...\..~-~\‘-~ Va. 4:4”!435111‘CX ~.. .~ - \ .4? ....;..~‘. ‘ if. ‘ _-..\ - ,xa...~.~. ~‘<~ 7'~¢;; 4:1..1- .. ..:i_ “A“. new“ x». .- - nwc.’ - J-'\'.. ' . u :' x u r“~4 .v \ ‘ ‘ kw... u 1. .' L<~ ~'6J “NE K T». .nnn a" u:— ~ . «fig .n. cu 44‘: <~ 2.. . ‘J‘. are; l 'x.\ .1 I ,3 {L , :r . .. ‘2’ ‘r " "rl-Lyrf l_ 1-" .. - .,’ '1. , j. x- {a'iut'r‘nll . - run: - rv-r . (- ,.. ., -,(w . , I. W. .. , ',,,,.,.,,.,4. . ,, ..,, . .1 1 , ,.,.- ,Jyu run I r“" 1"“ - .‘ .'1. u. :u 1‘,...,.,. w,“ .1, ,..,.,y .Hp. ,u wt v 1 .2 .1 ”.1: ~ n‘ ".1. .|., “My, ll ; ’v -4 “nan. ( nu. . “n... on u: .m‘i 'Ir; .,,.. . I .,.,..: ,- mm, '1' '1‘ “N” ; , .::.,. ’ <“r.r_» ..,... 1 . ,x. V . mum, q. -~ . . - k ,. 4. ' ,n 4 ha...- . 0.1-» ..u\r~:.1.1.'1u- . ’\ max k~ . u \n: hf: ~lk" ,, «.1 ”:4" M ‘ It; . "‘“ia‘ g, 1! m J . , .:=. 4-. 7!“:IIJ' Amman-n 11‘1-1 ".1 ,. ,5 I m, 4. ",l’ k, ,a-Avw (3‘15 '-!}J.I run ; v n 'J-‘\|I;~, .u/ N , ,. a“ . .' u“... .. . «4‘ M.» a.“ M. m um. Ill— ‘xv mgw' ‘ 4x A , 4'. ., ., '1‘. “I r': ' ‘vf‘vuv‘ um h.» ,rgr. - 1 ! .,.,.,... ,'..- U... ’ , J .r r ”Mm, - 5"” "I ”’4‘. ..~ ‘ ,~~,.., . ,_. ., . .uw -' . ., 'r 4 1315.333 mllHlllllWlllllll 31293 00897 5157 This is to certify that the dissertation entitled CROSS -CULTURAL VARIAT IONS IN IMPLICIT THEORIES OF REQUESTING BEHAVIOR presented by Min-Sun Kim has been accepted towards fulfillment of the requirements for Ph.D. . Communication degree in Major professor Date 5/9/92 MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 *——— LIBRARY Mlchlgan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE ' 35918202 MSU Is An Affirmative Action/Equal Opportunity institution c:\circ\datedm.pm3-p.1 CROSS-CULTURAL VARIATIONS IN IMPLICIT THEORIES OF REQUESTING BEHAVIOR By Min-Sun Kim A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1992 57% 4/04/ ABSTRACT CROSS-CULTURAL VARIATIONS IN IMPLICIT THEORIES OF REQUESTING BEHAVIOR By Min-Sun Kim While numerous studies have attempted to describe cultural preferences in communication tactics and strategies, in many cases the origins of those preferences are obscure. To test whether different cultural groups develop distinct beliefs about requesting behavior, first, a classification scheme of request tactics is established in terms of 12 mutually exclusive main tactics that vary along several contextual and syntactic dimensions. Then, the following five interactive constraints are identified based on the past literature and a series of pilot tests: (1) Concern for clarity; (2) Concern for effectiveness; (3) Concern for avoidingihurting the hearer's feelings; (4) Concern for nonimposition; and.(5) Concern for avoiding negative evaluation by the hearer. The primary aim of this dissertation is to compare the ways in which five interactive constraints are perceived and rated across cultures, and to trace the possible links between the perceptions of interactive constraints and the perceived likelihood of using specific tactics. The participants are a total of 595 undergraduates: 296 Koreans (native speakers of Korean) and 299 Americans (native American English speakers) studying in their respective countries. After being presented with one of six request situations, participants rate the perceived importance of each constraint. Then they evaluate one of the two different sets of 12 exemplar tactics along the five dimensions of interactive constraints as well as for likelihood of ii “1‘7...” Pl. use. The main findings of this study point to a picture that is generally consistent with the enduring cultural generalizations about request styles. Specifically, the results of this study seem to suggest different processes of arriving at goals: Americans as focusing on task constraints (conveying the message clearly and efficiently) and Koreans focusing on social-relation constraints (avoiding damage to the relationship or loss of face by the bearer). The research presented in this dissertation extends our knowledge about what kinds of general interactive constraints shape peoples’ beliefs on interaction, and how interactive constraints are anchored in the wider cultural milieu. P, To my brothers and sisters, my parents and grandmother for their love iv ,L' ACKNOWLEDGMENTS This is much more than a simple acknowledgment of the people who helped me in my dissertation project directly or indirectly. I owe each one of them genuine thanks for many different reasons. To begin with, I would like to thank Dr. Steven Wilson and Dr. Mary Bresnahan, the co-chairs of my dissertation committee. Throughout my graduate education, Dr. Wilson strongly influenced my professional development. I will always remember his wonderful guidance and encouragement. Besides her insightful comments and guidance, Dr. Bresnahan brightened some of my dismal days in Michigan with her optimism and kindness. She led me through some of the most difficult hurdles in completing this dissertation. The other members of my committee, Drs. Gerald Miller, David Johnson, and John Hunter also greatly influenced this project. Dr. Miller has been a source of inspiration from the beginning of my Ph.D. program. Dr. Johnson provided many constructive comments and useful criticism. Dr. Hunter spent countless hours with me throughout my years at Michigan State. I can never forget his generosity and remarkable character. I would like to thank every single one of the professors, staff, and graduate students, whom I had the privilege of knowing in the Department of Communication at Michigan State University. In particular, Sandy Starnaman, who was my office-mate for four years and shared good and bad times, deserves special mention. Narayan Raja, whom I met during the FTA (foreign teaching assistant) training program, has. helped me in every respect. I am indebted to Dr. Hyun Joo Kim, now a professor at Choong-Nam University, who helped me with collection of Korean Data. University of Hawaii offered me a job in 1990 when I was still an ABD. Since I arrived in Hawaii, my department provided an extremely supportive and friendly environment. I would like to thank several of my colleagues who supported me through the last stages of completing the dissertation: Drs. Kelly Aune, Krystyna Anne, Michael Basil, William Brown, Ronald Cambra, Michael Miller, Kazuo Nishiyama, Rodney Reynolds, Bill Sharkey, Ms. Bernadine Yim, and my favorite TA, Ron St. John. Living on the other side of the world are my brothers and sisters, to whom this dissertation is dedicated. They gave their youngest sister undue share of their love and support, after my parents passed away very early in my life. This dissertation is also dedicated to my parents and grandmother. I owe them everything. Vi TABLE OF CONTENTS CHAPTER Page I. CROSS-CULTURAL VARIATIONS IN IMPLICIT THEORIES OF REQUESTING BEHAVIOR INTRODUCTION ......................................... 1 Why Requests ................................... 4 Taxonomy of Request Tactics .................... 5 Characteristics of Interactive Constraints ..... 19 Contents of Interactive Constraint ............. 20 Concern for clarity ....................... 20 Concern for minimizing imposition ......... 21 Concern for avoiding hurting the hearer's feelings ................ 22 Concern for avoiding negative evaluation by others ................. 23 Concern for effectiveness ................. 24 Cross—Cultural Comparisons of Request Tactics along the Interactive Constraint Dimensions .... 26 Clusters of tactics ....................... 26 Placement of request tactics along the five dimensions ............ 27 Potential incompatibility among the interactive constraints .......... 28 Cross-Cultural Comparisons of the Importance of Interactive Constraints ..................... 29 The Perceived importance of interactive constraints .......................... 29 Effects of social status on the perceived importance of constraints ............ 30 Relative weights given to each constraint in predicting likelihood of use ...... 30 Intracultural Variability ...................... 31 Cultures Selected .............................. 32 Research Questions ............................. 32 CHAPTER II. METHOD .................................................. 34 Overview ....................................... 34 Participants .................................... 34 Request Situations ............................. 36 Pilot Tests .................................... 37 Realism of request situations ............. 37 Perceived social relationships and request size ......................... 38 Selection of Request Tactics ................... 38 Perceived Importance of Interactive Constraints. 41 Ratings of Request Tactics along the Dimensions of Interactive Constraints ......... 42 Assessment of Likelihood of Use ................. 46 Assessment of Individual Difference Variables .. 48 vii CHAPTER III. RESULTS ................................................ 51 Comparisons of the Perceptions of Request Situations ................................ 51 Results of the Main Study ...................... 56 Question l-A ..........., .................. 56 Question 1-B .............................. 72 Question 1-C .............................. 82 Question 2-A .............................. 84 Question 2-B .............................. 87 Question 2-C .............................. 96 Question 3-A .............................. 108 Question 3-B .............................. 115 CHAPTER IV. DISCUSSION AND CONCLUSION ................................ 119 Summary of Findings ............................ 119 Theoretical Implications: Cross—cultural Differences in the Implicit Theories of Requesting .................................... 124 Relationship between interactive constraints and perceived likelihood of use .......... 124 Salience of interactive constraints ....... 127 Practical Implications ......................... 130 Limitations and Directions for Future Research .134 Conclusions .................................... 135 BIBLIOGRAPHY ................................................. 137 APPENDICES ................................................... 154 viii LIST OF TABLES TABLE Page 1-a. Classification of Request Tactics .................... 6 l-b. Past Classification Schemes of Request/ Directive Tactics .......................... 10 2. Comparison of Demographic Characteristics of Korean and the U. S. Participants ....................... 35 3. Realism of Request Situations ........................ 39 4. Perceptions of Request Situations: Scale Items and Reliabilities ............................... 4O 5. Perceived Importance of Interactive Constraints: Scale Items, and Reliabilities .............. 43 6. Constraint Judgments of Utterances: Items and Reliabilities ............................... 47 7. Individual Difference Measures: Scale Items .......... 49 CD . Cross-cultural Comparisons of Perceived Request Size.. 52 9. Cross-cultural Comparisons of Perceived Social Status in Request Situations ............................ 54 10. Cross-cultural Comparisons of Perceived Social Distance in Request Situations ................... 55 11. Coefficient Alphas for Three Strategy Categories ..... 58 ll-a: Correlations among 12 Tactics (Clarity) .... 6O ll-b: Correlations among 12 Tactics (Feelings) ... 62 ll-c: Correlations among 12 Tactics (Imposition) . 64 ll-d: Correlations among 12 Tactics (Dislike) .... 66 ll-e: Correlations among 12 Tactics (Effectiveness) ................... 68 ll-f: Correlations among 12 Tactics (Likelihood of Use) ............... 7O 12. Ratings of Request Tactics 12-a: Clarity .................................... 73 12-b: Perceived Imposition ....................... 75 12-c: Consideration for the Other's Feelings ..... 77 12-d: Risking Disapproval for Self ............... 78 12—e: Effectiveness .............................. 8O 12-f: Likelihood of Use .......................... 81 13. Degrees of Incompatibility among Interactive Constraints and the Likelihood of Use l3-a: Korean Sample .............................. 83 13-b: U.S. Sample ................................ 85 14. Ratings of Perceived Importance of the Five Constraints ............................. 86 15. Descriptive Statistics for the Perceived Importance Ratings lS-a: Clarity .................................... 89 15-b: Concern for Feelings ....................... 9O lS-c: Avoiding Imposition ........................ 92 15-d: Avoiding Devaluation ....................... 93 lS-e: Effectiveness .............................. 95 ix l6. l7. 18. 19. 20. 21. 22. 23. Interactive Constraints and the Likelihood of Use: Regression Analysis without Effectiveness 16-a: Korean Participants ...................... 16-b: U.S. Participants ........................ Interactive Constraints and the Likelihood of Use: Regression Analysis without Effectiveness 17-a: Korean Participants ...................... l7-b: U.S. Participants ........................ Path Analysis of the Theoretical Model ............. Ratings of Individual Difference Variables across Cultures ....................................... Relationships among individual Difference Variables Correlation Analysis for Perceived Importance of Constraints and Individual Difference Variables Regression Analysis for Perceived Importance of Constraints and Individual Difference Variables Hierarchical Multiple Analysis for Perceived Importance of Constraints and Individual Difference Variables ........................... .. 98 .. 99 ..102 ..103 ..104 ..109 ..111 .112 .113 ..116 LIST OF FIGURES FIGURE Page 1. Theoretical Path Model: Relationships among the Interactive Constraints, Effectiveness, and Likelihood of Use ...... 101 2. Theoretical Path Model with Path Coefficient: Relationships among the Interactive Constraints, Effectiveness and Likelihood of Use .................................. 106 xi CHAPTER 1 CROSS-CULTURAL VARIATIONS IN IMPLICIT THEORIES OF REQUESTING BEHAVIOR INTRODUCTION Previous contrastive studies that establish cross-cultural differences in interaction styles have shown that speech communities share detectable patterns of language use (e.g., Tannen, 1981; Katriel, 1986; Rintell, 1981). Tannen (1981), for instance, shows that speakers of American English tend to be more direct than speakers of Greek. House and Kasper (1981) reveal a similar trend when comparing German with British English: German speakers tend to realize requests and complaints more directly than do English speakers. Blum-Kulka (1982, 1983) also notes that Hebrew speakers use higher levels of directness as compared to speakers of American English. Korean speakers also prefer to make directives in indirect ways in cases where American English speakers prefer direct ones (see Holtgraves & Yang, 1990). i It is very widely acknowledged that linguistic routines are embodiments of the socio-cultural values of speech communities that use them (i.e., Gumperz, 1978, 1982). However, many studies concentrate more on describing the use of routines rather than explaining the socio-cultural aspects of the meaning underlying the linguistic routines. It is the contention of this paper that research needs to go beyond descriptions to explanations of language use that are culturally and socially revealing. For instance, if such variations in the stylistic preference are indeed shown to exist, why are there those variations? To try to understand the origins of cultural preferences for l specific strategies, this paper proposes that people may possess culture-specific beliefs or "implicit theories" about request styles (see Steffen & Eagly, 1985; Jones, 1964; Rule & Bisanz, 1987; Wegner & Vallacher, 1977). The implicit theories portray the processes of social cognition shared by a cultural group. Communication researchers studying social cognition -- how people think about people -- have conducted research and developed formal theories about individuals' commonsense (or implicit) theories (e.g., implicit motivation, implicit personality theory, implicit social relations theory). For instance, Heider (1958) studies what he called the "naive psyhological theories" held by individuals. According to Rule and Bisanz (1987), a person's naive psychology about the relevant tactics used to achieve various interaction goals can be conceptualized as a social knowledge structure. As people seem to spend a large amount of time trying to get each other to do things, it is conceivable that they have implicit theories on which to base decisions about social influence tactics (see Wegner & Vallacher, 1977). A student who wants his/her professor to extend a homework deadline must decide whether to threaten, or reason with, or ingratiate. Cultural groups may develop culturally shared implicit theories that suggest not only which tactics are permissible, but also which tactics tend to produce results in each particular situations. As people hold a network of beliefs or implicit theories about people, themselves, and about the social situations they encounter, it is likely that people make requests on the basis of their conceptions of commonsense theories of requesting styles that are culturally preferred. Given that individuals regularly engage in influence attempts, tactical choices are no doubt guided by some implicit theoretical system. That is, many of the principles relevant to the perception of others also apply to the perceptions of requesting behavior. While people seldom reflect on the nature of their requesting behavior, a cultural group may share "commonsense theories" of requesting behavior. This dissertation explores how cultural groups may differ in the structure and content of their theories about requesting behavior, focusing on the perceptions and importance attached to interactive concerns in requesting situations. Although the implicit theories do not measure actual requests, people's perceptions of requesting behavior may provide insights about how requests are actually made. To test whether different cultural groups develop distinct beliefs about request styles, first, a classification scheme of request strategies is established in terms of 12 mutually exclusive main tactics that seem to be subsumed under three strategy categories (hints, query and direct statement). Then, the following five conversational constraints are identified based on the past literature and a series of pilot testing: (1) Concern for clarity; (2) Concern for effectiveness; (3) Concern for minimizing imposition; (4) Concern for avoiding hurting the hearer's feelings; and (5) Concern for avoiding negative evaluation by the bearer. The purposes of this study are: (1) to compare the ways in which five Interactive constraints are perceived and rated across cultures; and (2) to trace the possible links between these perceptions of interactive constraints and perceived likelihood of using specific request tactics. To accomplish these purposes, this chapter reviews and analyzes selected literature and studies on requesting and conversational goals. The review and analysis form the basis for establishing a theoretical framework for the study and in formulating the research questions. Chapter 2 present the methods used in conducting the study. Then, findings of this study are presented in Chapter 3. Finally, Chapter 4 discusses the important findings and present implications for intercultural communication theory and practice. Why Requests? This dissertation deals with the particular communication phenomenon of "requesting."1 In a conversational setting, requesting refers to communication that indicates a speaker's desire for the hearer to bring about some desired state or event, which would not have occurred otherwise (see Fraser, 1975; Becker, 1982; Hermann, 1983; Tracy, Craig, Smith & Spisak, 1984). Requests, which are used frequently even early in language acquisition, typically account for between 25 and 50 percent of utterances (Reiss, 1985; Becker, 1982; Dore, 1977; Ervin-Tripp, 1977). Skills that enable people to make the partner perform the requested action are critical for participation in society. According to Labov and Fanshel (1977), speakers must give more attention to the proper handling of "requests" than to any other form of face-to-face interaction. According to Brown and Levinson (1978), requests by definition are face-threatening: hearers can interpret requests as intrusive impingements on their freedom of action; speakers may hesitate to make the request for fear of exposing a —éh need, or risking the hearer's loss of face. Therefore, high social stakes are involved for both speaker and hearer in the choice of the specific way in which the request is made. Given their commonness and the fact that the use of requests involves an array of linguistic and social skills, the study of their realization across cultural groups is an excellent domain for investigation. Taxonomy of Request Tactics Conversational goals are realized by actions or strategies. Languages offer a proliferation of alternative forms or means to reach request goals. Just as there are different ways to begin an interaction and test affinity, so there are different ways to request the same action. The ideas of "strategy" and "tactics" have been postulated in describing relatively concrete communicative actions (Berger, 1987; von Cranach, Kalbermatten, Indermuhle & Gugler, 1982; Miller, Galanter & Pribram, 1960). Generally, strategies are viewed as a class of similar actions that are used to attain situation- specific interaction goals, and tactics are seen as concrete actions that persons manifest in their goal-directed interactions with others (see, Berger et a1., 1989; Street & Cappella, 1985). The notion of a request tactic is defined here by the particular sentential form and meaning that the speaker employs to instantiate a strategy and to reach a request goal (e.g., request to borrow money, request to be on time, etc.). A classification scheme of request tactics was established in terms of 12 mutually exclusive main tactics, that vary along several contextual features. Table 1-a presents the proposed twelve tactics for making a request. The present category scheme is an integration —;h Table l-a Classification of Request Tactics In the following, the examples are illustrated by means of the situational context, "The requester wants the bearer to repay a loan." Category 1. Mild hint The utterance contains no elements which are of immediate relevance to the intended illocution, but is still interpretable as a request through the context. This form of request relies largely on the interpretative capacity of the bearer. The utterance contains an implicit need for action preparatory condition in which the hearer is to infer that some action is necessary. I have run out of cash. 2. Strong hint The illocutionary intent is not immediately derivable from the utterance; however, the utterance refers to relevant elements of the intended illocutionary act. It requires less inferencing activity on the part of the hearer than mild hint. The utterance contains reference to a more specific "need for action" preparatory condition than mild hint. I could use the money I loaned you. 3. Question hint The illocutionary intent is embedded in what is seemingly an information question. While strong hints are typically assertions, information questions take the form of an interrogation. The utterance gives an escape route to the hearer who does not want to comply in treating the question directive as if it were a request for simple facts. Do you remember the money I lent you? 4. Syntactic downgraders Syntactic downgraders modify the utterance internally by mitigating the impositive force of the request by means of syntactic choices. The utterance is not complete without the mitigating part (underlined below) of the utterance. Would it be alright if I ask you to repay the loan? Would it be possible for me to ask you to repay the loan? Do you think you could possibly repay the loan? Table l-a (Continued) Do you mind if I ask you to repay the loan? I wonder if you could repay the loan. 5. Permission A requester is offering the bearer the authority to grant the permission to make a request. The focus is on the requester's activity, rather than the hearer's. May I ask you to repay the loan? Can I ask you to repay the loan? 6. Ability Query-Preparatory The utterance contains reference to an ability preparatory condition for the feasibility of the request by questioning about the hearer's ability to perform the desired action. This indicates that the requester will not interfere with the hearer's freedom of action. Could you repay the loan? Can you repay the loan? 7. Willingness Query-Preparatory The utterance contains the reference to the hearer's intention, willingness, or commitment to carry out an action. Unlike syntactic downgraders, the request is realized from the viewpoint of the hearer. Would you mind repaying the loan? Will you repay the loan? Would you repay the loan? Won't you repay the loan? 8. Suggestory The utterance is phrased as a suggestion to do an action. This indicates in some respects the hearer wants what the requester wants. How about repaying the loan? Why don’t you repay the loan? What about repaying the loan? Why not repay the loan? You'll repay the loan, Won't you? 9. Want The utterance expresses the requester’s desire or want that an action comes about. The utterance contains reference to a Table l-a (Continued) sincerity preparatory condition (the source's desire for change) for the feasibility of the request. I want you to repay the loan. I'd like you to repay the loan. I would prefer it if you repay the loan. I would appreciate it if you repay the loan. I'll be very happy if you repay the loan. 10. Performative The illocutionary verb denoting the requestive intent is modified, e.g., by verbs expressing the requester’s intention. I must ask you to repay the loan. I'm asking you to repay the loan. I ask you to repay the loan. I request you to repay the loan. 11. Obligation The requester implies that the bearer is under some obligation to do the desired action. It gives the bearer little choice in the matter. The obligation here is more severe than in the commitment requests. The illocutionary point is directly derivable from the utterance. In this category, the distinction between assertions and question (the second example) is not important since the latter is not typically heard as a literal question. You should repay the loan. Shouldn’t you repay the loan? You'll have to repay the loan. You must repay the loan. 12. Imperative Imperative is the literal form of a request. It makes it clear what the requester wants the bearer to be uniformly heard as a request in the contexts in which they occur. Repay the loan. of previous attempts to set up classifications of request or directive strategies in different languagesZ (see Table l-b). Table 1-b shows that it is possible to sort the previously developed category schemes into this current taxonomy. The 12 tactics are distinguished among three main strategy types. These three strategies are: (a) Higt, comprised of tactics 1,2, and 3; (b) ngry, comprising tactics 4, 5, 6, 7, and 8; (c) Direct Statement, comprised of tactics 9, 10, 11, and 12. In the hint category (mild, strong, and question hint), illocutionary force is not immediately derivable from the literal meaning of the utterance, even though the utterance refers to the necessary (felicity) conditions for performing requestive acts. On the other hand, the interpretation of query strategy (syntactic downgraders, permission, ability, willingness, and suggestory) is aided by query form. In the Direct Statement strategy (i.e., want, performatives, obligation, and imperatives), requestive force is marked explicitly, making little inferential demand (see also the somewhat different request classification schemes of Blum-Kulka, House & Kasper, 1989; Hoppe-Graff, Herrmann, Winterhoff—Spurk, & Mangold, 1985). Much has been learned about strategies or tactics for achieving various interaction goals. It has been suggested that one can distinguish strategy or tactic from overarching constraints that serve as criteria for making a choice between tactics in pursuing interaction goals. In the following section, the notion of interactive constraints is defined and the several distinguishing characteristics are noted. 