2‘in I 5mm”. a I . mell aJa. . aw L5. .t u 1) — \ wflhfi . . 1qu ._I,... 3‘ .. 7 .E . . ‘. . 9...: «I u, nuwhvh. - ‘ florwww NF. . C‘ ,1... My .21. w: . U2“. 1:. . . ,.A._, I . fl . :zwvrw} 1»?! . .mv...v.wufi«kww R. . . . . qukwwfi . , .... . ‘ 3% $3 $3.33.. Ea fig figfififififi in .fi ._ . , u 1 . .» {outfitt I y . fi... ..._f, . 5,10. ...,....:.. J. . v u I nl.o..r>.gv w r I\illlllllllllllllll v r W < \ fl .-——— This is to certify that the dissertation entitled I GENDER DIFFERENCES IN ADOLESCENTS' USE : OF SEX-TYPED LANGUAGE VARIABLES l AND CONVERSATIONAL PATTERNS g presented by Kristine Busk has been accepted, towards fulfillment ’ of the requirements for Ph-D- degreein 'PSyChOIqu ’ flhflfizfi Major professor Elaine Donelsqn, Dme June 6, 1985 MS U is an Afl'mmm've Action/Equal Opportunity Institution MSU LIBRARIES mm \. sz ‘UthNG MAllRlAlb Pldré in book draw to remove this cheukuut irom }our record. FIJLS wili be cnarged if book is )1 turned after the date 'm'*?¥"’"” 7 staltiped below. MW; 87 p; 17‘; lllfi Al N —+h ‘3 bef‘m [‘hs‘flqziu‘ I . :5. 300 n2f~‘ M: 77R 1 71339 ‘K- O yA’M. u' ‘LULHM Dec; 526100 [JUL 1 0i 2001 GENDER DIFFERENCES IN ADOLESCENTS' USE OF SEX-TYPED LANGUAGE VARIABLES AND CONVERSATIONAL PATTERNS By Kristine Busk A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1985 ©1986 KRISTINE BUSK All Rights Reserved ABSTRACT GENDER DIFFERENCES IN ADOLESCENTS' USE OF SEX—TYPED LANGUAGE VARIABLES AND CONVERSATIONAL PATTERNS By Kristine Busk Language samples gathered from 9th graders, 11th graders, and first year college students interacting in same and mixed-sex dyads were analyzed for gender differences in language variables and conversational patterns. There were no gender differences in the use of qualified speech, filled speech, intensifying adverbs, or swear words.‘ In mixed-sex dyads females used more questions than males, while in same-sex dyads males used more questions than females. Gender differences in conversation patterns indicated that males used more words and had longer utterances than females. In same-sex dyads female pairs took fewer, but longer turns than male pairs, while in mixed-sex dyads females took shorter turns than males. Gender differences in laughter were observed with females exhibiting more laughter than males. The pattern of results lend themselves to several important conclusions. First, in the use of specific language variables adolescents exhibited a lack of sex-typed behavior which may be due the experimental setting, or to changes in current sex-role standards. Second, gender differences in the pattern of turn-taking and question use indicates that male/male dyads are characterized by an interview style consisting of many questions and short turns. The female/ female dyads are characterized by a conversational style, with fewer questions and longer turns. This pattern suggests that the activity of talk may have different meaning for females and males. Third, the mixed-sex dyad is characterized by a female question-male response pattern, suggesting that females use questions to maintain conversational interaction when speaking with males. Fourth, while females exhibited more laughter than males, most of the laughter occurred in same-sex dyads. This was discussed as further indication of the different interaction styles of same and mixed-sex dyads. ACKNOWLEDGMENTS There are many individuals who contributed to the successful completion of this project. First, I would like to thank the members of my committee: Dr. Elaine Donelson, who has, with great skill and tact, guided my entire graduate career. To her I acknowledge a profound intellectual and professional debt. Dr. Ellen Strommen, whose support, respect, and good humor are gratefully felt by graduate students throughout the department. Dr. John McKinney, who insisted that I finish this project on time. Dr. Barrie Thorne, whose knowledge of the literature and editorial assistance are reflected throughout these pages. I would also like to express my appreciation to the fine undergraduate research assistants who helped on this project. They were an extremely bright, responsible, and loyal group of people: Carla Antico, R Cole Bouck, Jeannine Bryla, Seth Chafetz, Nancy Cohen, Lorne Demrose, Peggy Everson, Wendy Frankel, Bob Gonyea, Jim Helmer, Jack Hudson, Gail Holmes, Brian Karbel, Liz Katt, Heide Lindblom, Rhonda McDaniel, Tim Matuszewski, Keith Mellon, Mike Reznick, Liz Seaton, John Stewart, Kim Troy, Laura Wallace. 