconsistent with the hypothesis. 42 Measuring the perceived similarity may address the possibility. When asked to report their friends’ perceptions on climate, the results would provide participants’ perceptions on their friends’ perception on climate. If participants are asked to evaluate their fiiends’ perceptions on climate, it also would not violate the assumption of observation independence. Thus, the similarities with fiiends should be assessed by this method. Third, other XS should be examined. The current study only focused on climate in examining balanced state. Other factors may lead significant effects on the triadic relationships. If participants’ fiiends are coworkers under the same supervisory, similarity of perceptions on the superior between fiiends form new triad. Or participants’ values on fiiendships can be considered X. Friendships provide three values for a person: stimulation value, utility value, and ego support value (Wright, 1969). Iffriends prioritize different values from each other, it will lead to cognitive imbalance. Like these examples, various things can be possible Xs in the triads, subsequently influencing job satisfaction. Various Xs should be addressed in the further research. Conclusion It is very likely for employees to forrrr friendship with those whom they interact with within organizations. These blended fiiendships influence employees’ not only social behaviors but also organizational behaviors. The current finding showed that similarity in responsibility with fiiends was negatively associated with job satisfaction. This finding implies that the cognitive imbalance leads to effects on organizational behaviors. Better understanding of the cognitive balance related to workplace friends is valuable for organizations. Further research on the relationship between fiiendships and job satisfaction may lead to a useful extension of Balance Theory. 43 FOOTNOTES 1. For organization B, only 51 employees, not all 200 employees, were listed in the questionnaire. 2. Although the sample size in the current study was 81, 19 (4 males and 15 females) additional samples were included for factor analyses. These 19 participants were employees from various organizations in Korea. The average age of the employees was 31 years old (SD = 7.91, Median = 30, Mode = 25) ranging from 22 to 55. 3. Although the current study reported only the results using the second method, the results using the first method showed similar results with ones using the second method. If you would like to inquire about the results using the first method, please contact the author. 4. For participants who failed to identify their best fiiends, the best fiiend was estimated based on the highest score on closeness and longest time to communicate at the workplace and out of work. Although the estimated best fiiends increased the sample size fi'om 63 to 72, the result of t-test showed that the participants who reported their best friends are statistically significantly different from the participants who did not reported their best friends in terms of main independent variables (similarity in affective tone and similarity in responsibility with the best friend). Therefore, the current study did not include the estimated best fiiends in the data analyses. 5. The results from the two-way best friends are consistent with the results from the one-way best fiiends. You can get the results for the two-way best fiiends from the author. 6. The results from the two-way fiiends are consistent with the results from the one-way fiiends. You can get the results for the two-way friends to the author. 7. Only two people out of total 30 employees in the organization A did not complete their closeness with other employees. So four employees who do not have workplace friends indicate that they do not rate any person 4 [close fiiend] or 5 [very close friend]. 8. Only one person out of total 51 employees in the organization B did not complete their closeness with other employees. So two employees who do not have workplace friends indicate that they do not rate any person 4 [close fiiend] or 5 [very close fiiend]. 