10 Table l-b Past Classification Schemes of Request/Directive Tactics (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Mild hint Strong hint Question hint Syntactic downgrader Permission Ability Query-Preparatory Willingness Query-Preparatory Suggestory Want Performative Obligation Imperative s y,«-,..,—. ll Becker Blum- Tactic Anderson Becker et al. Kulka (1990) (1982) (1989) (1983) (1) --.- hint hint hint (2) question (3) directive ("Gotta match?") (4) permission permission permission (5) directive ("May I have") directives ("May I") embedded ability (6) imperative ("Could you") ("Can you") request * willingness (7) ("Would you") question ("Will you) "why not" (8) questions need or desrre (9) need desire need ("I want statement statement you") (10) - Obligation (ll) « ("You should") "Let’s" imperative (12) -imperative imperative imperative "You“ embedded -imperative imperative ........................................................................... Blum- Kulka (1987) 12 Blum- Kulka et al. (1985) query- preparatory conventionally indirect ("Could you") Brown & Carrell & Levinson Konneker (1978) (1981) hint (off-record) ------------ interrogative (no modal) ("Do you have") negative interrogative politeness (modal) ("Could ("Can you“) you") question-directive ("Will you") preparatory ("Would you mind") positive politeness ("How about") scope- stating ("I want you") (hedged) performative performative ("I am asking you") mood locution derivable declarative (need/ want) bald-on imperative -record (elliptical) Clark & Lucy (1975) Clark & Schunk (1980) 13 Ervin- Tripp (1976) Fraser (1978) hints memory ("Did I ask you") question directive ("Gotta match?") Imposition ("Would you mind") permission (”May In) permission permission ability ("Can you") commitment ("Will you") directive ("May I") ("May I") embedded ability imperative ("Can/could ("Could you") you") intention ("Will/would your!) directive intent ("How about") "I will be very happy" "I’d love if" desire ("I'd like you to") need S tatement intent to perform ("I request") obligation ("Shouldn't you") obligation ("Don't you have to") "Please color the circle" l4 Fraser et al. (1980) Garvey (1984) Cordon & Ervin—Tripp (1984) conventionalized hints ("Have you got") permission ("May I") permission ("Can/May I") imbedded imperative ("Could you") imbedded request ‘ ("Can/Could") "Why don't you" "I would appreciate it if" need statement Fraser & Tactic Nolen (1981) (1) (2) (3) "Would (4) you mind" (5) "Could (6) you" "Will/ (7) would you" (8) "Why don't you do that" "I‘d like (9) you to" "I must (10) ask you" "You have (11) to do that" (12) "Do that" imperative imperative imperative (ellipsis) 15 Holtgraves & Hill House & Tactic Hermann Yang et al. Kasper (1983) (1990) (1986) (1981) primary (1) goal mild hint ("I need to") (2) strong hint "Do you (3) have" "Would it (4) be alright" "Do you think" (5) "May I" "can I" (6) "Can you" "Could you" "Can you" query ---------------------------------------------------------- preparatory "Would you" "Would you (7) ”Would you ' mind" mind" (8) secondary "I want you "Lend me" (9) goal to" "Gimme" want ("I'd like") "I’d like" "A pen" hedged /explicit (10) performatives (ll) legitimization obligation ("You have to") choice of means mood (12) ("Please do") derivable ........................................................................ Tactic Jordan & Roloff (1990) 16 Kemper & Thissen (1981) Lim (1988) Parkhurst & Gottman (1986) need assertation ("I sure could use") "I think the leaves need to be raked" hint implied requests/ hints resource inquiry ("Did you bring") "Have the leaves been raked?" questions regarding the intention self- assessment ("Do you think") question imperative ("Can you"/ Will you") "Could you rake the leaves?" "Would you rake the leaves?" commitment imposition "I suggest you rake the leaves“ suggestion ("Why don’t you") reason ("Why not") "I would like you to rake the leaves" need statement ("I want") obligation ("You should") "(Please) rake the leaves" (Please) + imperative ........................................................................ 17 Sanford & Tactic Rintell Roach Searle Walters (1981) (1987) (1975) (1979) (l) (2) "You have (3) any?" elaborated (4) "Is it possible" request-to (5) "May I" request ' ("Can I ask") "Can you" O "Can you" (6) "Could you" wanting ability modal auxiliary --------------------------- "Could/ ----.--------------------- "Would you" would desire/ "Would/ (7) willingness will you" wanting (8) ("How about") suggestion claim of wish/ (9) preference want ("I'd like to") (10) demand ("I insist") reasons for you (11) doing A have to strong/weak (12) imperative imperative imperative l8 Tactic Winograd (1977) (1) "The door (2) needs shutting" (3) "Would it be (4) possible" (5) (6) "Could you" (7) "Won't/ Would/will you" (8) (9) (10) (11) "Shut (12) the door" l9 Characteristics of Interactive Constraints Much of the knowledge about communication goals (or objectives of a conversation) has been formulated in terms of the degree of abstractness (see Kellermann, 1989; McCann & Higgins, 1988; Read & Miller, 1989; Street & Cappella, 1985; Wilson & Putnam, 1990). At the most abstract, global level, several authors have argued for cross-situational constraints that serve as criteria for making a choice between tactics in the pursuit of interaction goals. In order to define the notion of global constraints, we have to define strategies/tactics and primary interaction goals. The idea of "strategy" and "tactics" has been postulated in describing relatively concrete communicative actions (Berger, 1987; von Cranach, Kalbermatten, Indermuhle & Gugler, 1982; Miller, Galanter & Pribram, 1960). Generally, strategies are viewed as action sequences that are used to attain goals, and tactics as specific behavioral actions that persons manifest in their goal- directed interactions with others (see, Berger et al., 1989; Street & Cappella, 1985). The next level of goals consists of the numerous outcomes or primary goals which may be desired from an entire interaction. An overall strategy and specific tactics for carrying out that strategy are implemented only if we select a functional outcome desired from interaction (e.g., gaining compliance, seeking information, de-escalating relationships, correcting others, and testing affinity). As people pursue interaction goals such as gaining compliance, seeking information, or altering relationships, they generate messages within a variety of constraints. These higher-level 20 constraints have been named "supergoals" (von Cranach et al., 1982), “life theme" (Schank & Abelson, 1977), "cross-situational goals" (Street & Cappella, 1985), "supermaxims" (Grice, 1975), "meta- strategies" (van Dijk & Kintsch, 1983), "meta-goals", "meta-plans" (Wilensky, 1981, 1983; Berger, 1987; Kellermann, 1988), and "ritual- constraints" (Goffman, 1967). The main implication is that, regardless of one's interaction goal, there exist higher-level concerns regarding how one will achieve that interaction goal. Interactive constraints are fundamental concerns regarding Ehg manner in which a message is constructed. They tend to affect the general character of every conversation one engages in, and an individual's conversational style in general (Wilensky, 1983; Kellermann, 1988). Global constraints contribute to consistent conversational performances across varying contexts (i.e., Street & Cappella, 1985). Contents of Interactive Constraints The current study is based on the assumption that human beings in any culture take into account certain major interactive constraints when choosing conversational strategies. Based on past literature and a series of interviews3, this paper proposes the following five interactive constraints: concern for clarity, concern for effectiveness, concern for avoiding hurting the hearer's feelings, concern for minimizing imposition, and the concern for avoiding negative evaluation by the hearer. In the following, each of these interactive constraints is defined. Concern for clarity. Concern for clarity as applied to conversational behavior is defined as the likelihood of an utterance 21 making one's intention clear and explicit. That is, the clarity concern controls the degree to which a strategy is explicitly and unambiguously communicating to a listener what the intention of the message is (i.e., Blum-Kulka, 1987). The global constraint (or preference) for clarity, as applied to conversational behavior is, therefore, a concern for achieving an outcome in the most explicit way possible. Movement toward increased clarity typically results in the choice of more pointed and direct tactical means. For example, if one's primary goal is to request an action, direct imperative forms (e.g., "Repay the loan" or "Lend me your book") will make the speaker's illocutionary point explicit. The current typology of request tactics seem to vary on a clarity dimension. Specifically, in the first three hint tactics, illocutionary force is not derivable from the literal meaning of the utterances, while the interpretation of such strategies as syntactic downgraders, permission, ability, willingness, and suggestory is aided by conventional usage. In the strategies of want, performatives, obligation, and imperatives, requester's intention is marked explicitly, making little inferential demand. The idea of clarity has been talked about frequently in the literature on conversation (Brown & Levinson, 1978; Blum- Kulka, Danet & Gherson, 1985; Leech, 1983). For instance, Grice (1975) put forward the ”Maxim of manner" in the use of language (e.g., be clear, be brief, try to avoid obscurity), which can be seen as guidelines for clear communication. Concern for minimizing imposition. This constraint pertains to the degree to which an utterance does not impose on the hearer or 22 interfere with the hearer's freedom of action (Brown & Levinson, 1978). An act of communication threatens the hearer's negative face to the extent that it imposes on one’s right to autonomy (Scollon & Scollon, 1981). Several authors (Blum-Kulka & House, 1989; Scollon and Scollon, 1981; Ting-Toomey, 1988) argue that the notion of "politeness" usually is associated in the Western world with "negative" or "deference" strategies: the show of deference expressed by the effort not to be heard as imposing on the bearer, by not assuming cooperation, and by leaving the hearer options for noncompliance. While the salience of this constraint might differ across cultures, prior research confirms the importance of nonimposition in conversational performance. The current request strategies also appear to vary along the dimension of negative-face imposition on the hearer (i.e., the threat to the desire to maintain autonomy and be unimpeded by others) (see Brown & Levinson, 1978). While the absolute amount of negative threat of a given tactic might vary across cultures (see Scollon & Scollon, 1981; Ting-Toomey, 1987), the degree to which tactics constrain the target's behavioral options may consistently increase across cultures as we move up the scale. The imperative (e.g., "Repay the loan") may impede the hearer’s freedom of action more than; say, the three forms of hints. Hints provide the listener the power to control the interaction by allowing him or her different response options (e.g., compliance, refusal, interpretation of the utterance as a statement). Concern for avoiding hurting the hearer's feelings. When planning to achieve interaction goals, people also may take into account how their projected actions might affect the hearer's feelings. "Concern for the other’s feelings" relates to the speaker's perceived obligation to save a hearer's desire for approval-seeking or the positive self-image that the hearer claims (Brown & Levinson, 1978; Ting-Toomey, 1988). Research on politeness focuses on strategies for minimizing threats to other's face and reinforces the importance of the concern for the other's feelings in conversational behavior (Brown & Levinson, 1978; Lakoff, 1977; Leech, 1983; Scollon & Scollon, 1981). Concern for the hearer's feelings has been proposed as a constraint under various pseudonyms including "the want to maintain the hearer's positive face" (Brown & Levinson, 1978), "identity goals" (Wilson & Putnam, 1990), and "concern with support" (need to show concern for other’s feelings) (Greene & Lindsey, 1989). Utterances involving bald imperatives (e.g., "Do X!") may risk a higher chance of hurting the other's feelings than hints (given the same situational contingencies), since the former, with a lack of request mitigation, potentially convey the implicit message that the speaker is not concerned about the relationship, but only with accomplishing the instrumental outcome (e.g., getting the money back). Concern for avoiding negative evaluation by the hearer. This interactive constraint represents the desire to avoid negative evaluation by the conversational partner. Specifically, it is an assessment that an utterance does not cause dislike, devaluation, or rejection by the bearer. According to self-presentation theory (Weary & Arkin, 1981), self-conceptions are formed by how people 24 believe others perceive them. Individuals in turn attempt to behave in ways that avoid devaluation by others. This constraint is consistent with Brown and Levinson's (1978) notion of a speaker's desire to save his or her own positive face. In recent years, several authors have suggested similar interactive constraints that are "motivating forces" in communication: "impression management goals" (Street & Cappella, 1985) and "approval-seeking strategies" (Ting-Toomey, 1988). While people, in any culture, will have the general desire to employ certain strategies minimizing the negative evaluation by others, it has been speculated that members of collectivistic cultures would tend to use more approval-seeking strategies (positive-face need) than would members of individualistic cultures (Ting-Toomey, 1988). Such tactics as imperatives, obligation, and performatives potentially could risk more devaluation for self since they demand more compliance with the requests than do three hint tactics. Concern for effectiveness. "Concern for effectiveness" is another major global constraint that may influence choices of conversational tactics and strategies. This constraint reflects an individual's desire for getting the end results from an interaction (i.e., perlocutionary effect). Similarly, communication effectiveness is defined as successful goal achievement or task accomplishment (see Canary & Spitzberg, 1987). Judgment of communicative competence has been shown to be related to the effectiveness with which goals are pursued (Canary & Spitzberg, 1987). While effectiveness alone is not sufficient condition of competence, it nonetheless indicates the importance of 25 this constraint in social interaction. One important distinction should be made between communication effectiveness and other interactive constraints. Effectiveness does not adhere to particular communication strategies or tactics; rather, effectiveness is a judgement of the use of particular tactics in specific settings for specific purposes. Effectiveness judgments are made of context-embedded factors rather than of language act types. Thus, tactics which are the most effective in one situation may be the least effective in others. Some scholars have noted the relative instability of effectiveness judgments (Burleson et al., 1988). Canary and Spitzberg (1987), for instance, found that although appropriateness evaluations of three conflict strategies (integrative, distributive, and avoidant) had a stable rank order regardless of episode type (opposite versus same sex), the rank order of perceived effectiveness across the strategies did vary across episodes. Judgments of strategies along the other constraint dimensions may be more stable across subject samples and situations than effectiveness judgments. In sum, one of the most important distinctions between other interactive constraints and effectiveness appears to be that the former primarily depends on the fundamental properties of a tactic, while effectiveness judgments depend heavily on contextual factors (i.e. "Will this tactic work for me in this situation?" or "will it successfully lead to my desired outcome in this situation?"). Thus, it is possible that effectiveness may function as an interactive constraint on a somewhat different level than the other four constraints. 26 In summary, the current study is based on the assumption that the above five interactive constraints may serve as general motivating forces in the selection of strategies and tactics and potentially serve as important determinants of "cultural ways of speaking" (cf. Katriel, 1986). Cross-cultural Cgmparisons of Request Tactics along the Interactive Constraint Dimensions Clusters of strategies on each dimension. In this study, the 12 request tactics have been distinguished among three main strategy categories: (a) Hint, comprised of strategies such as mild hint, strong hint and question hint; (b) ngry, comprising strategies such as syntactic downgraders, permission, ability, willingness, and suggestory; (c) Direct Statement, comprised of strategies such as imperatives, obligation, performatives, and wants. It is expected that request tactics display the three major clusterings regardless of cultural contexts. Hints, by alluding to shared knowledge, serve solidarity-enhancement, as do in-group jokes (Ervin-Tripp, 1976). Thus, the first three hint strategies (mild, strong, and question hints) are difficult to interpret without the use of some shared information. This kind of relationship frequently occurs in a solidarity society, in which a great deal of information is already embedded in the context (i.e. Ervin—Tripp, 1976). On the other hand, tactics that belong to Query strategy (syntactic downgraders to suggestory) are typically questions or interrogations that ask for an answer from the hearer, and specify the necessary conditions for performing acts (see, Labov & Fanshel, 1977; Searle, 1969; Wilson, Kim & Meischke, in press). Thus, the interpretation of such ....v .. ...»... 435W. magi. i. 1.111...- 7.... .. 27 strategies is aided by conventional usage. Finally, the tactics that belong to Direct Statement strategy (imperatives, obligation, performatives, and wants) manifest that the intent of the requester cannot be taken for granted, as in task-centered groups in offices and laboratories, where explicitness and clarity have value (Ervin- Tripp, 1976, p.44). The requesters are forced to elaborate their meanings and make the intentions both explicit and specific so that little meaning is left open to context. Given these characteristic differences, request tactics may cluster together across the three main groups along the five constraint dimensions. Placement of request tactics along the five dimensions. Although ’no cross-cultural investigation has explored the rank-ordering of interactive constraints, previous empirical Studies have tended to show cross-cultural stability in the ranks of the relative politeness of linguistic forms (Walters, 1979; Hill, Ide, Ikuta, Kawasaki & Ogino, 1986; Rintell, 1981). Recently, Holtgraves and Yang (1990) attempted to test the potential universality of Brown and Levinson's (1978) politeness theory. They found that the rank ordering of the politeness ratings of the four superstrategies was very similar for Americans and Koreans, even though the difference between the two most polite and two least polite strategies was greater for Koreans than for Americans. Walters (1979), in his contrastive analysis of the perceived politeness of request strategies, also found a strong rank order correlation between English and Spanish requests (ET-70)- Given the consistency of the results in the prior research, the rank ordering of constraint judgments is predicted to be similar across cultures, even though the absolute levels of the five constraints 28 judgments vary. Potential Incompatibility among the Interactive Constraints. One major question to be addressed here is the potential incompatibility among the five interactive constraints. Wilensky's (1983) analysis on interaction goals suggests that there are two kinds of goal relations: (a) goal competition, in which individuals' goals are negatively related to one another, and (b) goal concord, in which individuals’ goals are positively related. A consistent findings from analyses of social goals and social situations has been the differentiation of task goals and task situations from social goals and social situations (Argyle, et a1., 1981; Bales, 1950; Wish & Kaplan, 1977; Kellermann & Kim, 1991). While purely social goals such as pleasure and enjoyment might lack a.task component, task goals almost always carry both a task and a social component (see Argyle, et a1., 1981). Focusing on social appropriateness and efficiency of goal outcome as two major conversational constraints (metagoals), Kellermann and Kim (1991) found that a set of 49 primary conversational goals did differ among themselves in their degree of task orientation and their differences were associated with metagoal compatibililty. It is this overlaying of a task purpose onto the social exchange that may lead to increases in tension between interactive constraints. Given that the request goals are overlaid with a task purpose, it is expected that there will be a high degree of incompatibility between the clarity of strategies and the three other constraints (minimizing imposition, avoiding hurting the hearer's feelings, and avoiding negative evaluation by others) in both cultures. 29 Cross-Cultural Comparisons of the Importance of Interactive Constraints The perceived importance of interactive constraints. The perceived importance of each constraint may differ systematically across cultures. For example, Brown and Levinson (1978) allow for the possibility of urgency (or clarity) concerns being part of the considerations subject to cross-cultural variation: "Cultures may differ in the degree to which wants other than face wants (such as the need for efficiency, or for the expression of power) are allowed to supersede face wants" (p. 254). If there is a societal norm of sincerity and directness in talk, for example, sincere disapproval is less of a face threatening act than it is in societies not having such a legitimization of non-face wants. For instance, speakers of Hebrew attach a high value to~'sincerity' in speech. In her ethnographic study on the ethos of directness in Israeli society, Katriel (1986) shows that the notion of gugri ’direct, straight' talk, is positively associated with concepts such as sincerity (truthful expression), naturalness (simplicity, spontaneity), solidarity, and anti-style, thus legitimizing a conscious suspension of relational concerns over clarity or effectiveness of communication. The perceived importance of each conversational constraint may be different across cultures, due to differing social norms operative in different societies. While some evidence suggests that there are discrepancies between perceived and empirically observed weights of variables of this nature (Summers et a1., 1970), data on perceived importance may still allow us to investigate whether considerations 30 of each conversational constraint point to a general picture which is consistent with the enduring cultural generalizations (stereotypes) in interaction styles. While people seldom reflect on the nature of their requesting behavior, a cultural group may share "commonsense theories" of requesting behavior. It is likely that people make requests on the basis of their conceptions of commonsense theories of requesting styles that are culturally preferred. Effects of social status on the perceived importance of constraints. The status relationship between the speaker and the hearer has received attention in the study of request forms (Brown & Levinson, 1978; Ervin-Tripp, 1976; Donohue & Diez, 1985; Garvey, 1984; Scarcella & Brunak, 1981). "Status" refers to the relative dominance (superordinate, subordinate, or of the same rank) of one person with respect to another. When an individual continuously uses imperatives for requesting nontrivial information and the other complies with the requests, the requester's power position is enhanced (Ervin-Tripp. 