1'1 TABLE OF CONTENTS LIST OF TABLES. LIST OF FIGURES REVIEW OF THE LITERATURE. Introduction . . . Stereotypes and Perceptions of Sex-typed Speech . . Actual Talk: Linguistic and Conversational Patterns Development of Sex Differentiated Speech Styles . . . . . . . . . . . Statement of the Problem . Hypotheses . . . . METHODS . RESULTS . DISCUSSION. Sex- -typed Speech Speech Production and Interruptions. Listening Behavior . Conclusion APPENDICES APPENDIX A. APPENDIX B. APPENDIX C. APPENDIX D. REFERENCES. Page iv Table 10 11 12 13 14 15 16 LIST OF TABLES Number of Subjects Per Cell. Frequency of Use of Sex- -typed. Language Variables. . . Mean Rate of Use for Intensifying Adverbs. Mean Rate of Use for Tag Questions Mean Rate of Use of Tag Questions Without 11th Grade Dyad. Mean Rate of Use for Tag Questions by College Students, Sex by Dyad Interaction. Mean Rate of Questions for Sex by Dyad Interaction. . Mean Number of Words Used. Mean Number of Utterances. Mean Number of Turns Means for Turn Taking Interaction. Mean Number of Words Per Utterance . Mean Utterances Per Turn Mean Utterances Per Turn Interaction . Mean Number of Occurrences of Laughter . Mean Rate of Supportive Listening Response . . . . . . . . . iv Page 47 58 6O 61 61 62 63 64 65 65 66 67 68 69 70 7O Figure LIST OF FIGURES Sex by dyad interaction for college students' use of tag questions. Sex by dyad interaction for rate of questions Sex by dyad interaction, turns. Sex by dyad interaction, utterances per turn . Page 62 63 66 69 REVIEW OF THE LITERATURE Introduction Interest in the differential use of language by women and men has a relatively long history. As early as 1913, anthropologist and feminist Elsie Clews Parsons wondered "may we anywhere expect to find men and women speaking quite the same language?" (p. 149). Parsons observed that women are excluded from many male dominated activities such as sports, politics, and the military, and that as a consequence women and men develop different "dialects." She also observed that women and men use different expletives, with profanity reserved for men, and women using modified expletives such as "my goodness" or "oh dear." The distinguished linguist Otto Jespersen (1922) also noticed that women's speech differs from men's along a number of dimensions. Women, Jespersen claimed, show a tendency to use intensive adverbs, exaggerated stress, and phonetic emphasis more than men. He claimed that women have smaller vocabularies than men, and frequently leave sentences unfinished. Finally, Jespersen speculated that due to women's "flighty nature," they speak more readily and with less precision of thought than men. In a 1954 commentary, psychologist Theodore Reik proposed that the sexes actually speak "different languages." He noted I that a listener may interpret the same phrase differently when uttered by a female or a male speaker. Similarly, the spoken word may have different semantic value for the two sexes. Reik noted that women, in general, avoid rough, coarse language, but that both sexes avoid strong language when in mixed-sex company. As had both Parsons and Jespersen, Reik observed that certain adjectives and phrases, such as "divine,“ "sweet," or "adorable" are spoken more readily by women than men. Indeed, Reik concluded, "an attentive and perceptive semanticist could easily publish a vocabulary of the different expressions and colloquialisms men and women use" (p. 15). In her 1972 book, The Sex Game, sociologist Jessie Bernard includes an entire chapter on the talk of women and men. Among other things Bernard concludes that women excel in expressive talk, while men excel in instrumental talk, that women are uncomfortable with and unable to express themselves in a debating style, and that women and men generally have different topics of interest in conversational interaction. Mary Ritchie Key published an article in 1972, followed by a book in 1975, in which she speculates on some of the differences in women's and men's speech. Key maintains that women exhibit patterns of uncertainty and indefiniteness in their speech. Women, she claims, avoid the use of swear words and more frequently use tag questions, proper pronunciation and grammatical structure, role-related words, intensifying adverbs, and adjectives similar to those identified by Reik. Further, Key remarked upon the similarities between women's speech and the speech of children. In the same year Robin Lakoff (1975) published her influential book, Language and Woman's Place, in which she speculates on nine specific ways that women's language