45 Appendix A. Friendship Network Measure Departrn Name How close are you with the How much How much ent following person? average time a day average time a /Position 1: Just a coworker do you have with day do you have 2: An intimate coworker the following with the 3: A fiiend person WITHIN following person 4: A close friend the organization OUTSIDE of 5: A very close friend for drink, meal work for drink, and conversation meal and etc.? conversation Ex. 3 hr 10 min etc.? Ex. 2hr 5 min 1 2 3 4 5 _ hr min hr min 1 2 3 4 5 _ hr __ min min 1 2 3 4 5 hr min hr rrrin l 2 3 4 5 hr min hr min 1 2 3 4 5 hr rrrin hr min 1 2 3 4 5 __ hr _ rrrin hr min 1 2 3 4 5 __ hr _ min hr min 1 2 3 4 5 hr rrrin hr min 1 2 3 4 5 __hr__min hr min 1 2 3 4 5 _ hf __ min hr min 1 2 3 4 5 hr min hr min I 2 3 4 5 __ hr __ min hr min 1 2 3 4 5 __ hr __ min hr min 1 2 3 4 5 hr min hr min 1 2 3 4 5 hr min hr min 1 2 3 4 5 hr nrin hr min 1 2 3 4 5 hr nrin h! rrrin 1 2 3 4 5 hr min hr min 1 2 3 4 5 hr min hr rrrin l 2 3 4 5 hr min hr "Fin 1 2 3 4 5 hr min hr min 1 2 3 4 5 hr min hr min 2 3 4 5 hr rrrin hr min 46 Appendix B. Korean Version of Friendship Network Measure AM one see or an or: as are Lll0likl act are 110114 not ma w Morena? ilLl 4w. crisis til—l 4m. can: {told-E715 Zlolfi‘rtf’fi‘é 1: :13 545* SE BEE 0| Eli eae Ol El'l 2. 8.1-“i154 5%! 81E 5E EEEQE 81! 1937542.: Si! 3: will Ll-‘r-E Si 5 i Alfli Zlol 5 E Alli? 4: {'3 5E. enigma? ZIOI ELllalLlDl? 5. 013? that SE 01|)3A|Zl 10E 01) 2M1? SE 1 2 3 4 5 _Allt_E ___A|z_r__g 1 2 3 4 5 ____Al?_l_E __AlZl_E 1 2 3 4 5 __Alzl____E _Al?_t__E 1 5 ___Allt__E _A|z_r_ru_l 1 2 3 4 5 __ AI?! _ E __ Al?! _ E 1 2 3 4 5 _Allt___,E __AIZ!___E 1 2 3 4 5 __AI?_t__E _AIZ!_E 1 2 3 4 5 _Allt_E _A|11__g 1 2 3 4 5 _Al?_t_E _A|Zt_E 1 2 4 __Alzt__E _A|?.t__E 1 _Allt_E _A|z_r_g 1 2 3 4 5 _Alat_E __AI?J_E 1 2 3 4 5 _Al?_t_E __AI?J_E 1 2 3 4 5 _M?J____E __AIZ."_E 1 2 3 4 5 _Alll_E _A|7_r_g 1 2 3 4 5 _ Alli __ E _ Al?! _ E 1 2 3 4 _Alll__E __AI?J__E 1 2 4 _ Al?! _ E __ Am _ g 1 2 3 4 5 ___Allt___E ___Al?.l__E 1 2 3 4 5 _ Allt _ E _ Al?! __ E 1 2 3 4 5 _ Allt _ E __ Al?! _ E 1 2 3 4 5 _Ali’.t_E _Am__g 1 2 3 4 5 _Al?_t___E __Al?.t_E 1 2 3 4 5 _ Allt __ E __ Alll _ E 1 2 3 4 5 _A|?.t__E _A|7_r_p_l 47 Appendix C. Psychological Climate Strongly Strongly disagree agree 1. The jobs in this organization are clearly defined and 1 2 4 5 logically structured. 2. In this organization it is sometimes unclear who has 1 2 4 5 the formal authority to make a decision. 3. The policies and organizational structure of the l 2 4 5 organization have been clearly explained. 4. Red-tape is kept to a minimum in this organization. 1 2 4 5 5. Excessive rules, administrative details, and red-tape 1 2 5 make it difficult for new and original ideas to receive consideration. 6. Our productivity sometimes suffers from lack of l 2 4 5 organization and planning. 7. In some of the projects I’ve been on, I haven’t been 1 2 4 5 sure exactly who my boss was. 8. Our management isn’t so concerned about formal l 2 4 5 organization and authority, but concentrates instead on getting the right people together to do the job. 9. We don’t rely too heavily on individual judgment in 1 2 4 5 this organization; almost everything is double-checked. 10. Around here management resents your checking 1 2 4 5 everything with them; if you think you’ve got the right approach you just go ahead. 11. Supervision in this organization is mainly a matter of 1 2 4 5 setting guidelines for your subordinates; you let them take responsibility for the job. 12. You won’t get ahead in this organization unless you 1 2 4 5 stick your neck out and your subordinates; you let them take responsibility for the job. 13. Our philosophy emphasizes that people should solve l 2 4 5 their problems by themselves. 48 14. There are an awful lot of excuses around here when somebody makes a mistake. 15. One of the problems in this organization is that individuals won’t take responsibility. 16. We have a promotion system here that helps the best man to rise to the top. 17. In this organization the rewards and encouragements you get usually outweigh the threats and the criticism. 18. In this organization people are rewarded in proportion to the excellence of their job performance. 19. There is a great deal of criticism in this organization. 20. There is not enough reward and recognition given in this organization for doing good work. 21. If you make a mistake in this organization you will be punished. 22. Our business has been built up by taking calculated risks at right time. 23. Decision making in this organization is too cautious from maximum effectiveness. 24. Our management is willing to take a chance on a good idea. 25. A fiiendly atmosphere prevails among the people in this organization. 26. This organization is characterized by relaxed, easy- going working climate. 27. It’s very hard to get to know people in this organization. 28. People in this organization tend to be cool and aloof toward each other. 