1976). Research indicates that speakers of relatively high status are less polite than speakers of relatively low status (Blum-Kulka, Danet & Gherson, 1985). These differences, if they exist, might be accentuated in some cultures. Thus, the level of social status may moderate the variations in the perceived importance of interactive constraints. Relative weights given to each constraint in predicting likelihood of use: One purpose of this study is to seek out explanations that account for the perceptual variations in the likelihood of use of request tactics4. It is expected that there will be cultural differences in relative weights given to each 31 constraint in predicting likelihood of use ratings. Specifically, the relative regression weights of each constraint in predicting likelihood of use ratings is predicted to be consistent with each cultural group’s perceived importance ratings of each constraint. This prediction, if it turns out to be valid, may help us to understand the psychological determinants of people's beliefs about request strategy choice. Intra-Cultural Variability in Beliefs about Requesting A final question of interest in the current investigation is whether intra-cultural variability of the importance of each constraint will be smaller than cross-cultural variability. In this dissertation, the following individual-difference variables are examined since they seem to have the potential to account for cross- cultural similarities and differences in the constraint importance: (a) Need for approval: The degree to which an individual worries about what others might think of his/her action (Salzman & Hunter, 1983); (b) Self-esteem: The degree to which an individual respects his/herself or considers his/herself worthy (Buss, 1980); and (c) Need for Dominance: The degree to which an individual desires to control and to dominate social situations (Salzman & Hunter, 1983). Concern for other's feelings and devaluation seem to be at the heart of such individual-difference variables as need for approval. As concern for the reactions of others increases, the importance of public self-awareness or approval also may increase; and so may the importance of interactive constraints such as concern for the hearer's feelings and devaluation. It has been suggested that persons high in assertiveness, need for control, and self-confidence 32 tend to be more concerned with the clarity of goal orientation (i.e., Frese, Stewart & Hannover, 1987). Thus, these individual difference variables may systematically affect the importance of interactive constraints. Cultures Selected Korea and the United States were selected for comparison because they offer great variance in respondents’ attributes. The two countries seem to be very different in their cultural orientations, such as high versus low context (Hall, 1976; Hall & Hall, 1985) and collectivism-individualism (Triandis, Bontempo, Villareal, Asai & Lucca, 1988; Hofstede, 1979, 1980). Individualism-collectivism and low-versus high-context communication are broad dimensions of cultural variability that influence many different aspects of interpersonal communication (see Gudykunst & Ting-Toomey with Chua, 1988; Gudykunst & Kim, 1984). Specifically, the cultural variability appears to be consistent with discussion of variability in the use of directness versus indirectness and certainty and ambiguity in communication (see Okabe, 1987; Katriel, 1986; Yum, 1988). Given the differences in terms of context and relational orientation, Korea offers an excellent contrast to the United States. Research questions Consequent to this literature review, the following research questions were formulated. 1. Are there significant cross-cultural differences in the perceptions of tactics along the five interactive constraint dimensions? a. in the clusters of tactics in each constraint? 33 b. in the perceived placement of request tactics along the five interactive constraints as well as for perceived likelihood of use? c. in the degree of incompatibility among the five constraints? 2. Are there significant cross-cultural differences in the importance of the five interactive constraint dimensions? a. in the perceived importance of each constraint? b. in the moderating effect of the social status in the perceived importance of each constraint? c. in the relative weights given to each constraint in predicting likelihood of use? 3. Are there significant intra-group variabilities in the implicit theories about requesting styles? a. Will any of the individual difference variables (need for approval, self—esteem, need for dominance, and participants' gender) be highly correlated with the perceived importance of each constraint? b. Will culture still affect the perceived importance of each constraint after controlling for the effects of individual difference variables? CHAPTER II METHOD Overview The main purpose of this study is to test potential cross- cultural differences in implicit theories of requesting styles with participants from Korea and the United States. After a series of pilot tests, participants were randomly provided with a questionnaire containing one of the six situations with one of the two sets of exemplar tactics for that situation. After being presented with a request situation, participants were asked to rate the perceived importance of each of the five constraints. Then participants evaluated a set of 12 exemplar tactics along the five dimensions of constraints as well as for perceived likelihood of use. Finally they completed measures of self-esteem, need for approval and need for dominance. Participants The participants in the main study are a total of 595 undergraduates, 296 Koreans (native speakers of Korean) and 299 Americans (native U.S. English speakers), studying in their respective countries. The U.S. participants are undergraduate students enrolled in introductory communication courses at Michigan State University. Korean participants also are undergraduates at Choong-Nam University in Daejeon, Korea. Table 2 presents the summary of demographic characteristics of participants from Korea and the U.S. The mean age of Korean subjects was 21.14 years and it was 20.87 years for the U.S. subjects. Regarding sex composition of the subjects, females comprised 56 percent of the Korean sample and 68 34 Table 2 Comparison of Demographic Characteristics of Korean and the U.S. Participants Korean subjects U.S. subjects (N=296) (N=299) Female 166 (56%) 203 (68%) Sex ------------------------------------------------ Male 130 (44%) 95 (32%) missing 0 (0%) 1 (0.3%) Mean Age 21.14 (yrs) 20.87 (yrs) Communication 80 (27%) 101 (34%) Major --------------------------------------------------- Social Science 91 (31%) 63 (21%) Humanities 15 (5%) 17 (6%) Others 108 (36%) 115 (39%) missing 2 (1%) 3 (1%) Freshman 68 (23%) 25 (8%) School Sophomore 148 (50%) 76 (25%) year --------------------------------------------------- Junior 32 (11%) 98 (33%) Senior 43 (15%) 99 (33%) 36 percent of the U.S. sample. The distribution of academic major is similar across the two samples. Even though some variations exist regarding the composition of the school year, there is no reason to believe that that may substantively affect the results. In sum, the samples investigated share many participant traits. All were university students, of similar age, sex composition, and level of education. While the choice of subjects in this study is selective (they are all college students), they are the most accessible groups in both countries. As Johnson and Tuttle (1989) point out, the question of which population should be examined is not subject to much consideration in the United States due to the relative mobility of U.S. citizens and the lack of clearly defined status. As for the Korean samples, the selected college students are not only comparable to their U.S. counterparts, but are also more or less representative of the population since highly homogeneous societal norms, values and rules of conduct are continually reinforced throughout one's education in Korea. In both groups of subjects, those who lived in foreign countries for more than three months (by which time serious acculturation might start occurring), and those who fluently speak a second language were excluded. Request situations To enhance generalizability of situations, six vignettes describing a wide variety of request situations were written (requests for fulfilling an obligation, obtaining permission, borrowing goods, and soliciting agreement).5 The situations were designed to vary in terms of the participants' social status, so that W 37 in each of two situations the requester is superordinate, subordinate, or of the same rank as the requestee. Appendix B presents the full text for each situation. (Social status: HearerSpeaker) 85: A student asks his/her professor for'an extension on a homework deadline. (Homework) S6: A research assistant asks his/her professor for permission to take a day off. (Research) Pilot Tests / ealism of re uest situations. As a check for realism of each W request situation, two separate groups of participants (30 native Korean speakers and 30 native American English speakers), studying at a large midwestern university, were asked to read each of these situations and rated the degree to which they agree with each of the following statements on three 7—point scales ("It is easy to imagine myself in this situation," "The above situation is realistic," and "I find it difficult to place myself in the above situation"). Korean subjects were instructed to imagine that each situation was happe in Korea. All six situations yielded average realism ratings 38 exceeding 5.0, with the mean realism score of 5.94 (SD=.71) for Korean subjects and 5.59 (SD=.73) for the U.S. subjects (see Table 3). Perceived social relationships and reqpest size: Members of different cultures might differ in their perceptions of request situations. For example, if the degree of relative power universally dictates variation in linguistic behavior and the social status of a professor in one culture is perceived as much higher or lower than in another, this may lead to a difference in the "implicit theories" about request styles. Thus, checks were made regarding whether the two cultures were compatible in terms of perceptions of external role relationships such as social status (in which the requestor is superordinate, subordinate, or the same in rank relative to the requestee), social distance (in which they are close, familiar, intimate or more distant, unfamiliar, lacking shared interests or attributes) and in perception of factors more specific to the requesting situation (i.e., the size of the request). Two different groups of subjects (25 native Korean speakers and 30 native American English speakers) participated in the second pilot study. Each participant was asked to read one of the six situations and rate the relative social status and social distance between the interactants on two 7-point scales for each. Participants also were asked to rate the size of the request on three 7-point scales. Table 4 lists the items in each scale and the reliabilities of the measures. Selection of Reqpest Tactics Generally, tactics are viewed as concrete behavioral actions 39 Table 3 Realism of Request Situations (Pre-test) U.S. Korean Participants Participants (N—40) (N=36) (3.3;; """""" 1;; """" 513 """" a... """ $13 """"" fin}; """"""" 2.1}; """ i3; """" 33% """ I66 """" Repay 5.83 1.32 5.87 1.24 Homework 5.55 1.47 6.00 .91 Research 5.96 1.09 6.41 .90 Delay 5.79 1.19 5.34 1.47 Borrow 5.49 1.41 6.04 1.18 Total 5.59 73 5 94 71 Note: Larger mean scores indicate higher perceived realism. 40 Table 4 Perceptions of Request Situations: Scale Items, and Reliabilities. (Pre-test) Perceived Social Status Reliability - .94 (Korean) / .76 (American) 1. In the above situation, you and your friend (student/professor) are of equal social power. 2. In the above situation, you occupy a social status equal to that of your friend (student/professor). Perceived Social Distance Reliability - .85 (Korean) / .89 (American) 1. In the above situation, my friend (student/professor) and I are on very close terms. 2. I am closely acquainted with my friend (student/professor). Request Size Reliability - .81 (Korean) / .59 (American) 1. I find the request in the above situation is pp; a big deal. 2. The above situation describes a request of considerable magnitude (reverse item). 3. The request in the above situation is a trivial one. 41 that persons manifest in their goal-directed interactions with others (see, Berger et a1., 1989; Street & Cappella, 1985). For this study, two utterances were written to instantiate each of the 12 tactics in each situation. Thus, a total of 144 utterances (2 utterances instantiating 12 tactics in six situations) was evaluated within each culture (see Appendix B for lists of tactics). After all the stimulus materials were constructed in English, they were translated into Korean by the author. Then another person, bilingual in Korean and English, translated the materials back into English. The two persons discussed any discrepancies and made changes in the Korean translation accordingly to ensure cross- cultural equivalence in meaning. The author strove to make the translations as equivalent as possible while also sounding natural in Korean. For example, the author deleted the subject from strategy forms (e.g., Will ypp lend me some money?) whenever it sounds awkward in Korean. In addition, some language-specific politeness markers in Korean (e.g., the honorifics) were not used when the sentence is addressed to equals (e.g., friends). For instance, in translating "Repay the loan" into Korean, if honorifics (e.g., Ju Se Yo) are added to the bald imperative form (e.g., Don Dol Lyo Jueo) in the Korean translation, then it is no longer an imperative. Perceived Importance of Interactive Constraints Each participant was randomly provided with a description of one of the six situations and_was asked to rate the perceived importance of each constraint in that situation. Importance ratings for the five dimensions were collected before specific tactics were evaluated to avoid sensitizing the subjects. Responses to the items were 7” sari- .. 42 measured on two 7-point items for each constraint (1 = strongly disagree, 7= strongly agree). Table 5 presents items measuring perceived importance of each interactive constraint and reliabilities of the measures. To estimate these reliabilities, the g transformation was applied to the correlations between the two items in each situation (for each of the constraints). Then, the average 5 scores across six situations were calculated between two items for each constraint, which, in turn, were transformed back into E- The resulting estimated unit-length reliability coefficients range from .47 to .72 for Korean subjects with the average reliability of .65. For the U.S. subjects, the reliabilities were slightly higher: from .60 to .79 with the average reliability of .72. The relationship between the number of items in the measure and its reliability is given by the Spearman-Brown formula. Using the formula, we can calculate the estimated reliabilities of the two item measure based on the unit-length measure reliabilities provided. Using the Spearman-Brown formula, the reliability of the two—item measure of each constraint was calculated based on the estimated reliabilities of a unit-length measure: .64 to .84 (M—.79) for Korean subjects and .75 to .88 for the U.S. subjects (M—.84). Ratings of Request Tactics along the Dimensions of Interactive Constraints After making a series of judgments on the perceived importance of constraints, the same groups of subjects rated each of the 12 exemplar tactics along each of the five dimensions of conversational constraints. Participants again were presented with the description of one of the six situations (the same one used for judgments of 43 Table 5. Perceived Importance of Interactive Constraints: Scale Items, and Reliabilities Item Reliability : single item/two items Clarity: Reliability -.71/.83 (Korean), .79/.88 (American) 1. In this situation, I feel it is very important to make my point as clearly and directly as possible. 2. In this situation, I want to directly come to the point while conveying my message. Concern for the other’s feelings: Reliability -.48/.64 (Korean), .74/.85 (American) 1. In this situation, I feel it is very important to avoid hurting the other's feelings. 2. In this situation, being considerate towards the other's feelings is a major concern to me. Non-Imposition: Reliability -.76/.86 (Korean), .79/.88 (American) 1. In this situation, it is very important not to intrude upon the other person. 2. In this situation, it is very important to avoid inconveniencing the other. Concern about disapproval: Reliability -.67/.80 (Korean), .64/.78 (American) 1. In this situation, it is very important that the other person does Egg see me in a negative light. 2. In this situation, it is very important that my message does pp; cause the other person to dislike me. 44 Table 5 (continued) Effectiveness Reliability -.72/.84 (Korean), .77/.87 (American) 1. In this situation, it is very important to get the other person to do what I want. 2. In this situation, making the other person to comply with my request is very important. Note: * To estimate the reliability of the single and two-item measures of the importance of each constraint, the correlations between the two items in each situation were subjected to the g transformation. Next, the average 5 scores across six situations were calculated between two items for each constraint, which, in turn, were transformed back into E- The resulting estimated unit-length reliability coefficients are reported as single-item reliabilities. ' ** one-item - Reliability of a single item (the average correlations calculated between two items for each constraint). two-item - Reliability of two-item measure calculated from Spearman-Brown formula 45 importance of constraints). Paired with the situation description was a list of the corresponding 12 utterances to be rated along the five constraint dimensions. Clarity: For clarity of strategies, the following instruction was provided: Please rate each statement in terms of the degree to which it communicates your intention in a clear, explicit, and unambiguous manner. Judgments of the tactics along the dimension of clarity were measured on two 7—point scales (direct-indirect; to the point--pp§ to the point). Concern for the hearer's feelings: For this judgment, participants read the following instruction: In this section, please rate each statement as to the degree to which using that statement hurts the other's feelings or embarrasses the hearer. Responses to the items were measured on the following two 7- point scales (hurts the other's feelings -- does Egg hurt the other's feelings; considerate to the other's feelings -- p95 considerate to the other's feelings) Concern about disapproval: The following instruction was provided to measure the degree to which an utterance risks devaluation from the hearer: In this section, please rate each statement as to the degree to which using that statement may cause the other to dislike, disapprove of, or reject you. Tactics were measured on two 7-point scales (the other will dislike me -- the other will Egg dislike me; the other will think negatively of me -- the other will pp; think negatively of me). Concern for minimizing imposition: The instruction was: 46 In the same situation as in the previous section, we would like you to determine the degree to which each statement may inconvenience or intrude upon the hearer. The two items measuring the degree to which an utterance imposes on the bearer were: pushy (not pushy) and imposing (not imposing). Effectiveness: For effectiveness of tactics, the following instruction was given: In this section, we would like you to rate the effectiveness of sentences. Effective statements are those that successfully get the other person to do what you want, while ineffective statements are those that would not get the other person do what you want. Judgment of the tactics along the effectiveness dimension was measured with the following 7-point scales: effective--ineffective; will lead to compliance--will not lead to compliance). Assessment of Likelihood of Use To measure the perceived likelihood of use of each utterance, the following instruction was provided: In this section, we would like you to indicate how likely you would be to use a particular sentence in the following situation. Ratings of the likelihood of use were made on the following two 7-point scales: likely to use ~— unlikely to use; unwilling to use -- willing to use. Table 6 presents a summary of the estimated reliabilities of items measuring constraint judgments of tactics. To estimate these reliabilities, the 12 correlations between the two items were calculated for each tactic across the six situations. Then, the 12 correlations for each constraint were transformed into 5 scores and were averaged. The average correlations were, in turn, transformed back into E. An estimated reliability of a unit length measure Table 6 47 Constraint Judgments of Utterances: Items and Reliabilities (Main study) Korean U.S. Participants Participants Scale Reliability* Reliability one-item (two-item)** one-item (two-item) (l) Clarity .84 (.91) .76 (.86) (2) Effectiveness .85 (.92) .84 (.91) (3) Concern for feelings .93 (.96) .79 (.88) (4) Concern about disapproval .92 (.96) .88 (.94) ‘ (5) Avoid imposition .76 (.86) .81 (.90) (6) Likelihood of use .84 (.91) .77 (.87) Overall average .86 (.92) .81 (.89) Note: * To estimate the reliabilities of the single and two-item ** measures of each constraint judgment, the 12 correlations between the two items were calculated for each tactic across the six situations. Then, the 12 correlations for each constraint were transformed into 5 scores and were averaged. The average correlations were, in turn, transformed back into g to yield single-item reliabilities. one-item - Reliability of a single item (the average correlations calculated between two items for each constraint). two-item - Reliability of two-item measure calculated from Spearman-Brown formula 48 ranged from .76 to .93 (average=.86) for Korean subjects and .76 to .88 (average-.81) for the U.S. subjects. Using the Spearman-Brown formula, the estimated reliabilities of the two-item measures of constraint judgments of tactics were calculated: .86 to .96 (M=.92) for Korean subjects and .86 to .94 (M=.89) for the U.S. subjects. Thus it appears that the set of two items used to tap each interactive constraint in this study provided highly reliable measures of each construct. Assessment of Individual Difference Variables Three individual difference variables chosen in this study are: need for approval, self-esteem, and need for dominance. In the present investigation, items in each scale were selected only if they maintained cross-cultural similarity in face validity. Need for approval was measured by selecting four items from Salzman and Hunter's (1983) 8-item measurement scale (see Table 7). Self-esteem was assessed with four items selected from Buss's (1980) 7—item measurement scale. Finally, participants also completed a three-item need for dominance scale (see Salzman & Hunter, 1983) to provide a measure of individual differences in the need to dominate social situations. All three scales used a 7-point response format (1-strongly disagree, 7-strong1y agree). The reliabilities of the three scales measuring individual difference variables were also calculated. Across cultures, the average reliability coefficient for the 4-item need for dominance scale was .76 (.79 for Koreans, .73 for Americans). The reliabilities for self-esteem and need for approval scales were lower than need for dominance: .67 (.56 for Koreans and .77 for Americans) and .50 (.56 for Koreans and .44 for Americans), 49 Table 7 Individual Difference Measures: Scale Items Need for Approval Strongly Strongly agree 7 6 5 4 3 2 1 disagree 1. Before I raise my hand in class, I always worry about what other classmates might think of my question. 2. I usually maintain my original position even when my superiors disagree (Reverse Item). 3. I usually avoid doing something that might provoke criticism. 4. I find it hard to do anything that my parents would disapprove of. Self-esteem l. I often wish that I was someone else (reverse item) 2. I'm fairly sure of myself. 3. I feel like I disappoint other people (reverse item). 4. There are lots of things about myself that need to be changed (reverse item). Need for Dominance 1. I feel that I can dominate a social situation. 2. I feel that I can control a social situation, even though it may not be obvious to other people. 3. In most social situations, I emerge as the leader. respectively. 50 CHAPTER III RESULTS In this chapter, the findings of this study will be described under the following main headings: (1) Comparisons of perceptions of request situations, and (2) Results of the main study. Cpmparisons of Percgptions of Request Situations To tap native speakers' perceptions of the request situations, three parameters which prior research has identified as important characterizers of situations were analyzed. Specifically, checks were performed to ascertain whether social role relationship (i.e., social status and social distance) and attributes of request situations (i.e., request size) were perceived similarly or not across cultures. Table 8 presents descriptive statistics for the rated request size. Overall, the cross-cultural differences in perceived request size were significant at .05 level for two of the six situations. However, the effect sizes for the variations in perceptions of request size were relatively small except the situation involving "Delaying the presentation" (£--3.8l, £-.46, p<.000). Taken all the situations together, the effect size for cross-cultural differences in the perceived request size was small and non-significant (£--1.45, df-53, p<.153; g--.19). In any culture there is a generally agreed ranking of social status in a given context. These intra-culturally defined general rankings of social status (e g., professor-student, friend-friend) relative to the bearer should be similar across cultures, even though the mean levels of social status may differ. As expected, both the 51 Table 8 52 Cross-cultural Comparisons of Perceived Request Size (l-low, 7-high) 03 39 57 .00 .27 .26 Korean Participants (N=25) Goals Mean SD Time 3.87 l Repay 3.69 1. Homework 4.77 1. Research 4.51 1. Delay 3.88 l. Borrow 3.16 1. Total 3 98 1 Note: * - Significant at . The U.S. Participants (N=30) Mean SD 3 p 3.97 1.50 -.24 .808 3.70 1.27 -.02 .983 5.44 .93 -2.06 .046 3.68 1.54 1.97 .054 5 38 1 18 -3.81 000 3 84 l 54 -l.71 094 4 34 75 -l.45 153 01 level. 53 U.S. and Korean subjects rated the social relationship between a professor and a student (Time, Homework, Research, and Delay situations) as marked by higher inequality of status (M=3.42 for the Korean subjects, M=4.52 for the U.S. subjects) than was present in the friend-friend relationship (Repay and Borrow situations) (M=l.51 for the Korean subjects, M=2.56 for the U.S. subjects) (see Table 9). On the other hand, the findings indicate cross-cultural disagreement on the absolute level of social status given to both types of relational conditions: American subjects in both relational conditions generally ranked the relationship to be characterized by higher disparity than Korean subjects (M=3.87, M=2.78, respectively, E--4.25, Qf-53, p<.000; _=-.51). Regarding social distance among conversational partners (see Table 10), Korean subjects in the student-professor condition (M-3.59) rated the relationship to be less distant than the U.S. subjects (M=4.38, £--3.43, p-.001, £=-.43). In the friend-friend condition, however, there were no significant differences between the Korean and the U.S. subjects in their perceptions of social distance (M-2.62, M-2.45, §--.48, p<.631, £--.07). Across all situations, however, the U.S. subjects consistently viewed the relationships to be more distant than Korean subjects (M-3.74, M-3.27, E—-2.24, df—53, £--.29, p<.029). In sum, Korean and the U.S. subjects were similar in their perceptions of request size across the situations. On the other hand, the U.S. subjects perceived the relationship (both professor- student and friend-friend conditions) to be characterized by higher power disparity than the Korean subjects. Finally, American subjects 54 Table 9 Cross-cultural Comparisons of Perceived Social Status in Request Situations (7=high disparity, l=low disparity) Korean The U.S. Participants Participants (N=25) (N=30) (l) Professor-Student Situations Goals Mean SD Mean SD 5 p E 5:11.; """" 3'55} """" 1’91 """ 2.13% """" 5%"'Ié'éé"6éi"2'£.%* Homework 3.80 1.57 4.95 1.19 -3.01 .004 -.38* Research 3.12 1.69 4.12 1.15 -2.51 .016 -.32 Delay 3 18 1 89 4.05 1 69 -1 78 081 — 23 'ééfiiééfi'fiiié """" I56 """ 2.3; """" é;"'25'éé"6éé"l';i. Borrow 1 28 1 23 2.32 1 42 -2 91 005 - 37* Repay 1 74 l 41 2.80 l 50 -2 70 009 - 34* sub-total 1.51 1 17 2.56 1 14 -3 36 001 - 42* Total 2 78 1 02 3.87 73 -4 25 000 - 51* Note: * - Significant at .01 level. 55 Table 10 Cross-cultural Comparisons of Perceived Social Distance in Request Situations (l-low, 7-high) Korean The U.S. Participants Participants (N-25) (N-30) (l) Professor-Student Situations ................................................................... Goals Mean SD Mean SD 5 p 5 iii; """" 3'}; """" i116 """ 2.35 """ i'i;"'i3’£é"6éé"l';i. Homework 3.92 1.18 5.08 .81 -4.18 .000 -.50* Research 3.16 1.16 3.47 .94 -1.05 .298 -.14 Delay 3.52 1.31 4.12 1.41 ~l.63 .110 -.22 sub-cotai"3'éé """" I6} """ 2.35 """" éé"'35';;"ééi"3';é' Borrow 2.66 1.78 2.38 1.48 .62 .538 .08 Repay 2.58 1.33 2.53 1.88 .10 .918 .01 sub-total 2.62 l 37 2.45 l 24 - 48 631 - 07 Total 3 27 91 3.74 61 -2 24 029 - 29 Note: * - Significant at .01 level. 56 perceived the relationship between interactional partners to be more distant than Korean subjects. Results of the Main Study Question l-A (Comparison of clustering of tactics): In order to test whether tactics cluster together across the three main strategy categories (hint, query, and direct statement), confirmatory factor analysis was conducted with the CFA subroutine of the PACKAGE computer program (Hunter & Lim, 1987) for each of the constraints and for perceived likelihood of use. To test the fit of the model to the data, the observed correlations among the strategies within a category must conform to the product rules of internal consistency (Hunter & Gerbing, 1982). In the initial tests of internal consistency, quite a number of significant deviations were found across cultures in "question hint" in the Hint category and in "syntactic downgraders" in the Query category. After examining the patterns of deviations, the two tactics (question hint and syntactic downgraders) were exchanged to be in the other factor. The examination of the contents of the two tactics seems to make this move necessary, for the following reasons: (1) question hints (e.g. "Do you remember the money I lent you?") take the form of an interrogation which can be seen to be closer to Query strategy than to Hint; and (2) Syntactic downgraders (e.g. Would it be all right if I ask you to repay the loan?) seem to fit better than question hint in the hierarchy of increasing directness, as they modify the utterance internally by mitigating the requestive force of the utterance by means of syntactic choices. With this change in the factor structure, a second run of 57 internal consistency tests was conducted. 0n the second run, deviations between observed and reproduced correlations were minute in both cultures. Only 8 out of 372 observed correlations significantly deviated from the expected correlation at .01 level. This shows that the items in each strategy category represent the same underlying dimension. While the items in three strategy categories form internally consistent scales, the alpha levels for some of the categories were quite low. Specifically, across the two cultures, the average reliability coefficient alphas for these factors range from .59 to .73 (M-.66) for Koreans and from .62 to .80 (M-.72) for Americans (see Table 11). Table ll-a through ll-f present the 12x12 item correlation matrices for each of the constraints. While a number of strategy categories yielded relatively low reliabilities (especially hint category), it seems to be partly attributable to the magnitude of variance for ratings of each tactic. Correlation between standard deviations of the ratings of tactics and corresponding reliabilities yielded relatively weak, but significant effect sizes (£-.20 for Koreans, and r-.22 for Americans). In short, the relatively low reliabilities in some tactics seem to be caused by the low variability in the subject’s ratings of tactics. The low variances for the ratings of each tactic are partly due to the high agreement between peoples' perceptions of tactics in the light of each constraint. Overall, the results seem to suggest that items in the three major clusterings (with minor changes) form internally consistent scales. The relatively low alpha levels (especially in hint 58 Table 11 Coefficient Alphas for Three Strategy Categories Coefficient Alphas Interactive Constraint Korea U.S. Total 65 60 Hint 36 30 Clarity ----------------------------------------- Query 76 75 Direct Statement .84 .76 Total 65 77 Hint .56 70 Concern for ----------------------------------------- feelings Query .65 .77 Direct Statement .75 .83 Total 73 80 Hint 69 78 Avoiding ----------------------------------------- Imposition Query .72 .82 Direct Statement .78 .79 Total 71 80 Hint .66 82 Avoiding ----------------------------------------- Being Query .75 .81 Disliked ----------------------------------------- Direct 59 Table 11 (continued) Coefficient Alphas Interactive Constraint Korea U.S. Total 63 73 Hint .48 61 Effectiveness ---------------------------------------- Query 65 77 Direct Statement .75 .82 Total 59 62 Hint .45 56 Likelihood ----------------------------------------- of Use Query .59 .52 Direct Statement .71 .77 Grand Average .66 .72 * Note: Question hint and syntactic downgrader have been exchanged to be in the other factor. 60 Table ll-a Correlations among 12 Tactics: Clarity Key: Strategy 1 1-mild hint 2-strong hint 3-syntactic downgraders Strategy 2 4-question hint 5-permission 6-ability query-preparatory 7-willing query-preparatory 8-suggestory Strategy 3 9-want lO-performative ll-obligation 12-imperative (Korean Sample) 1 _- 2 24 _- 3 1o 14 -- 5 -0 21 ll 51 -- 6 2 15 12 38 51 -- 7 -9 11 11 23 32 59 -- 8 '4 14 l 28 28 41 39 -- 11 -9 . 6 3 -2 '11 21 41 24 61 53 -— Note: Correlations corrected for attenuation. 61 Table lloa (continued) (U. S . sample) 1 -- 2 2 -- 3 8 27 ~- 4 -20 8 26 ~- 5 ~20 3 -2 49 -- 6 -12 8 21 52 47 -- 8 —6 l9 l9 14 24 36 34 -- 9 -4 5 19 20 31 48 51 48 ~- 10 ~13 l4 5 9 19 31 24 50 37 -- Note: Correlations corrected for attenuation. 62 Table ll-b Correlations among 12 Tactics: Concern for the other's feelings (Korean sample) 1 -- 2 34 -- 3 31 25 -- 4 24 20 34 ~- 5 9 l 13 53 -- 6 14 6 12 34 42 -- 7 8 3 l7 l4 17 35 ~-- 8 1 -2 -16 11 16 27 17 -- 10 -7 6 7 ll 19 18 12 37 32 -- 11 -7 -1 -ll 1 13 9 19 41 47 62 ~- Note: Correlations corrected for attenuation. 63 Table ll-b (continued) (U.S. sample) 1 -- 2 52 -- 3 36 44 -- 4 38 43 56 -- 5 33 . 38 33 63 -- 6 3s 34 34 43 65 -- 7 28 34 24 32 s4 s3 -- 8 15 3o 4 10 20 30' 26 -- 9 21 30 18 27 33 40 64 31 -- Note: Correlations corrected for attenuation. 64 Table ll-c Correlations among 12 Tactics: Avoiding Imposition (Korean sample) 1 -- 2 57 -- 3 35 36 -- 4 18 27 42 -- 5 2 17 32 62 -- 6 11 14 21 39 51 -- 8 8 l4 -1 19 10 22 18 -- 9 4 10 9 7 18 20 32 51 -- Note: Correlations corrected for attenuation. 65 Table ll-c (continued) (U.S. sample) 1 -- 2 64 -- 3 40 56 -- 4 40 49 57 ~- 5 25 36 37 65- -- 6 29 37 32 56 69 -- 7 25 36 28 44 51 66 -- 9 17 29 18 31 36 46 66 26 -- 10 2 13 7 23 29 42 43 60 4O -- 11 5 12 6 15 19 25 33 35 55 54 -- Note: Correlations corrected for attenuation. 66 Table ll-d Correlations among 12 Tactics: Avoiding Dislike (Korean sample) 3 29 39 -- 4 29 33 51 -- 5 21 20 32 58 -- 6 17 11 17 36 48 -- 7 10 13 25' 3o 39 52 -- 10 4 l8 19 19 30 21 14 40 36 ~- 11 -3 4 7 8 16 12 28 34 37 62 -- Note: Correlations corrected for attenuation. 67 Table ll-d (continued) (U.S. sample) 1 -- 2 68 -- 3 49 63 -- 4 48 54 62 -- 5 38 51 43 67 -- 6 31 45 40 57 69 -- 7 30 40 31 37 54 63 -- 8 16 27 12 20 31 34 26. -- 9 29 32 20 26 34 39 66 33 ~- 10 24 30 18 23 40 40 37 7O 42 ~- 11 1 l3 -1 10 21 21 32 34 50 49 -- Note: Correlations corrected for attenuation. 68 Table ll-e Correlations among 12 Tactics: Effectiveness (Korean sample) 4 19 15 28 -~ 5 4 1 23 49 -- 6 11 4 19 36 48 —- 7 6 l 15 12 23 32 -- Note: Correlations corrected for attenuation. 69 Table ll-e (continued) (U.S. sample) ------- ...................................................... 1 -- 2 38 -- 3 23 42 ~- 4 4 28 43 -- 5 3 15 18 69 -- 10 -11 7 -3 3 18 17 22 57 39 -- Note: Correlations corrected for attenuation. Table ll-f Correlations among 12 Tactics: Likelihood of Use (Korean sample) 1 -- 2 30 —- 3 16 19 -- 4 18 14 25 ~- 5 6 10 13 51 -- 7 -14 6 21 4 20 19 -- 8 10 13 -12 l9 19 20 13 ~- 10 -6 2 15 8 15 4 18 26 34 -- Note: Correlations corrected for attenuation. Table ll-f (continued) (U.S. sample) 1 -- 2 36 -- 3 24 29 -- 4 21 20 34 -- 5 l7 l3 16 58 -- 8 1 19 -2 -8 -3 12 -3 -- 9 -2 12 6 -13 -9 22 51 16 -- 10 -2 18 -l -24 -8 0 7 50 23 —~ 11 -6 l7 -7 -30 -17 -1 7 37 41 54 -- Note: Correlations corrected for attenuation. 72 categories) show that the items in those categories may not be strongly related, even if all the items correlated with each other in the similar way.6 Question l-B (The perceived placement of request strategies along the five dimensions of constraints): The orderings of constraint judgments were predicted to be similar across cultures, even though there might be culture-specific differences in the mean ratings of judgments.7 To examine whether the ratings of tactics' follow a similar order across cultures, mean values obtained for each of the 12 tactics were converted to ranks in each culture. Then the two columns (one based on the ratings by the U.S. participants and the other by the Korean participants) of 12 values were correlated. That is, the rank-order coefficient on each interactive constraint judgment was calculated, based on mean values obtained for each of the 12 tactics from each culture (see Table 12-a through 12-f). The obtained rank order correlations across cultures range from .63 to .98. The highest concordance between the two cultures was found in "Avoiding Imposition" (rho-.98, p-12, p<.01), followed by "Concern for the other's feelings" (rho-.80, p-12, p<.01), "Likelihood of Use" (rho-.80, p§12, p<.01), "Concern for disapproval" (rho-.76, 5&12, p<.01), and "Concern for clarity” (rho-.63, p-lZ, p<.01). The rank order of effectiveness judgments was not statistically significant (rho-.26, p-12, p >.01). The results of this analysis generally support strong cross-cultural similarity in the ordering of interactive constraints at the level of aggregate data, with the exception of effectiveness judgments. In order to highlight the specific cross-cultural similarities 73 Table 12—8 Ratings of Request Tactics: Clarity (l-low, 7-high) Korea U.S. Strategy (N-294) (N-295) (1) Hint Mean SD Mean SD t p r Mild hint 3.96 (2) 1.82 4.45 (2) 1.84 -3.27 .001 -.13 Strong hint 4.22 (4) 1.83 4.63 (3) 1.73 -2.84 .005 -.12 Syntactic downgrader 3.92 (1) 1.74 4.89 (5) 1.54 -7.21 .000 -.29 4 03 1.18 4 66 96 E-value - -7.ll (df-587, p<.000), r--.29 Korea . U.S. (N-290) (N-295) (2) Query Mean SD Mean SD t p r Question hint 4 1 3 1 Permission 4. l. 4 1. . Ability 4.71 (8) 1.80 5.59 (9) 1.39 -6.69 .000 -.27 Willingness 5 1 4 1 Suggestory 4 l 4 1 E-value - -2.65 (dfa583, p<.008), r--.ll Korea U.S. (N-291) (N-298) (3) Direct Statement Mean SD Mean SD t p r Want 4 54 (6) 1.86 5.69 (10) 1.30 -8.70 .000 -.34 Performative 4.77 (9) 1.90 5.81 (11) 1.36 -7.67 .000 -.30 Obligation 5 37 (11) 2.02 5.47 (8) 1.70 -.65 .513 -.03 Imperative 5 79 (12) 1.91 5.94 (12) 1.65 -.99 .322 -.04 5 12 1 47 5 73 1 17 E-value - -5.55 (df-587, p<.0001), r--.22 ** Rank-order correlation between two cultures: rho-.63 74 and differences in the mean levels of tactical judgments, the judgments of tactics were summarized across the three main strategy types. First, the mean levels of tactics on the clarity scale, as shown in Table 12-A, reveal that both Korean and American participants agree in rating the imperative as the clearest way to make a request. Looking at the averages across three main strategy types, the results point to some cross-cultural disagreement on the absolute levels of clarity across all three categories: Hint (£-- 7.11, df-587, p<.000), Query (E--5.16, df-585, p<.0001) and Direct Statement (£--5.55, df-587, p<.0001) categories. Overall, the U.S. participants were found to perceive tactics as clearer (to the point) than Korean participants. On a specific tactic level, 10 out of 12 tactics showed the same trend. In sum, both cultures shared similar perceptions of the relative clarity of the 12 tactics but differed more in their perceptions of the absolute clarity of individual tactics. Earlier it was mentioned that there exists an almost perfect rank-order correlation (rho—.98) in the ratings of perceived imposition of tactics across cultures. The ratings of the perceived imposition reported in Table 12-8 show that there exists some disagreement between Korean and the American subjects on the absolute degree to which an utterance burdens the hearer. The most highly "imposing" strategy for both Korean and American subjects was Direct Statement (M-4.45, SD-l.21,'M=4.47, SD-l.14, respectively). There were no significant cross-cultural differences in the mean ratings of tactics in Direct Statement category (p-.26, gg-591, NS). On the other hand, Korean subjects perceived both Hint and Query categories Table lZ-b 75 Ratings of Request Tactics: Perceived Imposition (1-low, 7-high) Korea U.S. Strategy (N-295) (N-297) (1) Hint Mean SD Mean SD Mild hint 2.70 (12) 1.72 1.79 (12) 1. Strong hint 2.94 (9) 1.48 2.36 (8) l. Syntactic downgrader 2.71 (11) 1.37 1.82 (11) 1. 2.73 . 1.15 1.99 1 E-value - 7.88 (df-590, p<.000), r-.3l Korea U.S. (N-295) (N—298) (2) Query Mean SD Mean SD Question hint 3.17 (7) 1.56 2.23 (9) 1. Permission 2.93 (10) 1.47 2.04 (10) 1. Ability 3.06 (8) 1.48 2.47 (7) 1. Willingness 3.86 (5) 1.55 3.26 (5) 1. Suggestory 4.12 (4) 1.46 4.07 (4) 1. 3.43 1.00 2.81 1 E-value - 6.90 (df—591, p<.000), r-.27 Korea U.S. (N-295) (N-298) (3) Direct Statement Mean SD Mean SD Want 3.56 (6) 1.63 3.22 (6) 1. Performative 4.22 (3) 1.64 4.15 (3) 1. Obligation 4.95 (2) 1.45 5.07 (2) l. Imperative 5.06 (1) .l.61 5.47 (l) 1. 4.45 1.21 4.47 1 ** Rank-order correlation between two cultures: rho-. E-value - -.26 (df-S9l, NS), r-.01 98 76 as more highly "imposing” than American subjects (g—7.88, gg-59o, 5r.31, p$.000; £—6.90, gg-591, 5-.27, p<.000, respectively). All the specific tactics within the two strategy categories followed the same trend. In summary, the results show that the relative ranks of the tactics along the scale of imposition follow a nearly identical pattern in both cultures, although the two cultures seem to rate the specific absolute levels of imposition differently. Regarding the third interactive constraint (Concern for the hearer's feelings), examination of Table 12-C shows that for both Koreans and Americans, the consideration for the other's feelings increases in the following order: Hint, Query, and Direct Statement. However, American subjects (M-5.66, SD-1.05; M-5.23, SD-.99; M-3.54, SD-l.31, respectively for Hint, Query and Direct Statement) perceived all three strategies as showing more consideration towards the other's feelings than Koreans (M-5.07, SD-1.04; M-4.34, SD-.94; M-3.23, SD-1.24). This trend was found in 11 out of the 12 tactics. Thus, the perceptions regarding the absolute level to which an utterance shows considerations for the other's feelings seem to be culture-specific. Partly, these differences in the mean ratings can be attributed to the fact that Koreans may possess higher threshold for "acceptable" levels of consideration for the other's feelings (see Kim, in press a). Table 12-D presents the ratings regarding the degree to which request tactics risk disapproval for self. As expected, the Direct Statement category was ranked the highest in risking devaluation in both cultures (M-3.80, SD-1.20; M-3.81, SD—l.19, respectively Korean and American subjects; £--.13, gfi-591. NS). In addition, both groups Table 12-c Ratings of Request Tactics: Consideration for the Other's Feelings (l-low, 7-high) Korea U.S. Strategy (N-295) (N-298) (1) Hint Mean SD Mean SD t p r Mild hint 5.23 (12) 1.49 5.63 (9) 1.36 -3.41 .001 -.14 Strong hint 5.06 (11) 1.40 5.29 (7) 1.46 -1.94 .052 -.08 Syntactic downgrader 4.91 (9) 1.49 6.05 (12) 1.12 -10.55 .000 -.40 5 07 1.04 5 66 1 05 E-value - -6.83 (df-59l, p<.000), r--.27 Korea U.S. (N-295) - (N-298) (2) Query Mean SD Mean SD t p r Question hint 4.44 (6) 1.74 5 54 (8) 1 SO -8 27 .000 - 32 Permission 4.84 (8) 1.43 5 99 (ll) 1 09 -11 05 000 - 41 Ability 5.03 (10) 1.40 5.66 (10) 1.19 —5.96 .000 -.24 Willingness 3.78 (5) 1.64 4 91 (6) l 43 -8 94 .000 - 35 Suggestory 3.58 (4) 1.70 4 O3 (3) l 55 -3 37 .000 - l4 4 34 .94 5 23 99 g-value - -11.28 (df-591, p<.000), r--.42 Korea U.S. (N-294) (N-299) (3) Direct Statement Mean SD Mean SD t p r Want 4.64 (7) 1.62 4.83 (5) 1.62 —l 46 146 - 06 Performative 3.39 (3) 1.76 4.08 (4) 1.63 -4.96 .000 -.20 Obligation 2.42 (1) 1 54 2.97 (2) 1 59 -4 32 .000 - l7 Imperative 2.48 (2) 1 77 2.27 (l) l 50 3 52 .000 14 3 23 1.24 3 54 1 31 E-value - -2.98 (df-591, p<.003), r--.12 ** Rank-order correlation between two cultures: rho-.80 Table 12-d Ratings of Reguest Tactics: Risking Disapproval for self (1-1ow, 7-high) Korea U.S. Strategy (N—293) (N-299) (1) Hint Mean SD Mean SD t p r Mild hint 1.61 (12) 1.37 1.82 (8) 1.60 -1.69 .091 -.07 Strong hint 1.97 (11) 1.43 2.15 (7) 1.58 -1.41 .159 -.06 Syntactic downgrader 2.06 (9) 1.47 1.42 (12) 1.27 5.70 .000 .23 1 89 1 08 1 80 1 25 E-value - .97 (df-590, p<.334), r-.04 Korea . U.S. (N—294) (N-299) (2) Query Mean SD Mean SD t p r Question hint 2 1 l 1 Permission 2 l l 1 Ability 2.04 (10) 1.41 1.82 (9) 1.25 2.08 .038 .09 Willingness 2 1 2 1 Suggestory 3 l 3 1 E‘value - 4.59 (df-59l, p<.000), r-.19 Korea U.S. (N-295) (N-298) (3) Direct Statement Mean SD Mean SD t p r Want 2.45 (7) l 61 2.34 (6) 1.55 - 44 662 - 02 Performative 3.77 (3) 1.71 3.37 (3) 1.56 2.97 .003 .12 Obligation 4.46 (2) 1 50 4.38 (2) 1 51 70 .483 03 Imperative 4.53 (l) l 71 4.97 (l) l 37 -3 52 000 - l4 -- 3 80 1 20 3 81 1 19 E-value - -.13 (df-591, NS), r--.01 ** Rank-order correlation between two cultures: rho-.76 79 of subjects rated the Hint strategy (M-1.89, SD-l.08; M-l.80, SD-l.25) lower than the Query strategy (M-2.62, SD-1.00; MF2.23, SD-1.07) in causing dislike or devaluation by the hearer. Mean levels for the Hint strategy did not differ significantly across cultures (£--.97, d§e590, NS). On the other hand, Korean subjects perceived the Query strategy as leading to more disapproval from the other person than did American subjects (5&4.59, g§2591, p{.000), even though the effect size was rather small (£-.l9). This trend held true for all the tactics within the query category. The most striking cross-cultural differences in strategy ratings were found in effectiveness judgments. The findings show a marked disagreement between Korean and the U.S. subjects on the relative effectiveness perceived for three strategies (see Table 12-E). For the U.S. subjects, both Query and Direct Statement categories (M-4.65, SD-l.09; M=4.70, SD—l.50, respectively) were judged to be more effective than the Hint category (M-3.95, SD-1.19). On the other hand, for Korean subjects, Direct Statement (M-3.67, SD-l.30) was the least effective strategy, while the Hint and Query categories were perceived to be equally effective (M-4.15, SD-l.15, M-4.15, SD-1.02, respectively). Overall, it seems that the most noticeable difference in effectiveness judgments was that the U.S. subjects consider direct statements the most effective way of making a request, while Korean subjects rate it as the least effective. Examination of Table 12-F shows that perceived likelihood of use ’ is rated in similar fashion across the two cultural groups: for both groups of participants Hint is the most preferred strategy type (M-4.61, SD-l.25; M—4.74, SD-1.28). Query was the second most likely 80 Table 12-e Ratings of Request Tactics: Effectiveness (1-low, 7-high) Korea U.S. Strategy (N-294) (N-296) (1) Hint Mean SD Mean SD t p r Mild hint 4.43 (10) 1.69 3.90 (2) 1.82 3.65 .000 .15 Strong hint 4.32 (9) 1.60 4.15 (4) 1.46 1.36 .174 .06 Syntactic downgrader 4.14 (7) 1.67 4.58 (8) 1.49 -3.34 .001 .14 4 30 l 17 4.21 1 11 E-value - 1.00 (df-588, p<.3l9), r-.04 Korea U.S. (N—294) (N-298) (2) Query Mean SD Mean SD t p r Question . hint 3.71 (5) 1.66 3.80 (l) l 49 - 70 .487 - 03 Permission 4.15 (8) 1.58 4.76 (9) 1.41 -4.90 .000 -.20 Ability 4.74 (11) 1.47 5.09 (10) 1.36 -3.03 .003 -.12 Willingness 4.07 (6) 1.56 4.57 (7) 1 47 -4.02 000 - 16 Suggestory 3.66 (4) 1.71 4.28 (6) 1 53 -4.65 000 - 19 4 11 89 4.52 88 E-value - ~5.69 (df-590, p<.000), r--.23 Korea U.S. (N-295) (N-293) (3) Direct Statement Mean SD Mean SD t p r Want 4.78 (12) 1.62 5.33 (12) 1.48 -4.36 000 - l8 Performative 3.43 (3) 1.77 5.13 (11) 1.49 -12.69 000 - 46 Obligation 3.16 (1) l 86 4.26 (5) 2.06 -6 81 .000 - 27 Imperative 3.38 (2) 2 02 4.09 (3) 2.30 «3 98 .000 - l6 ......................................................................... E-value - -8.78 (df-591, p<.0001), r--.34 ** Rank-order correlation between two cultures: rho-.26 Table 12~f 81 Ratings of Reguest Tactics: Likelihood of Use (l-low, 7-high) .......................................................................... .612 .016 .000 -.22 24 (df-587, ........................................ 4-.------_------_--—_-----------. Korea Strategy (N—295) (1) Hint Mean Mild hint 4.87 (10) Strong hint 4.85 (9) Syntactic downgrader 4.12 (8) 4.61 E-value - -1. Korea (N-296) (2) Query Mean Question hint 4.08 (6) Permission 4.10 (7) Ability 4.89 (11) Willingness 3.85 (5) Suggestory 3.31 (4) 4.05 E-value - -2. Korea (N-296) (3) Direct. Statement Mean Want 4.37 (12) Performative 2.72 (3) Obligation 2.37 (1) Imperative 2.57 (2) 3.00 ** Rank-order correlation between two cultures: 1.30 E-value - -2.05 (df-59l, p<.04l), r--.08 U.S. (N-294) Mean SD 4.80 (10) 1.79 4.50 (6) 1.77 4.94 (11) 1.85 4.74 1 28 p<.216), r--.05 U.S. (N-296) Mean SD 4.56 (8) 1 87 4.75 (9) 1 80 5.06 (12) 1.57 3.68 (5) l 78 3.17 (3) l 93 4.25 1.08 p<.027), r-.09 U.S. (N-297) Mean SD 4.52 (7) 2.06 3.46 (4) 1.94 2.71 (2) 1.98 2.29 (l) 1.89 3.24 1.52 rho-.80 82 to use strategy, with a mean of 4.05 (SD-1,05, for Koreans) and 4.25 (SD-1.08, for Americans) and Direct Statement strategy was the least likely to use (M-3.00, SD-l.30; M-3.24, SD-1.52). There were no significant cross-cultural differences in the mean levels of likelihood of use in Hint category (M-4.60, SD-1.28, M-4.6l, SD-1.32, £-—.12, af-586, NS). Korean subjects, however, were found to be less willing to use both Query and Direct Statement categories than Americans (§--2.97, gg—59o, p<.003, £--.12; g—-2.05, af-591, p<.041, £--.08), even though the effect sizes are small. Question 1-C (Incompatibility among the constraints): In order to check whether subjects from Korea and the U.S. perceive a similar degree of incompatibility among the constraints, rank order correlations among mean tactic ratings on the five constraint judgments were calculated separately for each culture (see Table l3-A and 13-B). These analyses were performed on the aggregate level data (E - 12 tactics). First of all, for Korean subjects, rank-order correlations among interactive constraints ranged from perfectly compatible (rho-.1.00, 9-12, p<.01) to highly incompatible (rho--.92). The ratings of clarity yielded a consistently negative relationship with the rest of the constraints (rho--.80, -.92, -.80, -.59; all significant at .01) and the likelihood of use (-.64, a-12, p<.01). The four interactive constraints and the likelihood of use ratings correlated highly with each other (see Table l3-a). In sum, the judgments of interactive constraints by Koreans seem to reduce to a single dimension bounded by "clarity" at one end and "prosocial," "effective," and "likely to be used" at the other end. 83 Table l3-a Degrees of Incompatibility among Interactive Constraints and the Likelihood of Use: Korean Sample Concern for Avoiding Avoiding Effective- Clarity Feelings Imposition Dislike ness Clarity -- Concern for feeling -.80* -- Avoid Imposition -.92* .93* -- Avoid Dislike -.80* 1.00* ,93* -_ Effectiveness -.59* .86* .73* .86* -- Likelihood of Use -.64* .87* .76* .87* .99* *: Significant at .05 level Note: correlations calculated from the rank-orderings of tactics 84 Table l3—b presents the rank correlation coefficients among the five constraints and the likelihood of use for American subjects. The three most striking differences from the results on Korean sample were found: (1) in the relationship between clarity and effectiveness judgments; (2) in the relationship between face-oriented constraints and effectiveness; and (3) in the relationship between effectiveness and likelihood of use. The Korean sample perceived clearer tactics to be less effective than indirect tactics (rho=-.59, p=l2, p<.01), while among Americans clarity was positively correlated with effectiveness (rho=.60). The Korean sample perceived "prosocial" tactics (tactics which showed concern for the other's feelings and autonomy and for the speaker's own image) to be most effective, while the Americans saw those constraints as being unrelated to effectiveness. Finally, the relationship between effectiveness and likelihood of use was almost perfect for Koreans, while American subjects yielded relatively low compatibility (rho-.20, p—lZ, NS) between