29. There is a lot of warmth in the relationships between management and workers in this organization. 30. You don’t get much sympathy from higher-ups in this organization if you make a mistake. 49 31. Management makes an effort to talk with you about our career aspirations within the organization. 32. People in this organization don’t really trust each other enough. 33. The philosophy of our management emphasizes the human factor, how people feel, etc. 34. When I am on a difficult assignment I can usually count on getting assistance from my boss and co- workers. 35. In this organization we set very high standards for performance. 36. Our management believes that no job is so well done that it couldn’t be done better. 37. Around here there is a feeling of pressure to continually improve our personal and group performance. 38. Management believes that if the people are happy, productivity will take care of itself. 39. To get ahead in this organization it’s more important to get along than it is to be a high producer. 40. In this organization people don’t seem to take much pride in their performance. 41. The best way to make a good impression around here is to steer clear of open arguments and disagreements. 42. The attitude of our management is that conflict between competing units and individuals can be very healthy. 43. We are encouraged to speak our rrrinds, even if it means disagreeing with our superiors. 44. People are proud of belonging to this organization. 45. I feel that I am a member of a well functioning team. 50 46. As far as I can see, there isn’t very much personal 1 2 3 4 5 loyalty to the company. 47. In this organization people pretty much look out for l 2 3 4 5 their own interests. Note. Items 2, 4, 6, 7, 9, 14, 15, 16, 19, 20, 23, 25, 32, 40, and 46 were recoded. Items 24, 29, 31, 33, 35, 38 and 44 were used for Affective tone and 11, 12, 13, 40, and 46 were used for Responsibility. 51 Appendix D. Korean Version of Psychological Climate HHSa 594817“ PIE-Cl Oil-T- 5943121 1. filal ilAlOIIAi-E Zr ?J$7l #951, 521442; 4E£l017t1 titl. 1 2 2. ital 31410114115- EIHUJIE -'7-7l £9.13 711m 7.15% 81101 £113 SIEJXI italic! PIE 75“?” EM. 3. 5FBI SIM-E ’Slfliollkl 2W 2334311 353-! 41210" Elltll 5311817“ WEiECl. 4. $31 EIAHE @5151 332F918: 71533 £1.45: iii-Klimt. 5. 31153 111-831. 12.11?l HIE/£1901 Ile‘El Tau-*4. Elia gélfisfl [111801 ’élt'tl Ololtl017l 1345M Git-.151. 6. $El ilklollkl-E 25:15 Eat-ill 71m ‘FEEE 71E.1 S3210! ‘EOiXI-E @5127} ElEl. 7. ”Di 015101 £101.33 EE’HIEOIM‘I. 5‘71 L121 SMPJXI 23E 7é$7l 212161. 3. 42a: sine 5444421 44:14 use me ékltlxl an, are 44215 mass 9.152011 eels-1L: zdor 525:1. 9. eat 214112 7117110401 sworn are ass was 7101 us new 912. 25 use Dial AliliOll 944w OIIFOlEIEl. 10. SEE-1| ENE LlQ-l E’Sflfi EEfilEf-l. é. Lll7l £511 EJECIPJ. :lEHE flfiflfi Elli. 11. ea: iMlOllkl 3W” are are 2:2 51813419358 are means a e 72011. 4131215501 XitlQJ 9.1501 can 811218 act. 12. $21 EIAIOlIAiE 21%| Efltlalfi 2M9} lr‘ISI’SlinOl was él-‘Ptllolettl. 3: 7H7HE|0| 21$01l Ellallkl 819.1% Eltl. 13. $El 21419.1 mama Eli's ¢¢§7l X15594 2155011 an Ext! areas 7201:}. 14. $El filAloilklE fl-T-Oll Bil fl-i—I flfilfll. Hflifi- PIE- JEMII 9E1. 52 15. $31 ilAI’Q-I Exfl’élé filLl-E 34$ 7H7HE'OI $119.48- XIXI 91-35115 7101121. 16. $El EWE £11.91 HEOI MINIW Elli)0l(7l) ‘fi 4‘- Elt‘f 85' HIE? 71x13. 21121. 17. $El EWOIIH-E 2811 721217} 885422 elem Blatant EAR-BEL 18. $Bl ilAlOIIkIE ’Slfl’éol 9.151145%” BIBllfllkl EQOI OWNER 19. $al ilAloflAiE 215-0" as Eta HIEOI 212:1. 20. $3! ilAlollkl-E I83 SUE-$23011 CHE 3E3 EQOM PJ’SOI 21121. 21. ital Elkloilkl-E E191 LH7l Oi'r‘i Eli-I ElElPl'. 1011 El! tl'fl-S EQE ZiOICl. 22. ital ENE HES A|7|01| NINE HE‘S $15311 7M8 $121121. 23. SIZE-1| ilklolllell RIM @298 ENE—9.1 lifi'i’iollkl Li-T- E-T-Eiolfil (ZEéltlEl). 24. $31 £1415 732138 Quilt:- EE OlOICIOiI Eoli‘fl §U|7l Eloi‘RlEl. 25. $EI ilAlOlIAiE £85001! {'39} E$l7|7l Elfitltl. 26. $El 3MP:- E'PJSLT'. -291 21513238 71MB Slit 27. seat swarm-e are 51555 27171 are OiiiCl. 28. $El 1W HES-E AiEOlIHI tlZII'. THESE}. 29. $al ilkloilklf @9333} Elfllloll mitt B74l7l $235101 22th 30. 2191 BMOI all-“Fl ElElE. 2W @fiflgi lli‘E-l E38 5238 EII E21 710M. 31. @2138 team B’t'fll 34781—11 512131 Eliot Elltloi 0|0|=7| £14“- 225! :fitltl. 32. $el ilklNIHE Erase 412: we seem steer. 53 33. @‘é‘lfil—Ql fig halite 9.1334 9.3.: (’S-lfliol 01%‘71! :DIEXI) I 2.1253111. 34. L117} 01311? 215FI 4min LlE LL91 SAM—l EESE $51 as 538 E92 4‘- 2151. 35. -?-E'.| IWOIIHE 11$ ¢§01| CHEl 7l§0l 012:5 i111. 36. 7.321135% a {FEE—l EEEIE $12! 