effectiveness judgments and the likelihood of use ratings (rho-.99, p-12, p<.01; 5:32.49, p<.OOl). Question 2-A (Perceived importance of each constraint): It was predicted that cultures may differ in the relative importance attached to interactive constraints. To test whether the perceived importance of constraints differs significantly across cultural groups, a series of t-tests on the importance ratings for each of the five constraints was conducted. Table 14 presents the means, standard deviations, and t-values for the ratings of perceived importance of the five constraints. In general, Americans gave clarity a much higher value than did the Koreans. On the other hand, 85 Table l3-b. Degree of Incompatibility among Interactive Constraints and the Likelihood of Use: The U.S. Sample Concern for Avoiding Avoiding Clarity Feelings Imposition Dislike Effectiveness Clarity -- Concern for feeling -.44 -- Avoid Imposition -.66* .91* -- Avoid Dislike -.53* .97* .92* -- Effectivenss .60* .13 -.12 . .06 -- Likelihood of Use -.36 .94* .87* .91* .20 *: Significant at .05 level Note: correlations calculated from the rank-orderings of tactics 86 Table 14 Ratings of Perceived Importance of the Five Constraints (l=low, 7=high) N Mean SD t-value r Korea 295 4 15 1 81 Clarity ------------------------------ -14.94 (p<.0001) - 52 U S 299 6.03 l 21 Concern Korea 295 5.36 1.29 for ------------------------------ 3.23 (p< 001) .13 feelings U.S. 299 4.96 1.65 Avoiding Korea 293 5.14 1.57 imposition ------------------------------ 8.05 (p<.0001) .32 U S 299 4.05 1 72 Concern about Korea 295 4.96 1.53 disapproval ------------------------------ 1.06 (NS) .04 U S 298 4.83 l 58 Effectiveness Korea 295 5 37 l 49 ------------------------------ -l.46 (NS) - 06 87 Koreans seem to allow face-maintaining considerations to play a more central role in interaction: Korean subjects rated the "Concern for the other's feelings" as more important than American subjects. Korean participants also rated minimizing imposition as higher in importance than the U.S. participants. The effect size for the cross-cultural differences for the importance of imposition (£=.32) was much higher than that of concern for the other's feelings (£-.l3). The differences in the importance ratings of the effectiveness and concern about disapproval were not statistically significant. Overall, the findings seem to suggest that cultural factors may explain differences in the perceived importance ratings of at least three out of the five interactive constraints. Question 2-B (the moderating effect of the level of social status on the perceived importance of each constraint): In order to investigate the effects of social status on the ratings of perceived importance, the data were analyzed by means of a 2 x 3 ANOVA. The factors were nationality (Korea/U.S.) by requestee's social status relative to the requestor's (high/equal/low). In each ANOVA, the perceived importance of each of the five constraints served as the dependent variable. Tables 15-A through lS-E present descriptive statistics for the cross-cultural variations in perceived importance of constraints across the three status levels. A two-way (culture x social status) analysis of variance for the rated importance of clarity yielded a significant main effect of cultural factor (Eé236.46, gf-l/582, p<.000, eta squared-.27). The effect of social status on the rated importance of clarity, though 88 statistically significant, was negligible (E=6.65, g§=2/582, p<.001, eta squared=.02) (see Table 15-A). A two-way interaction effect between culture and social status was also minute (F=8.63, af=2/582, p<.000, eta-squared=.02). Overall, Korean subjects rated the importance of clarity lower in expressing their intentions to persons of higher social status (M=3.53, SD=1.75) than to persons of equal or lower status (M=4.69, SD=1.70; M=4.21, SD=l.82, respectively). Post hoc analysis using Tukey-HSD procedure indicated that while there was no difference between equal and higher status persons (p <.05), there were statistically signifcant differences between lower and equal status (p <.05) and lower and higher status (p<.05). On the other hand, differences in rated importance of clarity across social status were negligible for American subjects (.10 in a 7-point scale). Expectedly, Tukey-HSD test showed that no two groups were significantly different at the .05 level. Ratings of the importance of being considerate to the other's feelings follow a similar trend (see Table 15-B). First, the main effect of culture on the importance of showing consideration was statistically significant (F—11.l7, afxl/582, p<.001, eta- squared-.02) and interaction between culture and social status was also significant (E-8.14, af-2/582, p<.000, eta-squared=.01). For American subjects, the importance of showing consideration for the other's feelings increased as the other person's status increased (M-4.52, 5.11, 5.25, respectively for lower, equal, and higher status). The post hoc analysis using Tukey-HSD test revealed that there were statistically significant differences between the lower and equal status (p <.05) and lower and higher status (p<.05). 89 Table 15-a Descriptive Statistics for the Perceived Importance Ratings: Clarity Culture Korea U S M-4.21 M=6.13 M=5.18 Low SD=1.82 SD=1.3O SD=1,85 N=97 N=100 N=l97 Requestee M-4.69 M=5.94 M=5.31 Social Equal SD=1.70 SD=l.l7 SD=1.59 Status N=100 N=99 N=l99 M—3.53 M=6.03 M-4.79 High SD=1.75 SD=1.14 SD=1.93 N-98 N=100 N=l98 M-4.15 M=6.03 SD-1.81 SD-l.21 N-295 N=299 Table 15-b Descriptive Statistics for the Perceived Importance Ratings: Concern for feelings Culture Korea U S M=5.46 =4.52 M=4.99 Low SD-l.15 SD-l.63 SD-l.49 N=97 N=100 N=197 ‘ Requestee M-5.54 M=5.11 M=5.33 Social Equal SD=1.30 SD=1.73 SD=1.54 Status N=100 N=99 M=l99 M-5.06 M=5.25 M-5.l6 High SD-l.38 SD-l.49 SD-l.44 N-98 N=100 N=l98 M-5.36 M-4.96 SD-1.30 SD-1.65 N-295 N-299 91 In contrast, Korean subjects rated level for consideration for the other's feelings as being more important when the other was at equal or lower status rather than of higher status. The mean differences between equal and higher status, as well as between lower and higher status were statistically significant at .05 level. Descriptive statistics for the importance ratings of avoiding imposition across status levels are shown in Table 15—C. A two-way ANOVA for perceived importance of avoiding imposition showed significant main effects of culture (F=84.93, af=l/580, p<.000) and social status (£212.67, gf=2/580, p<.000), and an insignificant interaction effect between these two variables (§=2.69, af—2/580, p<.068). The main effect of culture (eta-square=.10) was three times larger than that of social status (eta-square=.03). Specifically, Koreans (M=5.14, SD=1.57) tended to attribute higher importance than Americans to avoiding imposition across all the three status levels (M—4.05, SD-1.72). Regarding the main effects of social status, subjects in the bearer-high status condition assigned lower importance ratings for nonimposition (M=4.18, SD=1.84) than subjects in the bearer-equal (M-4.83, SD-1.82) or hearer-low status conditions (M-4.76, SD-l.47). Pairs of groups significantly different at the .05 level were found between equal and higher status as well as between lower and higher status. The difference between lower and equal status conditions was not statistically significant (p <.05). As for the importance of avoiding devaluation (see Table 15-D), the influence of social status was significant (E-24.16, gf-2/581, p<.000, eta-square-.07). In contrast, the main effect of culture (£- 1.21, af-l/581, p<.271) was not statistically significant, and the 92 Table 15-c Descriptive Statistics for the Perceived Importance Ratings: Avoiding Imposition Culture Korea U S M=5.17 M=4.36 M=4.76 Low SD-l.39 SD=l.44 SD=1.47 N-96 N=100 N-196 Requestee MF5.56 M=4.10 M=4.83 Social Equal SD-l.52 SD=1.80 SD=1.82 Status N-lOO N=99 N=l99 M—4.68 M-3.69 M=4.18 High SD-l.69 SD-l.83 SD-l.83 N-50 N-50 N-197 Mr5.l4 Ms4.05 SD-l.57 SD-1.72 N-293 N-299 93 Table 15-d Descriptive Statistics for the Perceived Importance Ratings: Avoid Devaluation Culture Korea U S M-5.l7 M=5.35 M=5.26 Low SD-l.37 SD=1.47 SD=1.42 N=97 N=lOO N=l97 Requestee M-5.39 M=4.79 M=5.09 Social Equal SD=1.48 SD=l.58 SD=l.55 Status N=100 N=98 N=198 M-4.32 M=4.33 M=4.33 High SD-l.53 SD-l.53 SD-l.53 N-98 N-lOO N-198 M-4.96 M-4.83 SD-l.53 SD-1.58 N-295 N-298 94 effect size was small (eta-square=.04). The interaction between culture and social status, though statistically significant (§= 3.90, af-2/581, p<.021), also was small (eta-squared=.01). Regarding the effect of social status, both Korean and the U.S. subjects showed more concern for not being negatively judged by lower status persons than by higher status persons (§=49.94, af=2/581, p<.000) (see Table lS-D). A Tukey-HSD follow-up test was performed among the three status conditions within culture. This test indicated that there were significant differences between higher and lower conditions (p <.05) across both cultures. Interestingly, for the Korean subjects, the highest mean importance for avoiding devaluation was in the equal 'status condition. In short, social status yielded a stronger influence than culture on the importance rating of avoiding devaluation. Table 15-E presents descriptive statistics for the rated importance of effectiveness. The two-way analysis of variance showed that the effect of culture was not significant (F—2.l6, af-2/58l, p<.l42). While the analysis revealed a significant effect of social status (F—4.20, gf-2/581, p<.015), a Tukey-HSD procedure indicated that no two groups are significantly different at the .05 level in both cultures. In short, Koreans and Americans did not differ in the general level of the rated importance of effectiveness, and the two cultures were found to be similar in the perceived importance of effectiveness across three levels of status. In summary, it was found that the effect sizes of social status on the rated importance of interactive constraints were mostly negligible. Table 15-e 95 Descriptive Statistics for the Perceived Importance Ratings: Effectiveness M=5.68 SD=l.36 N=197 Requestee Social Equal Status M=5.27 SD=1.58 N=l98 M=5.43 SD=1.42 N=l98 Culture Korea U S M-5.55 M=5.80 SD-l.27 SD=1.44 N=97 N=100 M=5.14 M=5.40 SD-1.64 SD=1.51 N-lOO .N=98 M-5.44 M=5.43 SD—1.52 SD=1.33 N-98 N=100 M-5.37 M=5.55 SD-l.49 SD—l.44 N-295 N-298 96 Question 2—C (the relative weightsggiven to each constraint in predicting likelihood of use): It was predicted that peoples' ratings of the perceived likelihood of use of a given tactic would be a function of the five constraints, and that the relative weights of each constraint predicting the likelihood of use will be culture- specific. Initially, a series of 12 regressions (one for each tactic) was conducted separately for each culture. In each analysis, perceived likelihood of use was the criterion variable. The five constraints served as predictor variables in each analysis. The regression analysis revealed that, when all the variables are taken together, the multiple regression coefficients range from .33 to .55 for Koreans (Mean R=.48) and from .32 to .61 for Americans (Mean R-.48). When averaged across 12 tactics, "effectiveness" was the most important predictor of likelihood of use in both cultures (2233-.25 for Koreans; papa-.30 for Americans). On the other hand, the beta weights for the rest of the constraints were rather negligible. This suggests that effectiveness might function as a mediating variable between the four interactive constraints and perceived likelihood of use, and that there may not be a direct link between the four constraints and perceived likelihood of use. In sum, the series of regression analyses showed that likelihood of use is heavily influenced by effectiveness. Thus, although other interactive constraints may actually make an independent contribution to the prediction of the likelihood of use, a considerable part of the predictive power of these constraints is captured by the prediction made from effectiveness. The relevance of this theoretical specification was tested in two ways: first, by running 97 another series of multiple regression analyses with effectiveness deleted as a predictor variable, and second, by testing a theoretical path model. According to the results of the new regression analyses (see Tables l6-a and l6-b) without effectiveness, when all variables are considered together, the multiple regression coefficients range from .28 to .47 for Koreans (Mean R=.46) and from .28 to .54 for Americans (Mean R-.39). These multiple regression coefficients were slightly smaller than those from the analysis with effectiveness. As predicted, for Korean subjects, neither clarity (mean papa—.03, NS) nor nonimposition (mean papa=.03, NS) seems to have any predictive power at all, while concern for the other's feelings (mean papa=.23, p<.01) and concern for avoiding dislike (mean pa£a=.22, p<.01) had equally moderate predictive power._ A similar regression analysis of the likelihood of use on the U.S. sample revealed that the standardized regression coefficient papa for clarity was the highest among the four constraints (mean pa£a=.20, p<.01), followed by nonimposition (mean papa-.17, p<.01), concern for the hearer’s feelings (mean papa-.11, p<.01), and concern for avoiding dislike (mean papa-.08, p<.01). Overall, the data suggest that for Korean subjects, the constraints of clarity and nonimposition do not play as important a role as affectively-oriented constraints (i.e., concern for feelings, concern for avoiding dislike). In contrast, clarity was the most important constraint for Americans. These findings from the regression analysis are fairly consistent with each group's perceived importance ratings as shown in Table 14, with some disparity in the 98 Table 16-a Interactive Constraints and the Likelihood of Use: Regression Analysis without Effectiveness (Korean Participants) Beta-Weights Multiple R Clarity Feeling Imposition Dislike Tactic '1 """"""" 2.5; """ iéwmiié; """" ié; """" ii; """" 2 37* 09 .ll 00 29* 3 .47* .05 .31* .04 .19* 4 .46* .ll* .36* -.02 .13* 5 .36* .07 .10 -.01 .30* 6 .28* -.01 .16* .06 .13* 7 .39* -.Ol .25* .05 .18* 8 .46* -.02 .31* .01 .20* 9 .46* -.01 .24* -.04 .27* 10 .45* -.03 .34* -.03 .16* ll 46* — 06 .35* 07 13* 12 46* 05 .04 07 41* 11;; """"""" 4;; """ 63 """" £5; """" I); """"" £5; """" Note: * - Significant at .05 level 99 Table l6-b Interactive Constraints and the Likelihood of Use: Regression Analysis without Effectiveness (U.S. Participants) Beta-Weights Multiple R Clarity Feeling Imposition Dislike Tactic '1 """"""" is}; """ Si; '''' 3:65 """" 1;; """" ié; """"" 2 .35* .28* -.15* .08 .14* 3 .44* .24* .20* .24* -.04 4 .23* .18* .09 -.05 .07 5 .28* .18* .04 .12* .09 6 .32* .16* .19* .ll .01 7 .38* .16* .15* .17* .07 8 .46* .17* .06 .30* .10 9 .54* .18* .22* .26* .ll 10 .46* .17* .26* .24* -.04 ll .44* .26* .08 .20* .44 12 42* 08 .22* 16* 14* 1;; """"""" Sélm'Tié; """" ii """" i}; """" 62; """" Note: * - Significant at .05 level 100 case of avoiding imposition. Although the Korean subjects rated avoiding imposition as more important than the U.S. subjects, the regression weight for that constraint was higher for the U.S. subjects. While there were some inconsistencies for the rest of the constraints, it is important to note that these discrepancies do not disconfirm the proposed relation between perceived importance ratings and the empirically obtained beta weights, since the correlations were not corrected for attenuation due to random error of measurement. The path analysis corrects for underestimation of coefficients. Thus, the path analysis of a theoretical model provides us with a more accurate picture of the relationship between constraints and the likelihood of use. To formally test the assumed causal structure of constraints and the likelihood of use, a path model was drawn (see Figure 1). Figure 1 shows arrows from each of the interactive constraints to effectiveness. Thus, according to this path model, four constraints are causal antecedents to effectiveness. The resulting path model now loses the direct causal impact of these four constraints on perceived likelihood of use. The correlation matrices are obtained by collapsing across six situations since magnitude of beta weights are fairly consistent across situations (see Table l7-a and l7-b). Then the correlations in each matrix were corrected for attenuation due to measurement error. The corrected correlation matrices used for input in path analysis and the reproduced correlations are presented in Table 18. The hypothesized model, with path coefficients, is presented in Figure 2. The sum of squared deviation for correlations reproduced 101 Figure 1. Theoretical Path Model: Relationships amongithe Interactive Constraints, Effectiveness, and Likelihood of Use Clarity Concern for feelings \ Effectiveness ————--‘7 Likelihood of Use Avoiding imposition Concern about disapproval Table 17-a Interactive Constraints and the Likelihood of Use Across Situations: Regression Analysis without Effectiveness (Korean Participants) Beta-Weights Multiple R Clarity Feeling Imposition Dislike Situation Repay 38* .15* 09 00 20* Borrow .51* -.06 .ll .04 .32* Time .36* -.01 .16* -.10 .10 Delay .50* -.06 .28* .23* .03 Homework .49* .00 .24* .03 .20* Day 57* - 03 31* - O4 31* .47* -.00 .20* .03 .19* Note: * - Significant at .05 level Repay: Between friends (Request to repay a loan) Borrow: Between friends (request to borrow money) Time: From professor to student (request to be on time for the class) Delay: From professor to student (request to change date of presentation) Homework: From student to professor (request to extend the homework deadline) Day: From student to professor (request to permit a day off) 103 Table l7—b Interactive Constraints and the Likelihood of Use Across Situations: Regression Analysis without Effectiveness (The U.S. Participants) Beta-Weights Multiple R Clarity Feeling Imposition Dislike Situation Repay 42* .22* 21* 15* - 12* Borrow .49* .16* .03 .23* .14* Time .38* .16* .09 -.05 .05 Delay .47* .14* .16* .22* .04 Homework .47* .07 .23* .12* .09 Day 53* 26* 02 15* 27* """""""" Lé;"""i};m"':i£;"""'ii.;""""6é"""" Note: * - Significant at .05 level Repay: Between friends (Request to repay a loan) Borr: Between friends (request to borrow money) Time: From professor to student (request to be on time for the class) Delay: From professor to student (request to change date of presentation) Homework: From student to professor (request to extend the homework deadline) Day: From student to professor (request to permit a day off) 104 Table 18 Path Analysis of the Theoretical Model (1) Average zero-order correlations across the 12 tactics Concern for Avoiding Avoiding Effective Clarity Feelings Imposition Dislike ness Clarity -- .10 -.01 .10 .52 Concern for feeling .01 -- .48 .73 .36 Avoid Imposition .12 .29 -- .50 .10 Avoid Dislike .01 .58 .26 -- .25 Effectiveness .ll .41 .10 .44 ~- Likelihood of Use .03 .39 .16 '.44 .40 Likelihood of use .28 .28 * United States sample (N-299) above diagonal, Korean below (N-296) (2) Reproduced correlations Concern for Avoiding Avoiding Effective Clarity Feelings Imposition Dislike ness Clarity -- .10 -.01 .10 .52 Concern for feeling .01 -- .48 .73 .36 Avoid Imposition .12 .29 -- .50 .10 Avoid Dislike .01 .58 .26 -- .25 Effectivenss .11 .41 .10 .44 -- Likelihood of Use .04 .17 .04 .18 .40 Likelihood of use .04 .11 105 (3) Errors (actual-reproduced) Concern for Avoiding Avoiding Effective Likelihood Clarity Feelings Imposition Dislike ness of use Clarity -- .OO* .00* .00* .OO* .02 Concern for feeling .OO* -- .00* .00* .00* .13 Avoid Imposition .00* .00* -— .00* .OO* .24 Avoid Dislike .OO* .00* .00* -- .00* .17 Effectivenss .00* .00* .OO* .00* -- .00* Likelihood of Use -.01 .22 .12 .26 .OO* -- * Errors constrained to be 0 by the estimation process The sum of squared errors in the lower triangle is .125, the overall Chi-Square is 16.50 (df=4) The sum of squared errors in the upper triangle is .091, The overall Chi-Square is 10.62 (df—4) 106 Figure 2. Theoretical Path Model with Path Coefficient: Relationships among the Interactive Constraints, Effectiveness, and Likelihood of Use (Korean Sample) Clarity >\\ . ll Concern for \\ feelings .26 \\\§; .40 Effectiveness ______—9 Likelihood of Use Avoiding 7' 07/ imposition .30 / Concern about / disapproval (U.S. sample) Clarity .48 Concern for feelings -36 “\\\\\‘TTf> Effectiveness _.;£§__—fi> Likelihood -.05 of Use Avoiding _‘ / imposition -.O Concern about disapproval 107 from this model is .106 for the U.S. sample and .130 for Korean sample. The overall Chi-square goodness of fit test yielded a relatively small chi-square value for both groups of samples (12.48, af-4, NS for the U.S. sample; 16.50, af=4, p<.01, for Korean sample). Both of these chi-square values are low enough to indicate a good fit of the model to the data. To summarize, the original model more or less fits the actual data, even though none of the coefficients were perfectly reproduced for the data. As expected, for the Korean sample, the two statistically significant path coefficients were found for the link between consideration for the other's feelings and effectiveness (r-.26, p<.05), and concern for devaluation and effectiveness (r=.30, p<.05). On the other hand, the path coefficients for clarity was substantially smaller (r-.ll, NS) and minimizing imposition was counter productive (r--.07, NS). For the U.S. sample, clarity was an extremely strong predictor of effectiveness (r-.48, p<.05), followed by concern for feeling (r-.36, p<.05). Concern for avoiding dislike and concern for minimizing imposition were not significantly related to effectiveness (r--.03, r--.05, respectively). In both cultures, effectiveness was significantly related to the likelihood of use (r-.40, for Korean sample, r-.43 for the U.S. sample). In all, this model provides important theoretical insights into the role of four interactive constraints in the prediction of perceived effectiveness and likelihood of use of request tactics. First, effectiveness does not seem to function as an interactive constraint at the same level as the other four concerns. Rather, this model suggests that there is a causal impact of the four 108 interactive constraints on the effectiveness of a tactic, which in turn determine the perceived likelihood of use of a tactic. Second, effectiveness of goal attainment is a mediating judgment for both cultures, but the specific constraints which contribute to perceived effectiveness varies substantially across cultures. In sum, the results altogether indicate that cultures seem to differ in their perceptions of the relative importance of each constraint affecting the level of likelihood of use. Furthermore, these path coefficients are found to be more or less consistent with the perceived importance ratings of each interactive constraint. These findings seem to be in line with the general ethos of the face 'consideration in Korean society, and the discussion section will try to provide a clue to the social factors behind this motivation. Question 3-A: The Effects of Individual-Difference Variables on Constraint Importance. Cross—cultural comparisons for the ratings of individual difference variables are presented in Table 19. Overall, Koreans are found to be more concerned with the need for social approval (Eé6.54, E -.26, p<.0001) while the U.S. subjects showed higher need for dominance (£--15.12, 3-53, p<.0001) and self-esteem (ET'4-63: £e19, p<.0001). Specifically, the two groups of subjects showed a marked difference in self-esteem (r-.53) rather than need for approval (r-.26) or need for dominance (r-.19). However, there was a cross-cultural agreement on the relationship among the three personality variables: need for approval being negatively related to self-esteem (r--.l7 for Koreans, r--.30 for Americans) and need for dominance (r--.13 for Koreans, r--.22 for Americans). In addition, self-esteem was positively correlated with need for dominance in both 109 Table 19 Ratings of Individual Difference Variables Across Cultures (l=low, 7=high) N Mean SD t-value r Korea 294 4.35 1.03 Need for ------------------------------ 6.54 (p<.0001) .26 approval U.S. 294 3.79 1.05 Self-esteem Korea 295 3.70 1.04 ------------------------------ -15.12 (p<.0001) .53 U S 294 5.15 1 29 Need for Korea 294 4.34 1.29 Domianance ------------------------------ —4.63 (p<.0001) .19 110 cultures (r=.34 for Koreans, r=.25 for Americans) (see Table 20). Table 21 presents a zero-order correlation analysis for three personality variables (need for approval, self-esteem, need for dominance) and the perceived importance of each interactive constraint. For the Korean sample, individual differences in need for approval were significantly associated with the perceived importance of "concern for disapproval" (r=.18, p<.01) and "concern for avoiding hurting the other’s feelings" (r=.12, p<.05). The findings also showed that self-esteem and need for dominance do not significantly correlate with any of the importance ratings for Korean subjects. For the U.S. subjects, only one individual-difference variable, need for dominance, was significantly related to the importance of clarity and effectiveness (r-.l6 and r=.l3, respectively). Neither need for approval nor self-esteem had a significant relationship with the importance ratings of any constraints. A regression analysis of perceived importance of constraints on personality variables yielded a similar trend (see Table 22). For the Korean sample, need for approval yielded a significant regression coefficient papa of .13 (£-.2.26, gf-288, p<.025) for predicting concern for feelings, and .19 (p—3.17, gf—288, p<.002) for predicting avoidance of dislike. The results of regression analysis for Korean subjects also indicated that need for dominance could predict the perceived importance of effectiveness, (papa-.17, E—2.89, gf-288, p<.004) and self-esteem predicted effectiveness (papa--.12, E-—l.20, af-288, p<.045) to a lesser degree. As for the U.S. sample, "need for dominance" significantly Table 20 111 Relationships among Individual Difference Variables (Korean Participants) Need for Approval Self-esteem Need for Dominance -----'---------——---------——-—----—---------—---- Need for Approval Self-esteem Need for Dominance (The U.S. Participants) Need for Approval Self-esteem Need for Dominance Need for Approval Self-esteem Need for Dominance 112 Table 21. Correlation Analysis for Perceived Importance of Constraints and Individual Difference Variables (Korean Participants) Concern for Avoiding Avoiding Clarity Feelings Imposition Dislike Effectiveness Need for Approval -.11 .12* .11 .18** .08 Self-esteem -.02 -.03 -.06 -.04 -.07 Dominance .05 .02 -.06 .00 .13* Gender .01 .10 -.08 .07 -.18 Note: * - Significant at .05 level ** - Significant at .01 level (The U.S. Participants) Concern for Avoiding Avoiding Clarity Feelings Imposition Dislike Effectiveness Need for Approval .02 .06 .09 .11 -.01 Self-esteem .00 -.01 -.02 -.10 -.07 Dominance .16** .01 .01 .07 .13* Gender -.05 -.04 -.10 .00 -.05 Note: * - Significant at .05 level ** - Significant at .01 level Table 22 113 Regression Analysis for Perceived Importance of Constraints and Individual Difference Variables (All Participants) Beta-Weights Concern for Avoiding Feelings Imposition Avoiding Dislike Effectiveness Need for Approval Self-esteem Dominance Multiple-R (Korean Participants) .11* .13* - O6 -.l6* 03 -.00 14* 23* Beta-Weights Concern for Avoiding Feelings Imposition .15* .03 - 06 -.08 08 .17* 18* 16* Avoiding Dislike Effectiveness Need for Approval Self-esteem Dominance Multiple-R Note: * - Significant at .13* .11 -.04 - 04 05 -.03 .14 13 .05 level .19* .09 - 04 -.12* O4 .17* 19* .20* 114 Table 22 (continued) (The U.S. Participants) Concern for Avoiding Avoiding Clarity Feelings Imposition Dislike Effectiveness Need for Approval .05 .10 .10 .ll -.01 Self-esteem .06 .Ol —.00 -.09 —.ll Dominance .18* .02 .03 .ll .15* 'fiéiéiéi;li{"Ti}?""""f6%"""miié'"""Ti%;"""'fig; """" Note: * = Significant at .05 level 115 affected the importance of clarity and effectiveness, its standardized regression coefficient papa being .18 (E=2.95, g§=290, p<.004) for clarity and .15 (E=2.51, af=289, p<.013) for effectiveness. On the other hand, none of the other personality variables (i.e., self-esteem and need for approval) had a significant effect on the dependent variables. In sum, among the three personality variables, only need for dominance had cross-culturally significant influence on the perceived importance of effectiveness in achieving request goals, although the effect size was relatively small. Question 3-B: The Effects of Culture in predicting Constraint Importance after Controlling for Individual Difference Variables. Given that there exist some significant relationships among individual-difference variables and constraint ratings, it is interesting to see whether culture still makes a difference after controlling for the individual-difference variables. Thus, hierarchical multiple regression analyses were conducted, with personality variables simultaneously entered as step 1, gender as step 2, and culture as step 3. Summary results are shown in Table 23. Step 1 shows the statistics for the equation with all three personality variables entered simultaneously. Step 2 shows the same statistics when another variable, gender, is added. Finally, step 3 shows the statistics when culture is added to the equation. As for the perceived importance of clarity, examination of Table 23 shows that introducing culture into the equation has significant and substantial effects after the individual-difference variables are controlled (R-squared change-.19, 5-156,369, p<.000). Regarding the 116 Table 23 Hierarchical Multiple Regression Analysis for Perceived Importance of Constraints and Individual Difference Variables (Clarity) Multiple R2 F Step Variable R R2 change Change Signif Approval Dominance 1 Self-esteem .30 .09 .09 18.50 p<.000 2 Gender .31 .09 .01 3.73 p<.054 3 Culture .54 .29 .19 156.37 p<.000 (Concern for the hearer's feelings) Multiple R2 F Step Variable R R2 change Change Signif Approval Dominance 1 Self-esteem .14 .02 .02 3.63 p<.013 2 Gender .14 .02 .00 1.02 p<.314 3 Culture .16 .03 .01 3.97 p<.047 (Nonimposition) Multiple R2 F Step Variable R R2 change Change Signif Approval Dominance 1 Self-esteem .23 .05 .05 10.83 p<.000 2 Gender .23 .05 .00 .52 p<.472 3 Culture .33 .ll .05 33.94 p<.000 (Avoidance of devaluation) Multiple R2 F Step Variable R R2 change Change Signif Approval Dominance 1 Self-esteem .18 .03 .03 6.16 p<.000 2 Gender .18 .03 .00 .863 p<.353 3 Culture .18 .03 .00 .364 p<.546 117 Table 23 (Continued) (Effectiveness) Multiple R2 F Step Variable R R2 change" Change Signif Approval Dominance l Self-esteem .16 .03 .03 5.00 p<.002 2 Gender .21 .05 .02 11.83 p<.001 3 Culture .22 .05 .005 2.79 p<.095 118 importance of non-imposition, the R squared change in the final step (Step 3) due to the introduction of culture also was statistically significant (R-squared change=.05, F=33.94, p<000). On the other hand, the change in the size of R-squared in predicting "Concern for feelings" was negligible (.01), even though it was statistically significant (F=3.966, p<.047). Consistent with the earlier analyses, culture did not significantly contribute to the increase in the size of R-squared for either "concern for dislike" or "effectiveness." In sum, culture continued to predict the importance of clarity, non- imposition, and consideration for the other's feelings after intra— cultural variability had been controlled. CHAPTER IV DISCUSSION AND CONCLUSION Summary of findings The primary aim of this dissertation is to explore cross- cultural variations in peoples' perceptions regarding requesting behavior from an implicit psychological perspective. It compares whether the five interactive constraints are perceived similarly across cultures, and traces the possible links between these perceptions of interactive constraints and likelihood of use. This dissertation postulates the following interactive constraints that may serve as criteria for making a choice between tactics in the pursuit of interaction goals: (1) concern for clarity, (2) concern for nonimposition, (3) concern for avoiding hurting the hearer's feelings, (4) concern for avoiding negative evaluation by others, and (5) concern for effectiveness. Each constraint represents a particular type of interactional concern. Based on the five interactive constraint, the specific research questions were formulated regarding the potential cross-cultural variability of the perceptions (implicit theories) of requesting behavior. The analyses with 595 subjects (296 Korean and 299 American university students) uncovered a multitude of cross-cultural similarities and differences in particular perceptions of requesting behavior. With regard to research question l-a, we expected request tactics in different cultural groups we studied to display three major clusterings. Three categories were: (1) Hint category, tactics that realize the request either by partial reference to the object or element needed for the implementation of the act or by reliance on 119 120 contextual cues; (2) Query2‘tactics‘that realize the act by reference to contextual preconditions necessary for its performance; and (3) Direct Statement: tactics that are syntactically marked as such, or by other verbal means that name the act as a request. Overall, the results suggest that tactics in the three major strategy categories (with minor changes) form internally consistent scales. The finding that some of the alphas are quite low, however, raises questions about whether members of either culture view the three major strategies as coherent categories. As was expected (research question l-b), the results of the analysis show cross-cultural similarity in the relative ordering of tactics along interactive constraints, with some disparity shown in the case of effectiveness judgments. The overall ranks of the tactics along the scales of four constraints follow a similar pattern across cultures, although two cultures seem to rate the specific absolute levels of interactive constraints differently. Striking cross-cultural differences in the rank and mean strategy ratings were found in effectiveness judgments: U.S. subjects consider the direct statement strategy as the most effective way of making a request, while Korean subjects rated it as the least effective strategy. Regarding the incompatibility among interactive constraints (research question l-c), there was a high degree of incompatibility between the clarity of strategies and the three other constraints (minimizing imposition, avoiding hurting the hearer's feelings, and avoiding negative evaluation by the bearer). This trend was most striking for the Korean subjects. Marked cross-cultural differences were found in the relationship between clarity and effectiveness. Specifically, American subjects see clarity to be closely related to 121 effectiveness of tactics; for Korean subjects clarity of tactics was counterproductive to effectiveness. In addition, Korean subjects perceived "prosocial" constraints (concern for the other's feelings and autonomy, and for the speaker's own image) to be highly effective. On the other hand, for the American subjects, these constraints were unrelated to effectiveness. The findings also point to interesting cross-cultural differences in the perceived importance of interactive constraints (research question 2—a). People who belong to different cultural groups seem to share basic interactive constraints, but the prominence of each constraint differs. The results testify to a lack of concern for clarity on the part of Koreans, in this case superseded by face-maintaining considerations -- concern for feelings, and avoiding Imposition., On the other hand, Americans granted Clarity a much higher value than did the Koreans, thus allowing directness concern ('plain talk') to play a more central role in interaction. In research question 2-b, we investigated the moderating effect of the requestor's social status relative to the requestee in the perceived importance of each constraint. Some authors (Brown & Levinson, 1978; Cherry, 1988; Donohue & Diez, 1985; Garvey, 1984; Scarcella & Brunak, 1981) point out that request forms reflect the perceived or strategically implicated dominant—submissive status relationship. In order to investigate the effects of social status on the ratings of perceived importance, the data were analyzed by means of a 2 x 3 ANOVA. The effect sizes of social status on the rated importance of interactive constraints were mostly negligible, 122 even though some of them are statistically significant. In sum, the effects of culture were not moderated by status. There were cultural differences in relative weights given to each constraint in predicting likelihood of use ratings (research question 2-c). A series of regression analyses of data performed with the five constraints as independent variables and the likelihood of use as dependent variable suggest that effectiveness may function as a mediating variable between the four interactive constraints and the likelihood of use, and there may not be a direct link between the four constraints and the likelihood of use. This theoretical prediction was tested using path analysis. As predicted, the path analysis indicated that there is a causal impact of the four constraints on the effectiveness of goal attainment, which in turn serves as a mediating judgment for both cultures that leads to perceived likelihood of use. In general, American subjects yielded clarity a much higher weight than did the Korean subjects. Similar to their perceptual importance of constraints, Koreans seem to allow face maintaining considerations to play a more central role in predicting effectiveness and likelihood of use. Overall, the relative regression weights were found to be consistent with each group's importance ratings. The present study also found that the effect sizes of individual difference variables on the perceived importance ratings were small, though some were statistically significant (question 3-a). Finally, question 3-b asked whether culture still would affect the perceived importance of each constraint after controlling for the effects of individual difference variables? In sum, culture 123 continued to predict the importance of clarity, non-imposition, and consideration for the other's feelings after intra-cultural variability had been controlled. The general pattern of results that emerged clearly points to distinct effects of cultural beliefs on requesting behavior. One of the most striking aspects of cross-cultural differences was the role of ”clarity" constraints. "Clarity concern" takes on the highest importance for the American subjects; for Korean subjects, it was a relatively less important concern. This trend was found not only in each group's conscious ratings of perceived importance but also in the prediction of the effectiveness and likelihood of tactic use F(i.e., the path model). On the other hand, "disapproval concern" seems to take on more importance for the Korean subjects than for the American subjects, especially in the prediction of the effectiveness of tactics. The results of this investigation also provide a number of interesting and important cross-cultural similarities in requesting beliefs. First, the results of this study suggest that a certain level of "concern for the other's feelings" appears to be essential for effective interaction, regardless of cultural context. Second, according to the path model tested, effectiveness is a more immediate determinant of one's perceived likelihood of tactic use than the rest of the interactive constraints. While there were significant cultural differences regarding which tactics were perceived as effective, effectiveness played an equally important role in the prediction of likelihood of use in both cultures. One might perceive that there seem to exist some inconsistencies 124 between the rank order correlations among constraints, and the results of path analysis. For instance, there was a highly negative rank-order correlation between clarity and effectiveness for the Korean subjects, while the path coefficient for the effect of clarity was positive (r=.ll). It should be noted that the path coefficients are calculated based on the correlations among constraint judgments at the level of individual ratings. On the other hand, the rank- order correlations (see Table l3—a and l3-b) regarding the relationships among constraints and the likelihood of use are based on mean responses (N-12 tactics). Mean ratings of tactics along the constraint dimensions predict the relationships among constraints on an average. They do not reflect ratings at an individual level. Thus, it is not particularly surprising that there appear to be some inconsistencies between the results of path analysis and the rank- order correlations Theoretical Implications: Cross-cultural Differences in the Implicit Theories of Requesting Relationship between interactive constraints and perceived likelihood of use: To what extent does the present cross-cultural comparisons reveal similarities and differences between cultures in their beliefs on the "ways of speaking" (cf. Hymes, 1972, 1974)? Or can we detect culturally specific interactional styles in the requestive beliefs? First of all, the results provide important theoretical insights into the relationship among the five interactive constraints, and perceived likelihood of use. Earlier, it was speculated that effectiveness may operate on a somewhat different level than other constraints. This was because effectiveness 125 judgments depend more on contextual factors than the rest of the interactive constraints. According to the results, across cultures effectiveness does not function as an interactive constraint at the same level as the other four constraints. There are several possible explanations for this. First, effectiveness does not adhere to particular communication strategies or tactics; rather, effectiveness is a judgement of the use of particular tactics in specific settings for specific purposes. In this way, effectiveness judgments are highly situation specific, in the sense that judgments are made of context-embedded factors rather than of language act types. Second, communication effectiveness sometimes is equated with successful goal achievement or task accomplishment (Spitzberg & Cupach, 1989) or peoples' ability to achieve their primary goals (Kellermann, 1989). Thus, the effectiveness judgment of a tactic may be the immediate antecedent for choosing to use that particular tactic. Understanding how language functions is one of the main goals in communication research. "Why do people say what they say?" or "Why do speakers bother to choose one strategy over another?" Some linguists and communication scholars have wondered why speakers bother to choose one request form over another. This perspective has involved trying to understand people's tactics for accomplishing their goals within stated interactive constraints. In the past, several authors have suggested cross-situational constraints that are "motivating forces" in communication: "Be clear" and "Be polite" (Lakoff, 1977); "Concern for clarity" and "Concern with support" (Greene & Lindsey, 1989); and "Directness" and "Politeness" (Blum- Kulka, 1987). Grice (1975) also put forward the "Maxim of manner" in 126 the use of language (e.g., be clear, be brief, try to avoid obscurity), which can be seen as guidelines for direct communication. Brown & Levinson (1978) also posit such wants as: (l) the want to be efficient or indicate urgency; and (2) the want to maintain hearer's face to some degree. Leech (1983) listed rhetorical principles constraining communicative behavior: the Politeness Principle, the Cooperative Principle, Clarity, and the Economy Principle. Most recently, Kellermann (1988, in press; Kellermann & Kim, 1991) has suggested "efficiency of goal attainment" and "social appropriateness of behavior" as two cross-situational constraints (or metagoals) to account for strategies used to pursue various interaction goals. The current list of interactive constraints includes the major principles that have been proposed as being important in guiding the use and understanding of conversational strategies. More importantly, the findings reported here indicate that the two simple constraint categories suggested by many of prior researchers seem to be too general. The findings of this study point to the complex relationships among effectiveness and four different interactive constraints. Specifically, the correlations among the four different constraints were quite dissimilar within each culture as well as across cultures. These data suggest that the three face-related constraints (i.e., avoiding devaluation, avoiding hurting the other's feelings, and avoiding imposition) cannot be subsumed under a single construct, i.e.,"social appropriateness". The notion of social appropriateness is too general to help us in understanding the mechanism behind the interactive constraints. 127 While the overall approach taken by prior research provides a useful framework for investigating the role of the major constraints in the generation and production of conversational behavior, most authors make no predictions about the cross-cultural generalizability of the constraints. For instance, although Kellermann (1988) posits that metagoals are fundamental, cross-situational conversational constraints, she does not specify whether they are universal constraints that affect the selection of strategies. This dissertation provides a theoretical framework that can systematically explain how the interactive constraints guide the preference of communication strategies across cultures. The current model of implicit theories of conversational behavior is designed to help us understand psychological determinants of strategy choices. Salience of interactive constraints: The results of this study support general, intuitive judgments concerning the relative directness of American style in comparison with that of Korean style. The results of this study seem to suggest different processes of arriving at goals: Americans focus on task constraints (conveying the message clearly and efficiently) and Koreans focus on social— relation constraints (avoiding damage to the relationship or loss of face by the hearer). The overall picture that emerges from a series of findings advocates the enduring cultural generalizations on the differences in linguistic perceptions: the American standards or norms of "truth-telling" (Condon, 1985), "straight talk" (Stewart, 1972), "clarity" (Eisenberg, 1984), "linear" pattern (Ishii, 1981), "outcome orientation" and "direct communication emphasis" (Yum, 1988), versus the East Asian norms of "positive face concern" (Ting- 128 Toomey, 1988), "receiver centered" and "indirect communication emphasis" (Yum, 1988). Understanding peoples' perceptions of strategy use requires an understanding of the social context, social mechanisms, and human motivation. For instance, the cross-cultural differences in the salience of interactive constraints are consistent with the predominant forms of communication in individualism-collectivism or low-high-context cultures (see Hall & Hall, 1985; Gudykunst, Ting— Toomey with Chua, 1988). Individualism-collectivism and low-versus high-context communication are broad dimensions of cultural variability that influence many difference aspects of interpersonal communication. The cultural differences in request beliefs that emerged are generally consistent with different aspects of cultural dimensions. Specifically, the culture-specific differences in implicit theories of requesting behavior is in line with the theoretical predictions made by Kim (in press a). She argued that given the similar situational contingencies, the propriety of one’s tactics will become a more important consideration in judging one's strategic competence as the culture moves towards a more collectivist orientations. Levine (1985) describes communication in the United States (an individualistic and low-context culture) in this way: The [North] American way of life, by contrast, affords little room for the cultivation of ambiguity. The dominant [North] American temper calls for clear and direct communication. It expresses itself in such common injunctions as "Say what you mean", "Don't beat around the bush," and "Get to the point." (p.28). The contrast between cultural emphasis on different conversational norms about other-face-maintaining considerations, as against clarity, has been recognized by various authors. One of 129 Grice's maxims for cooperative conversation is "manner," which suggests that speakers should avoid ambiguity and obscurity of expression and ambiguity (Grice, 1975). Several writers have written that while this direct communication is a norm in North America, an individualistic society, Grice's principle would not be accepted as a norm in cultures with a different value orientation (see Yum, 1988). Okabe (1987), for instance, has shown that in Japan, the traditional rule of communication which prescribes against demanding, rejecting, asserting oneself, or criticizing the listener straightforwardly, is a more dominant principle than Grice's maxim of manner. Similarly, Park (1979) argued that Koreans do not make "negative responses" like 'no' or 'I disagree with you,’ or 'I cannot do it.’ Rather they like to use more frequently than North Americans.circumlocutionary expressions, such as ... 'I agree with you in principle' .. or 'I sympathize with you' (p.88). Park (1979) also mentions that the importance of preserving group harmony and the importance of nunchi (an affective sense by which Koreans can detect whether others really are pleased or satisfied or not) as the two primary reasons why most Koreans opt for the indirect style of communication in their everyday lives. The preference for indirect, ambiguous communication over direct, open communication is attributed to the importance of the face-honoring process whereby the concept of 31923 (respect for the other's sense of selfhood that includes their morale, face, self-esteem, and state of mind) has a highly valued place in the Korean culture. Gudykunst, Ting-Toomey with Chua (1988, p.102) write: The value orientation of individualism propels North Americans to speak their minds freely through direct verbal expressions. 