7Hfl’élol 2111111 EEC}. 37. SPEI ENGINE 75455501 Be; 7112134 3594 21$¢§S 7lltlfllol= E15115 9.134% 221:1. 38. 73151518- ElfliOl flfifi‘fi gtlfii §7lil halaln “Afr-l. 39. -?-BI ilklollkl @88184'6, 21$ 4%3 El tl-L': 3351f} ’Slflifll {l IILHE EL'OI E-l 52351. 40. $El ilAlOilklE 842453 ZlXIP-l Sig-PEN 12113101 Xl'i‘éli i3 7le7“ 918 712§ EE'F-l. 41. fizal EMMA-l i8 Pres Evil-:- 7151 :3 title 215.! EEOILI 9m EfliIOll/H ‘50le SEA-E- ’AOICl. 42. E‘EE'E- ’g’gflfiloll 22E ‘i‘HflQ—l 8513 012? EBB-“F (¥§¢) 21'le EHEI 711m 2151. 43. 54858- t‘dkl 2141521 5943170 9171! 51111315 XH‘JQ-l 946% E1815! Z‘lEi‘él-EEI. 44. 254E58- ilkloil #31011 Eloikl Xlfiléfifll :EEI. 45. LP: it 501711? E194 T’S‘Z‘lolP—lfl :77'El. 46. Lll7l £7l0flt- ’Slfliol ilklollfilltl HE'S-1&1 OHAIEIQ-l ii 21C}. 47. $EI 1141011411:- 5—1858 €le Kiel—91 0l9-loil 12118101 4591310101 ElCl. Appendix E. Job Satisfaction Strongly Strongly disagree agree 1. I am satisfied with being able to keep busy all the 1 2 5 time. 2. I am satisfied with the chance to work alone on the 1 2 5 job. 3. I am satisfied with the chance to do difl‘erent things 1 2 5 from time to time. 4. I am satisfied with the chance to be “somebody” in 1 2 5 the community. 5. I am satisfied with the way my boss handles his (her) 1 2 5 workers. 6. I am satisfied with the competence of my supervisor 1 2 5 in making decisions. 7. I am satisfied with being able to do things that don’t 1 2 5 go against my conscience. 8. I am satisfied with the way my job provides for l 2 5 steady employment. 9. I am satisfied with the chance to do things for other 1 2 5 people. 10. I am satisfied with the chance to tell people what to l 2 5 do. 11. I am satisfied with the chance to do something that 1 2 5 makes use of my abilities. 12. I am satisfied with the way company policies are put 1 2 5 into practice. 13. I am satisfied with the amount of work I do. 1 2 5 14. I am satisfied with the freedom to use my own 1 2 5 judgment. 15. I am satisfied with the chance to try my own 1 2 5 methods of doing the job. 16. I am satisfied with the working conditions. 1 2 5 17. I am satisfied with the way my coworkers get along 1 2 5 with each other. 55 18. I am satisfied with the praise I get for doing a good 1 2 3 4 5 job. 19. I am satisfied with the feeling of accomplishment I get fi'om the job. 1 2 3 4 5 20. I would recommend this job to a friend. 1 2 3 4 5 21. I find real enjoyment in my job. 1 2 3 4 5 22. I like my job better than the average worker does. 1 2 3 4 5 23. I am seldom bored with my job. 1 2 3 4 5 Note. Items 2, 6, 9, 13, 17, 19, and 23 were recoded. 1, 4, 5, 9, 12, 14, 15, 17, 18, 19, 20, and 21 were used for the final analyses. 56 Appendix F. Korean Version of Job Satisfaction Elllfi2 594817“ 9115121 DH? 5521th L115 243 ulE’AOIl Elltlol new. A O 1 2 2 L115 fi‘i‘l EX} 45%? 7l£l011 Ellfil 91:81“ ° Elli. E... 1 2 3. L115 71%} Cl! 24% Slit as 7I£l°l E11811 Elitltl. 4. Lit-:- llkloikl $215.1 HEIDI E4“- EJE 7|£l°l Elltll Elfitltl. 5. Lite L194 84171 ‘Ffl’slflifi CHE 2151011 EH3" Elfiflf-l. 6. Ll-E L191 Ski-9| 9M1 HE is?!“ Ellfiil 3:317" 9115121. 7. LHE L194 HEM OlELlE 2.1% Sill 9.1015 BE 710“ 215311. 8. LIE LlQ-l ’SI‘EJOll/H II‘N—lfi' 318-3 HEMP:- EM—loll 11113104 9.1313121. 9. L-l-E Cl! Alili $1311 21$ it 4‘- EI'E 7l£l01| EH‘SlOi Elitlxl HEEL 10. Ll-E- El! Aliloilfll LIE-l fi‘POll 12118101 931’ 7|£l°ll EH3" E15311. 11. Lit- Ll2l 554% mm 4‘- 21'5— 2.1% it 4‘- 2115 715M Elltll 91:511. 12. Lite 114171 E’ENE flfltlE 9’40! Ellflll Elitltl. 13. LIE LlQ-l 91$9—l SID-ll EHthi 9.13581“ EEC}. 14. LIE LlP—l 398 418?! ¢ 211‘: X1801 ”13101 57 PJ‘EIEI. 15. Ll-E L1 xltl 259.1 ears 1.15011 41821 4 2115 715101 c1501 astral. 16. Lil-:- 81? £17.30" CH8" 3352151. 17. L11? Sillfiiol HE OiEEl-E Elfloll 12113101 E15531“ PIE-El. 18. L115 L117} 8.1-TI El 45292122111 ELIE- ’S'EIOlI Elltloi Elaifitl. 19. L115 fi‘PP-l 3H3 EH SE :Eoil E13817“ 9115131. 20. Lite Lll Elite til-1101711 $5.1 it 7201121. 21. LIE- Lll 342101111 ERIE E7128 :771El. 22. L115 13834 513212111 Lll $918 Ci 55019151. 23. 1415 L11 51212 1:14:31 51:1. 