130 Individualistic values foster the norms of honesty and openness. Honesty and openness are achieved through the use of precise, straightforward language behaviors. The value orientation of collectivism, in contrast, constrains members of cultures such as China, Japan, and Korea from speaking boldly through explicit verbal communication style. Collectivistic cultures like China, Japan, and Korean emphasize the importance of group harmony and group conformity. Group harmony and conformity are accomplished through the use of imprecise, ambiguous verbal communication behaviors. Interactive constraints are essentially a generator of tactical preferences. If there were no overarching themes, people's choices of tactics would appear as isolated entities without connection to the rest of the knowledge. The current model has the potential to account for both cross-cultural similarities and differences in the manifestation of the role of interactive constraints across a wide variety of primary goals. Perceptions regarding the role of the interactive constraints across cultures may affect the organization of interaction, which are good candidates for being potentially important determinants of "cultural ways of speaking" (cf. Katriel, 1985; Hymes, 1974). Determining influence of interactive constraints on conversational behavior is an important and necessary endeavor, particularly since, in the "global village," intensive communication with speakers of differing backgrounds is becoming the rule rather than the exception. The interactive constraints can potentially account for fundamental structural distinctions underlying contrasts in expressive patterns. Practical Implications In our modern socially diversified and specialized urban societies, intensive communication with speakers of differing ' backgrounds and assumptions is the rule rather than the exception. Conversing and communicating competently across cultures is becoming 131 a major concern for many people. The current line of research has important practical implications for intercultural interaction and the cross-cultural communication competence. Knowing the cultural "way" of communication entails more than a command of syntax and lexicon-~it requires an understanding cultural perceptions and of the usages regarding different types of communication tactics. Dependence of the implicit requesting theories on culture (found in this dissertation) show that the transfer of the norms of one community to another community may well lead to "pragmatic failure" (Leech, 1983), and to the judgment that the speaker is in some way being impolite, uncooperative, etc. Tannen (1981), for example, has remarked on the pragmatic failure caused by the fact that indirectness in a second language (Creek) was different from that of her native-speaker hosts. While chatting, she mentioned that she associated Greece with grapes, and was surprised at not having seen any since she came to Greece. She also mentioned that Americans have a special way of fixing eggs, i.e., scrambling them. These utterances, meant as polite conversational topics, were taken as hints by her hosts, and she was faced with scrambled eggs and grapes for breakfast from then on. Unfortunately, neither was one of her favorite dishes. The current findings indicate that the similar intercultural misunderstandings can occur between Koreans and Americans. The cross-culturally differing estimates of requesting styles may lead to differing ways in which speakers choose to form their requests and the ways in which hearers are affected by these choices. The successful intercultural interaction may partly be achieved if interactants with different implicit theories on 132 requesting styles can correctly interpret the observed requesting behavior. The competent use of language lies in knowing how to use words to get listeners to make the right inferences about what is meant (Gibbs, 1985). Pragmatically, this research suggests that simply knowing the meanings of individual words is not sufficient to ensure the proper accomplishments of conversational goals. One of the most striking aspects of this research was the cross-cultural differences in the salience of interactive constraints. Likewise, people need to have additional information about the value ("importance") attached by one another to interactive constraints. The results indicate that this pragmatic information constitutes the shared or mutual knowledge that allows interlocutors to achieve communication. Failure to produce successful strategies might be seen as reflecting either a general knowledge deficit, or communicative incompetence. This factor probably accounts for much stereotypical cross-cultural misunderstanding. Highly relevant in this context is the work of Gumperz (1982), which shows that cross—cultural differences in expectations of linguistic behavior, can lead to breakdowns in intercultural communication. Examining interactions between British-English and Indian-English speakers in England, Gumperz (1978) found that differences in cues resulted in systematic miscommunication about whether a question was being asked or whether a person was being rude or polite, among others. The current investigation points out that when speakers of different cultural backgrounds interact, the problems that develop in communication can be accounted by 133 misunderstanding of the salience of each others' global goals or constraints in conversation. In sum, one of the most important implications of this study is that different cultures do not necessarily share the same assumptions concerning what constitutes appropriate choices of strategies. Blum- Kulka (1987) explains some of the learner's speech act behavior as resulting from overgeneralization, simplification, or reduction of sociopragmatic knowledge. To the extent that strategies are linked to interactants' assessments of constraints and to the extent that cultures differ in their assessment of these values, misunderstandings can occur. Given that global constraints contribute to consistent performances across varying contexts, and that they can influence the establishment of stable and/or preferred interaction exchanges, the idea of the different restraining force of various interactive constraints can provide a useful framework for explaining intercultural misunderstandings. Are cultural stereotypes of the "direct," or even "blunt" American or the "roundabout" Japanese anchored in people's strategic knowledge on the interactive constraints? The general conclusion is that successful accomplishment of interpersonal goals require general regulatory processes that are domain-independent. Then it is necessary to include instruction regarding regulatory interactive constraints, domain-general metacognitive knowledge on other cultures. To get things done, we must communicate intensively with individuals whose backgrounds are different from our own. Their differences in language use and communicative style often look like and may be interpreted as 134 deficits in competence or intellectual ability. Limitations and Directions for Future Research This research has several limitations. First, the current empirical evidence focuses on peoples’ perceptions of interactive constraints, rather than on their actual strategy use. The issue of performance is not a minor one, though external to the perspective studied here. While the Korean and the U.S. subjects seem to be equally concerned about the effectiveness of a tactic, the idea of what constitutes an "effective" tactic seem to differ across the two cultures. The "actual" effectiveness (outcome) of different utterances in achieving primary goals has not been studied in this dissertation. The results also do not provide an answer as to whether two cultural groups are able to employ those means that are perceived to be "effective". Lack of volitional control over the choice of strategies (e.g., verbal fluency, communicative anxiety, lack of strategy repertory, etc.) can prevent one from acting in accordance with the perceptions of the importance of constraints. Future studies should be conducted to test how the knowledge of the interactive constraints guides communication performance and outcome. The studies that deal with subsequent remedial moves (e.g., dealing with resistence, accounts, and apologies) should also be conducted since this diseertation deals only with initial requests. Second, the current research is limited to the particular communication phenomenon of "requesting". While requests are particularly interesting as they constitute face-threatening acts and can stretch over large parts of conversations, future studies should attempt to empirically document conversational behavior in other 135 interaction goals (e.g., apologies, refusals, criticisms) and compare them across different speech communities for deeper understanding of the phenomena involved. Third, it should be noted that specific cross-cultural and cross- linguistic studies including the current one tend to focus on two or three cultural and linguistic systems at most. The studies contrast the interaction styles of the following: Greeks and Americans (Tannen, 1981), Spanish and English (Walters, 1979; Rintell, 1981; Fraser & Nolen, 1981); Tamil, Tzetal, and English (Brown & Levinson, 1978); Korean and English (Holtgraves & Yang, 1990); German, Danish, and British English (House & Kasper, 1981); and Japanese and English (Hill, Ide, Ikuta, Kawasaki & Ogino, 1986). Furthermore, in many cases, English serves as the yardstick for comparison. Thus, wider variety of languages and different populations with a variety of linguistic and cultural backgrounds should be investigated. Fourth, to date relatively few studies have been conducted on nonnative speech act performance. Clearly, there is a definite need for studies examining second-language learner populations. Requests, like some other primary goals, reflect cultural values. Deeply held cultural values are not easily given up. Studies should be conducted regarding the pragmatic transfer. Conclusion The current research raises the larger issues of what kinds of general interactive constraints shape interaction beliefs, and how these constraints are anchored in the wider social structure. The results obtained so far for the primary goal of requesting support enduring cultural generalizations based on perceptions regarding 136 linguistic behavior. The five conversational constraints serve as "pressures" which shape and give rise to distinct social interaction patterns. The interactive constraints account for fundamental structural distinctions which underlie cultural contrasts in expressive patterns. The research presented in this paper extends our knowledge about what kinds of general interactive constraints shape people's beliefs on interactions, how interactive constraints are anchored in the wider cultural milieu. Recent years have seen developments in the higher-level constraints in interaction. This approach is particularly useful for the understanding of social interaction. What will the future bring to the area of interactive constraint? In the future, more careful and critical examinations of interactive constraints and related concepts will probably occur. Undoubtedly, the list of interactive constraints introduced in this paper should be further refined, and clarified. Some methodological advances, better ways to measure and assess importance of interactive constraints than is presently available, should also develop. The notion of interactive constraints is an extremely important topic, eminently worthy of further theoretical and empirical investigation. 137 NOTES 1. Request as a primary speech act has been discussed under various pseudonyms in the literature: "directive" (Andersen, 1990; Ervin-Tripp, 1976; Gordon & Ervin—Tripp, 1984; Leech, 1983; Searle, 1975), "exercitives" (Austin, 1962), "regons" (Soskin & John, 1963), "requestives" (Bach & Harnish, 1979), and "impositive" (Green, 1973). In the communications field, this class of conversational goals have been labeled as "compliance gaining" (Hunter & Boster, 1987; Marwell & Schmitt, 1967; Miller, Boster, Roloff & Seibold, 1987). While both compliance-gaining and requesting obviously focus on behavior change of the targetl requesting is only one approach to gaining compliance. Warning, ordering, ingratiation, and threatening, among others, can also function as different approaches for gaining compliance. Thus, compliance gaining is a broader term than requesting. 2. Spanish and English (Walters, 1979; Rintell, 1981; Fraser & Nolen, 1981); Hebrew, German, English, and Danish (Blum-Kulka, Danet & Gherson, 1985; Blum-kulka, 1987); Tamil, TZetal, and English (Brown & Levinson, 1978); Korean and English (Holtgraves & Yang, 1990); German, Danish, and British English (House & Kasper, 1981); and Japanese and English (Hill, Ide, Ikuta, Kawasaki & Ogino, 1986). The scheme also follows from various scales of request and directive forms: illocutionary transparency or opacity (Searle, 1975; Ervin- Tripp, 1976; House & Kasper, 1981, 1987; Blum-Kulka, 1982, 1987; Blum-Kulka et a1., 1985; Weizman, 1989), syntactic directness (Becker, Kimmel & Bevill, 1989), politeness or deference (Brown & Levinson, 1978; Fraser, 1978; Fraser, Rintell & Walters, 1980; Fraser & Nolen, 1981; Hill et al., 1986; James, 1978; Kemper & Thissen, 138 1981a; Rintell, 1981; Sanford & Roach, 1987; Winograd, 1977), imperative force or coerciveness (Jordan & Roloff, 1990; Ervin & Tripp, 1976), syntactic features (Carrell & Konneker, 1981; Walters, 1979), and cost and benefit to the interlocutor (Clark & Schunk, 1981). 3. To see if any other constraints in the selection of request strategies were overlooked, 10 Korean and 10 U.S. students at Michigan State University were interviewed regarding the reasons why they would or would not use particular request strategies. Generally speaking, the students reported back the originally identified five kinds of constraints. The responses of interviewees are summarized in Appendix A. 4. The construct and predictive validity of the "likelihood of use" procedure is a subject of controversy (Burleson, Wilson, Waltman, Goering, Ely & Whaley, 1988; Seibold, 1988; Hunter, 1988; Boster, 1988; Burleson & Wilson, 1988; Dillard, 1988). The ratings of the likelihood of using an utterance are not necessarily the same as the actual use of those utterances. Thus, this paper prefers to view the results as tapping perceptions and beliefs regarding language use. 5. The context of each request situation chosen for this study is not a standard requesting one (i.e. Asking for time, passing the salt, etc.). Hoppe-Graff, Hermann, Winterhoff-Spurk and Mangold (1985) note that in standard recurring situations, request can be highly indirect without the risk of misunderstanding. 6. The second statistical test for unidimentionality is the test for external consistency or parallelism (Hunter & Gerbing, 1982). 139 The parallelism test is to ascertain whether all the items correlate in the same way with relevant outside factors. The tests for external consistency yielded quite a number of deviations between the actual and predicted correlations. The attempt to revise the factor structures were not successful. Thus, the claims on the potential unidimensionality of each strategy categories should be treated with caution. 7. The similar analyses were conducted to test the effect of participants' gender. The findings showed that the gender effects on the ratings of tactics and the perceived importance of constraints were mostly small and insignificant (refer to Tables 21 and 23). APPENDICES 140 Appendix A Summary of Interview Results Regarding the Reasons why People Might Not Use a Certain Strategy (1) Concern for clarity I wouldn't use hints ("I have run out of money") because it may not make sense to the other person. (2) Concern for effectiveness I don't have to use the most direct strategies since some indirect ones ("Could you be on time?") can still communicate my intention. The direct and some conventional indirect requests will be equally effective. So why be so direct and risk offending the other? (3) Concern for the hearer’s feelings I don't want to embarrass others by using statements which are too abrupt or direct. I don't want to cause inconvenience to others. I want to avoid hurting the other’s feelings. (4) Concern about risking devaluation for self I would be concerned not to be perceived by the bearer as rude and inconsiderate. When the topic of a request entails potentially high relational consequences, I would use a more indirect request. (5) Concern for minimizing imposition I feel sorry for the other when I have to ask something that potentially burdens the other. So I want to be indirect in my request. Some requests can lay a psychological, or sometimes monetary, burden on the other. The reason why I prefer to use indirect strategies is that I don't want to leave the impression that I am forcing him/her to do certain things. (l) (2) (3) (4) (5) (6) (7) ' (8) (9) 141 APPENDIX B List of Request Situations and Specific Tactics Mild hint Strong hint Question hint Syntactic downgrader Permission Ability Query-Preparatory Willingness Query-Preparatory Suggestory Want (10) Performative (11) Obligation (12) Imperative 142 (Repay) Tactic 1: \OCXJVONU‘J—‘LONH 10. ll. 12. Situation: Imagine that one of your female friend, whom you have known for several years, has the habit of borrowing money and then not repaying it for long periods of time. In fact, it seems that she has been late not only when repaying money borrowed from you but also when she has received loans from other people. Two weeks ago the person borrowed 20 dollars from you. Again, the person did not repay it as promised. You waited a few days more but found that you really need some cash. Now you want to ask her to return the money. "I have run out of cash." "I could use the money I loaned you." "Do you remember the money I loaned you?" "Would it be alright if I ask you to repay the loan?" "May I ask you to repay the loan?" "Could you repay the loan?" "Will you repay the loan?" "How about repaying the loan?" "1 would like you to repay the loan." "1 must ask you to repay the loan." "You should repay the loan." "Repay the loan." Tactic 2: QVO‘U'IDWNH 9. 10. ll. 12. "I don't have any money." "I wish I could get the money I loaned you." "Do you have any money?" "I wonder if you could repay the loan?" "Can 1 ask you to repay the loan?" "Can you repay the loan?" "Won't you repay the loan?" ”You will repay the loan, won't you?" "I would appreciate it if you repay the loan." "I'm asking you to repay the loan." "You will have to repay the loan." "Give me my money." .3... .k!. .. . 11.435313. . . 143 (Borrow money) Situation: Imagine that you missed breakfast and are about to have lunch at a university cafeteria. When you search for money, you notice that you have forgotten to bring your wallet. Given your class schedule, you have just enough time to eat but not enough time to go back home and get money before your next class. Just then, you happen to notice one of your classmates (male), whom you have known for several years, sitting nearby. Now you decide to ask him to lend you some money. Tactic 1: \OGJVO‘U‘J-‘WNH 10. ll. 12. "I forgot to bring my wallet." "I wish I could borrow some money." "Do you have any money?" "Would it be alright if I ask you to lend me some money?" "May I ask you to lend me some money?" "Could you lend me some money?" "Will you lend me some money?" "How about lending me some money?" "I would like you to lend me some money." ”I must ask you to lend me some money." ”You should lend me some money." "Lend me some money." Tactic 2: QNOUIJ-‘UONH 9. 10. ll. 12. "I seem to have misplaced my wallet." ”I could use some money." ”Do you have any extra cash?" ”I wonder if you could lend me some money?" "Can I ask you to lend me some money?" "Can you lend me some money?" "Won't you lend me some money?" ”You will lend me some money, won't you?" "I would appreciate it if you lend me some money." "I'm asking you to lend me some money." "You will have to lend me some money." ”Give me some money.” 144 (Taking a day off) Situation: Imagine that you are a graduating senior working on a professor's (male) research project from whom you had taken several classes before. You are supposed to work in the professor's office every Tuesday and Thursday. Next Tuesday, however, you have an important interview with a prospective employer. Since the interview coincides with your working hours you need to take time off to attend the interview. In this situation, you want to ask the professor (male) for permission to take the time off. Tactic 1: \DCDNO‘UJ-‘WNH "I have an important interview next Tuesday." "I could use next Tuesday off." "Do you need me to work next Tuesday?" "Would it be alright if I take next Tuesday off? "May I take next Tuesday off?" "Could you let me take next Tuesday off?" "Will you let me take next Tuesday off?" "How about letting me take next Tuesday off?" "I would like you to let me take next Tuesday off." 10. "I must ask you to let me take next Tuesday off." 11. "You should let me take next Tuesday off." 12. "Give me next Tuesday off." Tactic 2: 1. "I haven't missed a day of work yet." 2. "It will be difficult for me to work next Tuesday." 3. "Do you have anything important for me to do next Tuesday?" 4. "I wonder if you could let me take next Tuesday off?" 5. "Can I take next Tuesday off?" 6. "Can you'let me take next Tuesday off? 7. "Won't you let me take next Tuesday off?" 8. ”You will let me take next Tuesday off, won't you?" 9. "I would appreciate it if you let me take next Tuesday off." 10. ”I'm asking you to let me take next Tuesday off." 11. "You will have to let me take next Tuesday off." 12. "Let me take next Tuesday off." fatalities-r... ......i:i...:.. .-.. i it... .. 145 (Being on time) Situation: Imagine that you are a professor in a university. In your class, group activities and participation is weighted heavily. From the start of the semester, one student (male) is continually late. He seldom makes it to class on time. Other students in the class appear to be disturbed by the student coming into class late. After class, you want to ask him to come on time for future class sessions. Tactic 1: l. "The class seemed to be a bit disturbed today." 2. "You were 30 minutes late for class today." 3. "Do you know what time the class begins? 4. "Would it be alright if I ask you to be on time for the next class?" 5. "May I ask you to be on time for the next class?" 6. "Could you be on time for the next class?" 7. "Will you be on time for the next class?" 8. "How about being on time for the next class?" 9. "I would like you to be on time for the next class." 10. "I must ask you to be on time for the next class." 11. "You should be on time for the next class." 12. "Be on time for the next class." Tactic 2: 1. "Arriving late disturbs the class." 2. "The class began 30 minutes before you arrived." 3. "Did you forget the time of our class?" 4. "I wonder if you could be on time for the next class?“ 5. "Can I ask you to be on time for the next class?" 6. "Can you be on time for the next class?" 7. "Won't you be on time for the next class?" 8. "You will be on time for the next class, won't you?" 9. "I would appreciate it if you would on time for the next class." 10 "I’m asking you to be on time for the next class." 11 "You will have to be on time for the next class." 12 "Don't be late for the next class." 146 (Delay a presentation) Situation: Imagine that you are a professor. For your class, you require individual presentations on class material. The presentation counts for 40% of the final grade and it involves demonstrating some experiments. Today is the first day of presentations but due to a backlog of material, you find it necessary to lecture for part of the time to cover material for the upcoming exam. Therefore, the last presenter (female) who had to bring various devices and electronic equipment, will not be able to present today. Now as the professor you want to ask her to delay her presentation. Tactic 1: 1. "Our upcoming test requires me to lecture for part of the class today." 2. "We won't be able to have everyone present today." 3. "Would you like to have more time to prepare your presentation until our next class?" "Would it be alright if I ask you to delay your presentation." "May I ask you to delay your presentation?" "Could you delay your presentation?" "Will you delay your presentation?" "How about delaying your presentation?" . "I would like you to delay your presentation." 10. "I must ask you to delay your presentation." 11. "You should delay your presentation." 12. "Do your presentation in the next class." \oooxloxuwb Tactic 2: "I have to lecture for part of the class today." "There is not enough time left today for everyone to present." "Is it important for you present today?" "I wonder if you could delay your presentation?" "Can I ask you to delay your presentation?" "Can you delay your presentation?" "Won't you delay your presentation?" "You will delay your presentation, won't you?" 9. "I would appreciate it if you could delay your presentation." 10. "I'm asking you to delay your presentation." 11. "You will have to delay your presentation." 12. "Delay your presentation until the next class." WVONUI-FWNH (Homework) Situation: Imagine that you had a cold last week. Your symptoms were severe enough to stay home and rest but not severe enough to go and see a doctor. While you are almost over with your cold now, you will not be able to finish the assignment due tomorrow in one of your classes. Your professor (female) made it clear that no points will be given for late homework without a legitimate excuse. While you do not have an official medical excuse, you cannot afford to get a zero point on the homework. Suppose you do dot know the professor very well except for the class. Now you want to ask the professor (female) to give you an extension. Tactic 1: l "I was not able to finish the homework." 2 "I could use an extra day for the assignment." 3. "Is an extension allowed for this assignment?" 4 "Would it be alright if I ask you for an extension in submitting this homework?" 5 "May I ask you for an extension in submitting this homework?" 6 "Could you give an extension in submitting this homework?" 7. "Will you give me an extension in submitting this homework?" 8 "How about giving me an extension in submitting this homework?“ 9 "I would like you to give me an extension in submitting this homework." . 10. "I must ask you to give me an extension in submitting this homework." 11. "You should give me an extension in submitting this homework." 12. "Give me an extension in submitting this homework." extension. Tactic 2: 1. "I was sick last week." 2. "I wish I could have another day to do this assignment properly." 3. "Is it possible to get an extension for this assignment?" 4. "I wonder if you could give me an extension in submitting this homework?" 5. "Can I ask you for an extension in submitting this homework?" 6. "Can you give me an extension in submitting this homework?" 7. "Won't you give me an extension in submitting this homework?" 8 "You will give me an extension in submitting this homework, won't you?" 9. "I would appreciate it if you give me an extension in submitting this homework." 10. ll. 12. "I'm asking you to give me an extension in submitting this homework." "You will have to give me an extension in submitting this homework." "Let me have an extension in submitting this homework." BIBLIOGRAPHY BIBLIOGRAPHY Andersen, E.S. (1990). Acquiring communicative competence: Knowledge of register variation. In R.C. Scarcella, E.S. Andersen & S.D. Krashen (Eds.), Developing communicative competence in a second language (pp.5-25). New York: Newbury House Publishers. Applegate, J.L. (1982). The impact of construct system development on communication and impression formation in persuasive contexts. Communication Monographs, 49, 277-289. Argyle, M., Furnham, A. & Graham, J.A. (1981). Social situations. Cambridge: Cambridge University Press. Austin, J.L. (1962). How to do things with words. Cambridge, Mass: Harvard University Press. Bach, K & Harnish, R. (1979). Linguistic communication and speech acts. The MIT Press. Bales, R.F. (1950). Interaction process analysis. Cambridge: Addison-Wesley. Becker, J.A. (1982). Children’s strategic use of requests to mark and manipulate social status. In S.A. Kuczac 11 (Ed.), Language development (pp.l-35), Hillsdale, NJ: Lawrence Erlbaum Associates. Becker, J.A., Kimmel, H.D. & Bevill, M.J. (1989). The interactive effects of request form and speaker status on judgments of requests. Journal of Psycholinguistic Research, 18, 521- 531. Beebe, L.M., Takahashi, T. & Uliss-Weltz, R. (1990). Pragmatic transfer in ESL refusals. In R.C. Scarcella, E.S. Andersen & S.D. Krashen (Eds.), Developing communicative competence in a second language (pp.55-73). New York: Newbury House Publishers. Berger, C.R. (1987). Planning and scheming: Strategies for initiating relationships. In R. Burnett, P. McGhee & D.D. Clarke (Eds.), Accounting for relationships: Explanation, representation and knowledge (pp.lS8-l74). London: Methuen. Berger, C.R. (in press). A plan-based approach to strategic communication. In D.E. Hewes (Ed.), Cognitive bases of interpersonal communication. Hillsdale, NJ: Lawrence Erlbaum Associates. Berger, C.R. & Kellermann, K. (1983). To ask or not to ask: Is that a question? In R. Bostrom (ed.), Commmunication Yearbook 7 (pp.342-368). Sage: Beverly Hills. 