58 Appendix G Expected Correlations from Hunter’s Confumatory Factor Analysis (Lower Triangle) and Deviations between Observed Correlation and Expected Correlation (Upper Triangle) 25 31 33 35 37 40 47 1 1 12 13 r_42 r_49 25 .01 -.01 -.02 -.01 .12 -.09 -.07 .10 .07 -.06 -.01 31 .42 .03 .02 .02 -.04 -.02 -.10 -.15 -.13 -.16 .12 33 .39 .42 -.04 -.03 -.05 .12 -.06 .05 -.02 .04 .13 35 .43 .46 .43 -.03 .06 .03 .01 -.07 -.07 -.06 .08 37 .44 .47 .43 .48 .02 .03 .04 -.01 .02 -.16 -.07 40 .39 .42 .38 .43 .43 -.09 -.13 -.05 .03 -.19 -.05 47 .38 .41 .38 .42 .43 .38 .18 .05 .07 .19 .38 11 .22 .23 .22 .24 .24 .22 .21 .14 .05 -.13 -.08 12 .23 .24 .22 .25 .25 .22 .22 .36 .11 -.11 -.13 13 .28 .30 .29 .31 .31 .28 .27 .45 .46 -.08 -.08 r_42 .23 .25 .23 .25 .26 .27 .22 .37 .38 .47 .30 r_49 .20 .22 .20 .22 .23 .20 .20 .32 .34 .42 .34 59 REFERENCES Abdel-Halim, A. (1981). Effects of role stress-job design-technology interaction on employee work satisfaction. Academy of Management Journal, 24, 260-273. Argyris, C. (1957). Personality and Organization, New York: Harper Collins. Aronson, E., & Cope, V. (1968). My enemy’s enemy is my fiiend. Journal of Personality & Social Psychology, 8, 8-12. Bedian, A. G, & Armenakis, A. A. (1981). A path analytic study of the consequences of role conflict and ambiguity. Academy of Management Journal, 24), 417-424. Boyd, N. G, & Taylor, R. R. (1998). A developmental approach to the examination of fiiendship in leader-follower relationships. Leadership Quaterly, 9, 1—25. Brass, DJ. (1981). Structural relationships, job characteristics, and worker satisfaction and performance. Administrative Science Quarterly, 26, 331-348. Bridge, K., & Baxter, LA. (1992). Blended relationships: Friends as work associates. Western Journal of Communication, 56, 200-225. Brief, A. P., & Roberson, L. (1989). Job attitude organization: An exploratory study. Journal of Applied Social Psychology, 19, 717-727. Byme, D., & Clore, G L. (1970). A reinforcement model of evaluative responses. Personality: An International Journal, I, 103-128. Cheloha, R., & Farr, J. (1980). Absenteeism, job involvement, and job satisfaction in an organizational setting. Journal of Applied Psychology, 65, 467-473. Cohen, S. J. (1995). Broadening the climate change debate. Ecodescision, 17, 34-37. Corbett, J. M., Marti, R., Wall, T. D.,& Clegg, C. W. (1989). Technological coupling as a predictor of intrinsic job satisfaction: A replication study. Journal of Organizational Behavior; 10, 91-95. Cranny, C. J., Smith, P. C., & Stone, E. F. (1992). Job satisfaction: How people feel about their jobs and how it aflects their performance. New York, NY: Lexington Book. Cross, S. E., Gore, J. S., & Morris, M. L. (2003). The relational-interdependent self- constral, self-concept consistency, and well-being. Journal of Personality & Social Psychology, 85, 933-944. 60 Curry, T. J., & Emerson, R. M. (1970). Balance theory: A theory of interpersonal attraction? Sociometrjy, 33, 216-238. Davis, K. E. (1985). Near and dear: Friendship and low compared. Psychology Today, 19, 22-30. Downey, K., Hellriegel, D., Phelps, M., & Slocum, J. (1974). Organizational climate and job satisfaction: A comparative analysis. Journal of Business Research, 2, 233- 248. Duchon, D., Green, S. G, & Taber, T. D. (1986). Vertical dyad linkage: A longitudinal assessment of antecedents, measures, and consequences. Journal of Applied Psychology, 71, 56-60. Duck, S. W. (1983). Friends for life: The psychology of close relationships. New York: St. Martin’s. Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Orlando, FL: Harcourt Brace Edwards, J. R. (2001). Ten difference score myths. Organizational Research Methods, 4, 265-287. Epitropaki, O., & Martin, R. (1999). The impact of relational demography on the quality of leader-member exchanges and employees’ work attitudes and well-being. Journal of Occupational and Organizational Psychology, 72, 237-240. Fang, C. F., & Kenrick, D. T. (2002). Interpersonal relations and group processes- repulsion or attraction? Group membership and assumed attitudes similarity. Journal of Personality & Social Psychology, 83, 111-125. F estinger, L., Schachter, S., & Back, K. (1950). Social pressures in informal groups. New York: Harper and Row. Festinger, L. (1957). A theory of cognitive dissonance. Stanford: Stanford University Press. Fleming, R., & Baum, A. (1986). Social support and stress: The buffering effects of fiiendship. In V. J. Derlega & B. A. Winstead (Eds.), Friendship and social interaction (pp. 207-226). New York: Springer-Verlag. Folger, R., & Greenberg, J. (1985). Procedural justice: An interpretive analysis of personnel systems. Research in Personnel and Human Resources Management, 3, 141-183. 