148 149 Berger, C.R. & Kellermann, K. (in press). Acquiring social information. In J. Daly & J. Weimann (Eds.), Communicating strategically. Hillsdale: Erlbaum. Berger, C.R., Karol, S.H. & Jordan, J.M. (1989). When a lot of knowledge is dangerous thing: The debilitating effects of plan complexity on verbal fluency. Human Communication Research, 1g, 91-119. Berry, J. (1980). Introduction to methodology. In H. Triandis 6 J. Berry (Eds.), Handbook of Cross-cultural Psychology (Vol.2). Boston: Allyn & Bacon. Blum-Kulka, S. (1982). Learning to say what you mean in a second language: A study of the speech act performance of learnerrs of Hebrew as a second language. Applied Linguistics 3, 29- 59. Blum-Kulka, S. (1983). Interpreting and performing speech acts in a second language--A cross-cultural study of Hebrew and English. In N. Wolfson & E. Judd (Eds.), Sociolinguistics and language acquisition (pp.36-55). Rowley, MA: Newbury House Publishers. Blum-Kulka, S. (1987). Indirectness and politeness in requests: Same or different? Journal of Pragmatics, 11, 131-146. Blum-Kulka, S., Danet, B. & Gherson, R. (1985). The language of requesting in Israeli society. In J. Forgas (Ed.), Langpage and social situation (pp.ll3-l4l). New York: Springer Verlag. Blum-Kulka, S. 8 House, J. (1989). Cross-cultural and situational variation in requesting behavior. in S. Blum-Kulka, J. House 6 G. Kasper (Eds.), Cross-cultural pragmatics: Requests and apologies (pp.l23-154). Norwood, NJ: Ablex Publishing Co. Blum-Kulka, 8., House, J. 6 Kasper, G. (Eds.) (1989). Cross- cultural pragmatics: Requests and apologies. Norwood, NJ: Ablex Publishing Co. Boster, F.J. (1988). Comments on the utility of compliance- gaining message selection tasks. Human Communication Research, 12, 169-177. Brislin, R. (1983). Cross-cultural research in psychology. Annual Review of Psychology, 33, 363-400. Brown, P. & Levinson, S. (1978). Universals in language usage: Politeness phenomena. In E.N. Goody (Ed.), Questions and politeness (pp.56-289). Cambridge: Cambridge University Press. 150 Burke, J.A. & Clark, R.A. (1982). An assessment of methodological options for investigating the development of persuasive skills across childhood. Central States Speech Journal, 33, 437-445. Burleson, B.R., Wilson, S.R., Waltman, M.S., Goering, E.M., Ely, R.K. & Whaley, 8.3. (1988). Item desirability effects in compliance-gaining. Human Communication Research, 13, 429- 486 Burleson, B.B. & Wilson, S.R. (1988). On the continued undesirability of item desirability: A reply to Boster, Hunter, and Seibold. Human Communication Research, 13, 178- 191. Buss, A.H. (1980). Self-consciousness and social anxiety. San Franscisco: W.H. Freeman and Co. Canary, D.J., Cody, M.J. & Marston, P.J. (1986). Coal types, compliance-gaining and locus of control. Journal of Language and Social Psychology, 3, 249-269. Canary, D.J. & Spitzberg, B.H. (1987). Appropriateness and effectiveness perceptions of conflict strategies. Human Communication Research, 14, 93-118. Canary, D.J. & Spitzberg, B.H. (1989). A model of the perceived competence of conflict strategies. Human Communication Research, 13, 630-649. Carrell, P.C. & Konneker, D.H. (1981). Politeness: Comparing native and nonnative judgments. Language Learning, 31, 17- 30. Chen, G-M (1989). Relationships of the dimensions of intercultural communication competence. Communication Qmarterly, fl, 118-133. Cherry, R.D. (1988). Politeness in written persuation. Journal of Pragmatics, 13, 63-81. Clark, H.H. & Lucy, P. (1975). Understanding what is meant from what is said: A study in conversationally conveyed requests. Journal of Verbal Learning and Verbal Behavior, 13, 56-72. Clark, H.H. & Schunk, D.H. (1980). Polite responses to polite requests. Cognition, 3, 111-143. Clark, H.H. & Schunk, D.H. (1981). Politeness in requests: A rejoiner to Kemper and Thissen. Cognition, 2, 311-315. 151 Condon, J. (1985). So near the United States: Notes on communication between Mexicans and North Americans. in L.A. Samovar and R.E. Porter Eds.), Intercultural communication: A reader (pp.86-9l). Belmont, CA: Wadsworth Publishing Co. Conley, J.M., O'Barr, W.M. 6 Lind, E.A. (1978). The power of language presentational style in the courtroom. Duke Law Journal, 1978, 1375-1399. Craig, R.T., Tracy, K. 6 Spisak, F. (1986). The discourse of requests: Assessment of a politeness approach. Human Communication Research, 33, 437-468. Davison, M.L. (1983). Multidimensional scaling. New York: John Wiley 6 Sons. - Delia, J.G., Kline, S. 6 Burleson, B.R. (1979). The development of persuasive communication strategies in kindergarteners through twelfth-graders. Communication Monographs, 33, 241- 256. Derlega, V.J. 6 Grzelak, J. (1979). Appropriateness of self disclosure. In G.J. Chelune 6 Associates (Eds.), Self- disclosure: Origins, patterns, and implications of openness in interpersonal relationships (pp.151-l76). San Francisco: Jossey-Bass. Dillard, J.P. (1988). Compliance-gaining message-selection: What is our dependent variable? Communication Monographs, 33, 162-183. Dillard, J.P. 6 Burgoon, M. (1985). Situational influences on the selection of compliance-gaining messages: Two tests of the predictive utility of the Cody-McLaughlin Typology. Communication Monographs, 33, 289-304. Donohue, W.A. 6 Diez, M.E. (1985). Directive use in negotiation interaction. Communication Monographs, 33, 305-318. Dore, J. (1977). Children's illocutionary acts. In R. Freedle (Ed.), Discourse production and comprehension (Vol.1) (pp.227-244). Norwood, NJ: Ablex. Douglas, W. (1987). Affinity-testing in initial interactions. Journal of Social and Personal Relationships, 3, 3-15, Duncan, 6. 6 Fiske, D.W. (1977). Face-to-face interaction: Research, methods, and theogy. Hillsdale: Erlbaum. Eisenberg, E.M. (1984). Ambiguity as strategy in organizational communication. Communication Monographs, 33, 1713-1722. 152 Emmons, R.A. (1989). The personal striving approach to personality. in L. Pervin (Ed.), Goal concepts in personality 6 social psychology. (pp.87-l26). Lawrence Erlbaum Associates. Ervin-Tripp, S. (1976). Is Sybil there? the structure of some American English directives. Language in Society, 3, 25-66. Ervin-Tripp, S. (1977). Wait for me, roller skate! In S. Ervin- Tripp 6 C. Mitchell-Kernan (Eds.), Child discourse (pp.165- 188). New York: Academic Press. Faerch, C. 6 Kasper, G. (1984). Pragmatic knowledge: Rules and procedures. Applied Linguistics, 3, 214-225. Fraser, B. (1975). Hedged performatives. In P.Cole 6 J.L. Morgan (Eds.), Symtax and semantics 3 (pp.187-210). New York: Academic Press. Fraser, B. (1978). Acquiring social competence in a second language. RECL Journal, 2 (2), 1-21. Fraser, B. 6 Nolen, W. (1981). The association of deference with linguistic form. International Journal of the Sociology of Language, 33, 93-109. Fraser, 8., Rintell, E. 6 Walters, J. (1980). An approach to conducting research on the acquisition of pragmatic competence in a second language. in Larsen-Freeman, D. (Ed.), Discourse analysis in second langmage research (pp.75-90). Rowley, MA: Newbury House Publishers. Frese, M., Stewart, J. 6 Hannover, B. (1987). Coal orientation and planfulness: Action styles as personality concepts. Journal of Personality and Social Psychology, 33, 1182- 1194. Garvey, C. (1984). Children's talk. Cambridge, Mass: Harvard University Press. Goffman, E. (1967). Interaction ritual. Garden City: Doubleday/Anchor. Gordon, D. 6 Ervin-Tripp, S. (1984). The structure of children's requests. in R.L. Schiefelbusch 6 J. Pickar (Eds.), The acquisition of communicative competence (pp.295-321)T—— Baltimore, MD: University Park Press. Gordon, D. 6 Lakoff, G. (1975). Conversational postulates. In P. Cole 6 J. Morgan (Eds.), Syptax and semantics 3: Speech acts (pp.83-106). New York: Academic Press. 153 Green, G. (1973). How to get people to do things with words: The question of whimperatives. In R. Shuy (Ed.), Some new directions in linguistics. Washington, D.C.: Georgetown University Press. Greene, J.O. 6 Lindsey, A.E. (1989). Encoding processes in the production of multiple-goal messages. Human Communication Research, 33 (1), 120-140. Grice, H.P. (1975). Logic and conversation. In P. Cole 6 J. Morgan (Eds.), Syntax and semantics 3: Speech acts (pp.107- 142). New York: Academic Press. Gudykunst, W.B. 6 Kim, Y.Y. (1984). Communicating with strangers. New York: Random House. Gudykunst, W.B. 6 Ting-Toomey, S. with Chua, E. (1988). Culture and interpersonal communication. Newbury Park: Sage Publications. Gudykunst, W.B., Yang, S-M. 6 Nishida, T. (1985). A cross- cultural test of uncertainty reduction theory: Comparisons of acquaintances, friends, and dating relationships in Japan, Korea, and the United States. Human Communication Research, 13 (3), 407-454. Gumperz, J.J. (1978). The conversational analysis of interethnic communication. In E. Lamar Ross (Ed.), Interethnic communication. Athens, CA: University of Georgia Press. Gumperz, J.J. (1982). Discourse strategies. Cambridge, England: Cambridge University Press. Hall, E.T. (1976). Beyond culture. New York: Doubleday Hall, E.T. 6 Hall, M.R. (1985). Hidden differences: Doing business with Japanese. Garden City, NY: Anchor Press. Hample, D. 6 Dallinger, J.M. (1985). Unused compliance gaining strategies. In J.R. Cox, M.O. Sillars 6 6.8. Walker (Eds.), Argument and Social Practice: Proceedings of the fourth SCA/AFA conference on argumentation (pp.675-691). Annandale, VA: Speech Communication Association. Hample, D. 6 Dallinger, J.M. (1987). Individual differences in cognitive editing standards. Human Communication Research, 33, 123-144. Hayes-Roth, B. 6 Hayes-Roth, F. (1979). A cognitive model of planning. Cognitive Science, 3, 275-310. Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. 154 Hermann, T. (1983). Speech and situation: A psychological conception of situated speaking. New York: Springer-Verlag. Hill, 8., Ide, S., Ikuta, 8., Kawasaki, A. 6 Ogino, T. (1986). Universals of linguistic politeness: Quantitative evidence from Japanese and American English. Journal of Pragmatics, 19, 347-371. Hofstede, G. (1979). Value systems in forty countries: Interpretation, validation and consequences for theory. in L. H. Eckensberger, W.J. Lonner 6 Y.H. Poortinga (Eds.), Cross-cultural contributions to psychology (pp.389-407). Lisse: Swets and Zeitlinger. Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly Hill, CA: Sage. Holtgraves, T. 6 Yang, J.N. (1990). Politeness as universal: Cross-cultural perceptions of request strategies and inferences based on their use. Journal of Personality and Social Psychology, 33 (4), 719-729. Hoppe-Graff, S., Herrmann, T., Winterhoff-Spurk, P. 6 Mangold, R. (1985). Speech and situation: A generral model for the process of speech production. in J.P. Forgas (Ed.), Langaage and social situations (pp.8l-95). New York: Springer-Verlag. House, J. 6 Kasper, G. (1981). Politeness markers in English and German. in F. Coulmas (Ed.), Conversational routine: Explorations in standardized communication situations and prepatterned speech (pp.157-185). New York: Mouton Publishers. House, J. 6 Kasper, G. (1987). Interlanguage pragmatics: Requesting in a foreign language. in W. Lorscher 6 R. Schulze (Eds.), Perspectives on language in performance. (pp. 1250-1288). Gunter Narr Verlag Tubingen. Hui, C.H. (1984). Individualism-collectivism: Theory,gmeasurement and, its relation to reward allocation. Unpublished doctoral dissertation, University Illinois at Urbana-Champaign. Hui, C.H. 6 Triandis, H.C. (1986). Individualism-collectivism: A study of cross-cultural researchers. Journal of Cross- cultural Psycholggy, 31, 225-248. Hui, C.H. 6 Villareal, M.J. (1989). Individualism-collectivism and psychological needs: Their relationships in two cultures. Journal of Cross-Cultural Psychology, 39 (3), 310- 323. Hunter, J.E. (1988). Failure of the social desirability response set hypothesis. Human Communication Research, 13, 162-168. 155 Hunter, J.E. 6 Boster, P.J. (1987). A model of compliance-gaining message selection. Communication Monograph, 33, 61-69. Hunter, J.E. 6 Gerbing, D.W. (1982). Unidimensional measurement, second order factor analysis, and causal models. Research in Organizational Behavior, 3, 267-320. Hunter, J.E. 6 Lim, T-S (1987). LIMSTAT. Unpublished manuscript, Department of Communication, Michigan State University, E.Lansing, MI. Hymes, D. (1972). Models of interaction of language and social life. In J.J. Gumperz 6 D. Hymes (Eds.), Directions in sociolingm1stics: The ethnography of communication (pp.35- 72). New York: Holt, Reinhart 6 Wilson. Hymes, D. (1974). Foundations in sociolingmistics: An ethnographic approach. Philadelphia, PA: University of Philadelphia Press. Ishii, S. (1985). Thought patterns as modes of rhetoric: The United States and Japan. in L.A. Samovar 6 R.E. Porter (eds.), Intercultural communication: A reader (pp.97-102). Belmont, CA: Wadsworth Publishing Company. James, S.L. (1978). Effect of listener age and situation on the politeness of children's directives. Journal of Psycholinguistic Research, 3, (4), 307-317. Johnson, J. 6 Tuttle, J. (1989). Problems in intercultural research. Handbook of International and Intercultural Communication Research (pp.461-483). Sage. Jones, E. E. (1964). Ingratiation. New York: Appleton-Century. Jordan, J.M. 6 Roloff, M.E. (1990). Acquiring assistance from others: The effect of indirect requests and relational intimacy on verbal compliance. Human Communication Research, 13 (4). 519-555. Kaplan, R. (1970). Cultural thought patterns in intercultural communication. Language Learning, 33, 3. Katriel, T. (1986). Talking straight: Dugri speech in Israeli Sabra culture. Cambridge: Cambridge University Press. Kellermann, K. (1988) Metagoals in conversation. Paper presented at the Temple University Discourse Conference, Philadelphia. Kellermann, K. (in press). A theory of meta-goals in conversation. In K. Tracy (Ed.), Goals in discourse. Englewood Cliffs: Lawrence Erlbaum. 156 Kellermann, K. 6 Berger, C.R. (1984). Affect and the acquisition of social information: Sit back, relax, and tell me about yourself. In R. Bostrom (Ed.), Communication Yearbook 8 (pp.412-445). Newbury Park: Sage. Kellermann, K. 6 Kim, M.S. (1991). Tactical choices in the pursuit of social goals. Paper presented to the Information Systems Division of the International Communication Association, Chicago. Kellermann, K., Reynolds, R. 6 Chen, J. (1989). Strategies of conversational retreat: When parting is not a sweet sorrow. Paper presented at the annual meeting of the Speech Communication Association, San Francisco. Kemper, S. 6 Thissen, D. (1981a). How polite?: A reply to Clark 6 Schunk. Cognition, 3, 305-309. Kemper, S. 6 Thissen, D. (1981b). Memory for the dimensions of requests. Journal of Verbal Learning and Verbal Behavior, 39, 552-563. Kim, M.S. (in press a). Culture-based conversational constraints in explaining cross-cultural strategic competence. International and Intercultural Communication Annual 17. Kim, M.S. (in press b). A comparative analysis of nonverbal expressions as portrayed by Korean and American print-media advertising. Howard Journal of Communication. Labov, W. 6 Fanshel, D. (1977). Therapeutic discourse: Psychotherapy as conversation. New York: Academic Press. Lakoff, R. (1977). What you can do with words: Politeness, pragmatics and performatives. in A. Rogers, B. Wall 6 J. Murphy (Eds.), Proceedings of the Texas conference on performatives, presuppositions and implicatures (pp.79- 105), Arlington, VA: Center for Applied Linguistics. Leech, G.N. (1983). Principles of pragmatics. New York: Longman. Leung, K. (1987). Some determinants of reactions to procedural models for conflict resolution: A cross-national study. Journal of Personality and Social Psychology, 33 (5), 898- 908. Lim, T-S (1988). A new model of politeness in discourse. Unpublished doctoral dissertation, Department of Communication. Michigan State University. Marwell, G. 6 Schmitt, D.R. (1967). Dimension of compliance- gaining behavior: An empirical analysis. Sociometry, 39, 350-364. 157 McCann, C.D. 6 Higgins, E. T. (1988). Motivation and affect in interpersonal relations: The role of personal orientations and discrepancies. in L. Donohew, H. Sypher 6 T. Higgins (1988). Communication, social cognition, and affect (pp.53- 79), Hillsdale, NJ: Lawrence Erlbaum Associates. McClelland, D.C. (1985). Human motivation. Glenview, IL: Scott, Foresman 6 Company. Miller, G.R., Boster, F.J., Roloff, M.E. 6 Seibold, D.R. (1987). MBRS rekindled: Some thoughts on compliance gaining in interpersonal settings. in M.E. Roloff 6 C.R. Miller (Eds.), Interpersonal processes: New directions in communication research (pp.89-1l6). Newbury Park, CA: Sage Publications. Miller, G.A., Galanter, E. 6 Pribram, K.H. (1960). Plans and the structure of behavior. New York: Holt, Rinehart 6 Winston. Okabe, K. (1987). Indirect speech acts of the Japanese. in D. Kincaid (ed.) Communication theory: Eastern and Western perspectives (pp.127-136). New York: Academic Press. Park, M. (1979). Communication styles in two different cultures: Korean and American. Seoul, Korea: Han Shin Publishing Company. Read, S.J., 6 Miller, L.C. (1989). Inter-personalism: Toward a goal-based theory of persons in relationships. in L. Pervin (Ed.), Goal concepts in personality and social psychologyt (pp.413-472), New York: Lawrence Erlbaum Associates. Reiss, N. (1985). Speech act taxonomy as a tool for ethnographic description: An analysis based on videotapes of continuous behavior in two New York households. Amsterdam: John Benjamins Publishing Company. Rintell, E. (1981). Sociolinguistic variation and pragmatic ability: A look at learners. International Journal of the Sociology of Language, 33, 11-34. Rintell, E.M. 6 Mitchell, G.J. (1989). Studying requests and apologies: An inquiry into method. in S. Blum-Kulka, J. House 6 G. Kasper (Eds.), Cross-cultural pragmatics: Requests and apologies (pp.248-272). Norwood: NJ: Ablex Publishing Co. Rokeach, M. (1973). The nature of human values. New York: The Free Press. Rubin, K.H. 6 Borwick, D. (1983). The communicative skills of children who vary with regard to sociability. In H.E. , Sypher 6 J.L. Applegate (Eds.), Understanding interpersonal communication: Social cognitive and strategic processes in children and adults. Beverly Hills: Sage Publications. 158 Rubin, R.E., Perse, E.M. 6 Barbato, C. (1988). Conceptualization and measurement of interpersonal communication motives. Human Communication Research, 33, 602-628. Rule, B.G. 6 Bisanz, G.L. (1987). Goals and strategies of persuasion: A cognitive schema for understanding social events. in Zanna, M., Olson, J., 6 Herman, C. (Eds.), Social influence: The Ontario symposium 5 (pp.l85—206). Hillsdale, NJ: Lawrence Erlbaum Associates. Salzman, K. 6 Hunter, J.E. (1983). The voicer/nonvoicer distinction: A dimension in the experience of conscious thought. Unpublished manuscript. Sanford, D.L. 6 Roach, J.W. (1987). Imperative force in request forms: The demanding vs. pleading dimension of directives. Paper presented at the International Communication Association Convention, Montreal, Canada, May. Scarcella, R. 6 Brunak, J. (1981). On speaking politely in a second language. International Journal of sociology of Language, 33, 59-75. Schank, R.C. 6 Abelson, R.F. (1977). Scripts, plans, goals, and understanding. Hillsdale, NJ: Lawrence Erlbaum Associates. Schank, R.C. 6 Childers, P. (1984). The cognitive computer: On language, learning and artificial intelligence. Reading, Mass.: Addison-Wesley Publishing Co. Schonpflug, W. (1985). Goal directed behavior as a source of stress: Psychological origins and consequences of inefficiency. In M. Frese 6 J. Sabini (Eds.), Goal directed behavior: The concept of action in psychology (pp.l72-188). Hillsdale, NJ: Lawrence Erlbaum Associates. Scollon, R. 6 Scollon, S. (1981). Narrative, literacy and face in interethnic communication. Norwood, NJ: Ablex Publishing Corporation. Searle, J. (1969). Speech acts. Cambridge, England: Cambridge University Press. Searle, J. (1975). Indirect speech acts. in P. Cole 6 J.L. Morgan (Eds.), Syptax and Semantics 3 (pp.59-106). New York: Academic Press. Searle, J. (1979). Expression and meaning: Studies in the theory of speech acts. Cambridge, England: Cambridge University Press. Seibold, D.R. (1988). A response to "Item desirability in compliance-gaining research". Human Communication Research, 12. 152-161. 159 Semmer, N.6 Frese, M. (1985). Action theory in clinical psychology. In M. Frese 6 J. Sabini (Eds.), Goal directed behavior: The concept of action in psychology (pp.296-310). Hillsdale, NJ: Lawrence Erlbaum Associates. Slobin, D.I. (1975). The more it changes...on understanding language by watching it move through time. Papers and Reports on Child Language Development (pp.l-30). University of California, Berkely. Soskin, W.F. 6 John, V.P. (1963). The study of spontaneous talk. in R.C. Barker (Ed.), The stream of behavior: Explorations of its structure and content (pp.228-28l). New York: Appleton-Century-Crofts. Spitzberg, B. H. 6 Cupach, W.R. (1984). Interpersonal communication competence. Sage. Spitzberg, B. H. 6 Cupach, W.R. (1989). Handbook of interpersonal competence research. New York: Springer-Verlag. Steffen, V.J. 6 Eagly, A.H. (1985). Implicit theories about influence style: The effects of status and sex. Personality and Social Psychology Bulletin, 33, 191-205. Stewart, E.C. (1972). American cultural patterns: A cross- cultural perapective. LaGrange Park, Ill: Intercultural Network Inc. Street, R.L. 6 Cappella, J.N. (1985). Sequence and pattern in communicative behavior: A model and commentary. in R.L. Street 6 J.N. Cappella (Eds.), Sequence and pattern in communicative behavior. Baltimore, MD: Edward Arnold Publishers. Summers, D.A., Taliaferro, J.D. 6 Fletcher, D.J. (1970). Subjective vs. objective description of judgement policy. Psychonomic Sicence, 33, 249-250. Tannen, D. (1981). Indirectness in discourse: Ethnicity as conversational style. Discourse Processes, 3 (4), 221-238. Tannen, D. (1984). Conversational style: Analyzing talk among friends. Norwood, NJ: Ablex. Ting-Toomey, S. (1987). Intercultural conflicts: A face- negotiation model. Paper presented at the annual meeting of the Temple University Discourse Conference, Philadelphia, PA. March. Ting-Toomey, S. (1988). Intercultural conflict styles: A face- negotiation theory. in Y.Y.Kim 6 W.B.Gudykunst (Eds.), Theories in intercultural communication (pp.213-238). Newbury park, CA: Sage publication. 160 Tracy, K. (1990). The many faces of facework. In H. Giles 6 W.P. Robinson (Eds.), Handbook of Langpage and Social Psychology (pp.209-226). New York: John Wiley 6 Sons. Tracy, K., Craig, R.T., Smith, M. 6 Spisak, F. (1984). The discourse of requests: Assessments of a compliance-gaining approach. Human Communication Research, 39 (4), 513-538. Triandis, H.C., Bontempo, R., Villareal, M.J., Asai, M. 6 Lucca, N. (1988). Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 33 (2), 323-338. Triandis, H.C., et al. (1986). The measurement of the etic aspects of individualism and collectivism across cultures. Australian Journal of Psychology, 33 (3), 257-267. van Dijk, T. 6 Kintsch, W. (1983). Strategies of discourse comprehension. New York: Academic Press. von Cranach, Kalbermatten, U., Indermuhle, K. 6 Gugler, B. (1982). Goal-directed action. London: Academic Press. von Cranach, M., Machler, E. 6 Steiner, V. (1985). The organization of goal-directed action: A research report. In Ginsberg. G.P., Brenner, M. 6 von Cranach, M. (Eds.), Discovery strategies in the psychology of action (pp. 19- 62). London: Academic Press. Walters, J. (1979). Strategies for requesting in Spanish and English: Structural similarities and pragmatic differences. Language Learning, 33, 277-293. Walters, J. (1981). Variation in the requesting behavior of bilingual children. International Journal of the Sociology of Language, 77--92. Weary, G 6 Arkin, R.M. (1981). Attitudinal self-presentation. In Harvey, J.H., Ickes, W.J. 6 Kidd, R. (Eds.), New directions in attribution theory and research 3. Hillsdale, NJ: Erlbaum. Wegner, D.M. 6 Vallacher, R.R. (1977). Implicit psychology: An introduction to social cognition. New York: Oxford University Press. Weizman, E. (1989). Requestive hints. In S. Blum-Kulka, J. House 6 G. Kasper (Eds.), Cross-cultural ppagmatics: Reqpests and apologies (pp.71-95). Norwood, NJ: Ablex Publishing Corporation. 161 Wheeler, L., Reis, H.T. 6 Bond, M.H. (1989). Collectivism- Individualism in everyday social life: The middle kingdom and melting pot. Journal of Personality and Social Psychology, 33 (1), 79-86. Wiemann, J.M. 6 Backlund, P. (1980). Current theory and research in communicative competence. Review of Educational Research, 39, 185-199. Wierzbicka, A. (1985). Different cultures, different languages, different speech acts. Journal of Pragmatics, 3, 145-178. Wiezman, E. (1989). Requestive hints. in S. Blum-Kulka, J. House 6 G. Kasper (Eds.), Cross-cultural pragmatics: Requests and apologies (pp.71-95). Norwood, NJ: Ablex Publishing Co. Wilensky, R. (1981). Meta-planning: Representing and using knowledge about planning in problem solving and natural language understanding. Cognitive Science, 3, 197-233. Wilensky, R. (1983). Planning and understanding: A computational approach to human reasoning. Reading: Addison-Wesley. Wilson, S.R., Kim, M.S. 6 Meischke, H. (in press). Evaluating Brown and Levinson's politeness theory: A revised analysis of directives and face. Research on Language and Social Interaction. Wilson, S.R. 6 Putnam, L.L. (1990). Interaction goals in negotiation. in J.A. Anderson (Ed.), Communication Yearbook 33, (pp. 374-427). Newbury Park: Sage Publications. Winograd, T. (1977). A framework for understanding discourse. in M.A. Just 6 P.A. Carpenter (Eds.), Cognitive processes in comprehension (pp.63-88), Hillsdale, NJ: Lawrence Erlbaum Associates. Wish, M. 6 Kaplan, S. (1977). Toward an implicit theory of interpersonal communication. Sociometry, 39, 234-246. Yum, J.0. (1988). The impact of Confucianism on interpersonal relationships and communication patterns in East Asia. Communication Monographs, 33, 374-388. ~14 -1 ~ .n-I-H... .. 1 \.,..,..'..,_, 4 -‘ "T“ -. .. . . .44-H. .~4.. p‘~:‘<-‘\ .- .. . ,, pm." 0 174' \_ . uw .~... ..~ '- --. .... ';" .‘4. . up“... ... ...,.., , 1 ...”... .,,,"’ 1. T" . 0.“. '4 n. pm '3 ‘4' .24.. ’3"... '13” ’ t "3441‘; w- “a. urn-.41.- - .»"'. - 4......“ ,, -- -L. .....L..‘- . 'J" -. . . - y- >- L3. 6.23:. ‘IN- m.‘ 4.1“... 11.44:.“ , «"VI‘ZMH- MICHIGQN STQTE 31,2110 Immrmnrmmmumwmm ”a“...