61 Fryxell, G E., & Gordon, M. E. (1989). Workplace justice and job satisfaction as predictors of satisfaction with union and management. Academy of Management Journal, 32, 851-866. Glisson, C., & Durick, M. (1988). Predictors of job satisfaction and organizational commitment in human service organizations. Administrative Science Quarterly, 33, 61-81. Graber, D. A. (2003). The power of communication: Managing information in public organizations. Washington, DC: CQ press. Graen, G, Novak, M., & Sommerkamp, P. (1982). The effects of leader-member exchange and job design on productivity and satisfaction: testing a dual attachment model. Organizational behavior and human performance, 30, 109- 1 3 1 . Guilford, J. P. (1954). Psychometric methods. New York: McGraw-Hill. Hackrnan, J. R., & Oldham, G R. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 55, 259-286. Hackrnan, J. R., & Oldham, G R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Decision Processes, 16, 250-279. Hackrnan, J. R., & Oldham, G R. (1980). Work redesign. Reading, MA: Addison-Wesley. Hamilton, R. F., & Wright, J. D. (1986). The state of the Masses. New York: Aldine. Hammer, M. R., Gudykunst, W. B., & Wiseman, R. L. (1978). Dimensions of intercultural effectiveness: An exploratory study. International Journal of Intercultural Relations, 2, 382-383. Hays, R. B. (1988). Friendship. In S. W. Duck (Ed.), Handbook of personal relationships (pp.39l-408). New York: Wiley. Heider, F. (1958). The psychology of interpersonal relations. New York: John Wiley. Hellriegel, D., & Slocum, J. W. Jr. (1974). Organizational Climate: Measures, Research and Contingencies. Academy of Management Journal, I 7, 255-280. Hofinann, D. A., & Stetzer, A. (1996). A cross-level investigation of factors influencing unsafe behaviors and accidents. Personnel Psychology, 49, 307-339. Insko, C. A. (1981). Balance theory and phenomenology. In R. E. Petty, T. M. Ostrom & T. C. Brock (Eds.), Cognitive responses in persuasion (pp.309-338). Hillsdale, NJ: Erlbaum. 62 Ironson, G H. (1992). Job stress and health. In C. J. Cranny, P. C. Smith & E. F. Stone (Eds.), Job satisfaction (pp.219-240). New York: Lexington Books. James, L., & Jones, A. (1974). Organizational climate: A review of theory and research. Psychological Bulletin, 81 , 1096-1 1 12. Johnson, M. S. (1989). Variables associated with friendship in an adult population. The Journal of Social Psychology, 129(3), 379-390. Jordan, N. (1953). Behavioral forces that are a function of attitudes and of cognitive organization. Human Relations, 6, 47-56. J Oreskog, K. G. (1993). Testing structural equation models. In K. A. Bollen & J. S. Lang (Eds.), Testing structural equation models (pp. 294-316). Newbury Park, CA: Sage. Joyce, W. F., & Slocum, Jr., W. M. (1984). Collective climate: Agreement as a basis for defining aggregate climates in organizations. Academy of Management Journal, 27, 721-742. Katzell, R. A., Thompson, D. E., & Guzzo, R. A. (1992). Howjob satisfaction and job performance are and are not linked. In C. J. Cranny, P. C. Smith & E. F. Stone (Eds.), Job satisfaction (pp.195-218). New York: Lexington Books. Kilduff, M., & Krackhardt, D. (1994). Bringing the individual back in: A structural analysis of the internal market for reputation in organizations. Academy of Management Journal, 3 7, 87- 1 08. Kopelrnarr, R. E., Brief, A. P., & Guzzo, R. A. (1990). The role of climate and culture in productivity. In B. Schneider (Ed.), Organizational climate and culture (pp.282-318). San Francisco, CA: J ossey—Bass. Kozlowski, S., & Hults, B. M. (1987). An exploration of climates for technical updating and performance. Personnel Psychology, 40, 539-563. Kozlowski, S., 8; Doherty, M. (1989). Integration of climate and leadership: Examination of neglected issue. Journal of Applied Psychology, 74, 546-553. Krackhardt, D., & Porter, L. W. (1985). When fiiends leave: A structural analysis of the relationship between turnover and stayers’ attitudes. Administrative Science Quarterly, 30, 242-261. Lambert, S. J. (1991). The combined effects of job and family characteristics on the job satisfaction, job involvement, and intrinsic motivation of men and women workers. Journal of Organizational Behavior; 12, 341-363. 63 Lee, J. (1998). Effective maintenance communication in superior-subordinate relationships, Western Journal of Communication, 62, 181-208. Lee, J. (1999). Leader-member exchange, gender, and members’ communication expectations with leaders. Communication Quarterly, 47, 415-429. Lee, J., & J ablin, F. (1995). Maintenance communication in superior-subordinate work relationships, Human Communication Research, 22, 220-257. Lincoln, J. R., & Miller, J. (1979). Work and friendship ties in organizations: A comparative analysis of relation networks. Administrative Science Quarterly, 24, 181-199. Litwin, G, & Stringer, R. (1968). Motivation and organizational climate. Cambridge, MA: Harvard University Press. Locke, E. A. (1976). The nature and causes of job satisfaction. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp. 1297-1350). Chicago: Rand Mcnally. Markiewicz, D., Devine, 1., & Kausilas, D. (2000). Friendships of women and men at work: Job satisfaction and resource implications. Journal of managerial Psychology, 15, 161-201 . Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50, 370-396. Money, R. B., & Graham, J. L. (1999). Salesperson Performance, pay, and job satisfaction: Tests of a model using data collected in the United States and Japan. Journal of International Business Studies, 30, 149-172. Monge, P. R., & Contractor, N. S. (2001). Emergence of communication networks. In F. M. J abiln & L. L. Putnam (Eds.), The new handbook of organizational communication: Advances in theory, research, and methods (pp.732-818). Thousand Oaks, CA: Sage. Monge, P. R., & Contractor, N. S. (2003). Theories of communication networks. New York: Oxford university press. Muchinsky, P. M. (197 7). Organizational communication: Relationships to organizational climate and job satisfaction. Academy of Management Journal, 20, 592-607. Mueller, B., & Lee, J. (2002). Leader-member exchange and organizational communication satisfaction in multiple contexts. Journal of Business Communication, 39, 220-244. 64 Muraskin, J., & Iverson, M. A. (195 8). Social expectancy as a function of judged social distance. The Journal of Social Psychology, 48, 11-14. Naumann, E. (1993). Organizational predictors of expatriate job satisfaction. Journal of International Business Studies, 21, 61—81. Newcomb, T. (1953). An approach to the study of communicative acts. Psychological Review, 60, 393-404. O'Keefe, D. J. (2002). Persuasion: Theory and research. Thousand Oaks, CA: Sage. Osgood, C. E., & Tannenbaum, P. H. (1955). The principle of congruity in the prediction of attitude change. Psychological Review, 62, 92-155. Oshikawa, S. (1970). Consumer pre-decision conflict and post-decision dissonance. Behaviral Science, 15, 132-140. Park, H. S., & Boldman, J. (1998, November). The relationship between similarity of attitudes toward issues and activities and attraction in fiiendship. Paper presented at the annual convention of National Communication Association. New York, NY. Patterson, M., Payne, R., & West, M. (1996). Collective climates: A test of their sociopsychological significance. Academy of Management Journal, 39, 1675- 1691. Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and persuasion: Classic and contemporary approaches. Dubuque, Iowa; Wm. C. Brown. Pollock, T. G, Whitbred, R. C., & Contractor, N. S. (2000). Social information processing and job characteristics: A simultaneous test of two theories with implications for jab satisfaction. Human Communication Research, 26, 292-330. Porter, L. Lawler, E., & Hackrnan, R. (1975). Behavior in organization. New York: Mc Graw Hill. Ratiu, I. (1983). Thinking internationally: A comparison of how international executives learn, International Studies of Management and Organization, 13, 139-150. Redding, W. C. (1972). Communication within the organization: An interpretive review of theory and research. New York: Industrial Communication Council. Reichers, A., 8c Schneider, B. (1990). Climate and culture: An evolution of constructs. In B. Schneider (Ed.), Organizational culture and climate (pp.5-39). San Francisco, CA: Jossey—Bass. 65 Riordan, C. M., & Griffeth, R. W. (1995). The opportunity for fiiendship in the workplace: The underexplored construct. The Journal of Business and Psychology, 10, 141-154. Rizzo, J ., House, R. & Litzman, S. (1970). Role conflict and ambiguity in complex organizations. Administrative Science Quarterly, 15, 150-163. Roberts, K.H., & O’Reilly, GA. 111. (1979). Some correlates of communication roles in organizations. Academy of Management Journal, 22, 42-57. Rogers, E. M., & Kincaid, D. L. ( 1980). Communication networks: Toward a new paradigm for research. New York: The Free Press. Salancik, G R., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23, 224-253. Schneider, B. (1975). Organizational climate: Individual preferences and organizational realities revisited. Journal of Applied Psychology, 60, 459-465. Sias, P. M., & Cahii, D. J. (1998). From coworkers to fiends: The development of peer fredships in the workplace. Western Journal of Communication, 62, 273-299. Smith, P. C. (1992). In pursuit of happiness. Why study general job satisfaction? In C. J. Cranny, P. C. Smith & E. F. Stone (Eds.), Job satisfaction (pp.5-20). New York: Lexington Books. Solano, C. H. (1986). People without friends: Loneliness and its alternatives. In V. J. Derlega & B. A. Winstead (Eds.), Friendship and social interaction (pp. 227- 246). New York: Springer-Verlag. Suh, E. M. (2002). Culture, identity, consistency, and subjective well-being. Journal of Personality and Social Psychology, 83, 1378-1391. Watson, D., Hubbard, B., & Wiese, D. (2000). General traits of personality and affectivity as predictors of satisfaction in intimate relationships: Evidence from self- and partner-ratings. Journal of Personality & Sales Management, 68, 413- 449. Watson, D., & Slack, A. K. (1993). General factors of affective temperament and their relation to job satisfaction over time. Organizational Behavior and Human Decision Process, 54, 181-202. Wayne, S.J., Shore, L., & Liden, R. (1997). Perceived organizational support and leader- member exchange: a social exchange perspective. Academy of Management Journal, 40, 82-111. 66 Weick, K. E. (1969). The social psychology of organization. Reading, MA: Addison- Wesley. Weiss, D. J ., Dawis, R. V., England, G W., & Lofquist, L. H. (1967). Manual for the Minnesota Satisfaction Questionnaire. Minnesota studies in vocational rehabilitation, 45. Weiss, H. M., Nicholas, J. P., & Daus, C. S. (1999). An examination of the joint effects of affective experiences and job satisfaction and variations in affective experiences over time. Organizational Behavior and Human Decision Processes, 78, 1-24. Werner, C., & Parrnelee, P. (1979). Similarity of activity preferences among friends: Those who play together stay together. Social Psychologi Quarterly, 42, 62-66. Wexley, K. N., Alexander, R. A., Greenawalt, J. P., & Couch, M. A. (1980). Attitudinal congruence and similarity as related to interpersonal evaluations in manager- subordinate dyads. Academy of Management Journal, 23, 320-330. Winstead, B.A., Derlega, V. J ., Montgomery, M. J ., & Pilkingston, C. (1995). The quality of fiiendships at work and job satisfaction. Journal of Social and Personal Relationships, 12, 199-215. Woodside, A. G, & Chebat, J. (2001). Updating Heider’s balance theory in consumer behavior: A Jewish couple buys a German car and additional buying-consuming transformation stories. Psychology and Marketing, 18, 475-495. Wright, P. H. (1969). A model and a technique for studies of friendship. Journal of Experimental Social Psychology, 5, 295-309. Wright, P. H. (1987). The acquaintance description form. In S.W. Duck & D. Perlrnan (Eds.), Sage Series in Personal Relationships: Vol. I. Intimate relationships: Development, dynamics, and deterioration (pp.39-62). London: Sage. Wright, R. Jr., King, S. W., Berg, W. E., & Creecy, R. F. (1987). Job satisfaction among Black female managers: A casual approach. Human Relations, 40, 489-506. Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology Monograph Supplement, 9, 1-27. Zajonc, R. B., & Burnstein, E. (1965).The learning of balanced and unbalanced social structures. Journal of Personality, 33, 153-163. 67 Table 1. The Effects of Climate Similarities with the Best F riend(s) on Job Satisfaction. B SE. ,6 t Affective tone 0.054 0.140 .052 0.387 Responsibility -0.205 0.127 -.215 -1.618 Type of - 0.041 0.124 -.O43 -0.327 organization Affective tone x 0.207 0.298 .152 0.694 Type of organization Responsibility x 0.053 0.261 .038 0.205 Type of organization F (3, 59) = 0.968, p =.414, adjusted R2 =— .002 Fchang, (2, 57) = 0.330, p = .720, Ram. = .011 Note. * p < .05 68 _._. .. .. .... ..__ m r- .m .. was Table 2. The Eflects of Climate Similarities with the Friends on Job Satisfaction. B SE. B t Affective tone -0.062 0.176 -.045 -0.352 Responsibility —0.379 0.157 -305 2413* 1““ Type of 0.091 0.109 .097 0.838 organization Affective tone >< 0.150 0.360 .084 0.416 j Type of organization Responsibility x 0.308 0.336 .207 0.915 Type of organization F (3, 68) = 2.878, p <05, adjusted R2 = .074 F.,.,,. (2, 66) = 0.708, p = .496, Ram, = .019 Note. * p < .05 69 Table 3. The Effects of Climate Similarities with the Weighted Friendship on Job Satisfaction. Affective tone Responsibility Type of organization Affective tone x Type of organization Responsibility X Type of organization B SE. B t 0.108 0.231 .065 0.467 -0.215 0.214 -.132 -1.002 -0.004 0.117 -.005 -0.038 -0.321 0.483 -.131 -0.664 -0.213 0.445 -.085 -0.478 F (3, 74) = 0.335, p = .800, adjusted R2 = -027 Fem, (2, 72) = 0.621, p = .540, chhang. = .017 Note. * p < .05 70 111111111111111111111111111 3