differs from men's. Women's language, according to Lakoff, is characterized by the use of words related to women's work, such as cooking or sewing. Women's language is also characterized by the use of empty adjectives, such as "divine" or "lovely," the use of questions rather than declaratives, especially tag questions, and the use of hedges and other words, phrases, or intonations conveying a sense of uncertainty. Further, she argues that women's language includes more intensifying adverbs, hypercorrect grammar, and superpolite forms. Finally, Lakoff claims that women rarely tell jokes, and that women often add emphasis to their speech, or as Lakoff puts it "speak in italics." Such speculation about differences in language use by men and women has prompted researchers to attempt empirical examinations of gender differentiated speech styles. Reviewed will be three areas of research relating to gender differences in speech styles; 1) stereotypes and perceptions associated with sex-typed speech, 2) the differential use by women and men of language and conversational variables, and 3) gender differences in language directed toward and used by children. It should be noted that this constitutes an update of material previously reviewed by Busk (1982). Stereotypes and Perceptions of Sex-typed Speech People tend to be quite consistent in their perceptions of what is considered appropriate speech for women and men. Often women are viewed as more talkative and polite than men, and women's speech is perceived as uncertain, trivial, and powerless. Men, on the other hand, are often stereotyped as terse and inexpressive, while their speech is perceived to be direct, strong, and assertive. Attempting to identify language forms associated with male and female speakers, Edlesky (1976a) presented 24 written statements containing different language variables to groups of adults and children. The language forms included items such as swear words, intensifying adverbs, politeness forms, tag questions, and direct and indirect imperatives. These forms were chosen because they have been discussed by writers such as Key and Lakoff as more characteristic of one sex or the other. Subjects were asked to assign each of the statements as typical of either a female or male speaker. A subjects' response was considered "correct" if in agreement with the literature on sex-typed speech. The results indicated a developmental progression in the identification of statements as sex-typed. First graders, as a group, identifed only two of the 24 statements as sex-typed. Third graders identified ten of the statements as sex-typed. Interestingly, sixth graders were more stereotypic in their assignments of sex-typed statements than were adults, identifying 14 statements as specifically spoken by a female or male, with the remaining ten statements identified as most probably spoken by a female or male. Adults, on the other hand, assigned only eight statements to females or males exclusively. The remaining 16 statements were considered variable, although most adults assigned them to the "correct" gender category. Edelsky contends that the development of the ability to recognize statements as sex-typed is an important aspect of communicative competence, or "the ability to use language in socially appropriate ways" (p. 47). In a similar study using written statements, Siegler and Siegler (1976) found that college students attributed strong assertive statements to male speakers, and tag questions to female speakers. Modified assertions occupied a middle ground, being attributed to both female and male speakers. A series of related studies have been conducted which examine the perceptions associated with sex-typed speech. For example, Siegler and Siegler (1976) assessed perceptions of the relative intelligence associated with each of the three types of statements found to be assigned to male or female speakers. They found that the strong assertions were rated as most intelligent, tag questions as least intelligent, and modified assertions again occupied an intermediate position. Newcombe and Arnkoff (1979) looked at three language forms, tag questions, qualified speech, and compound requests, which are stereotypically associated with greater use by females. They found that these forms, whether spoken by a male or female stimulus person, were rated by college students as less assertive, more qualified, and warmer than corresponding statements which deleted these forms. Sex-typed speech forms were varied in written messages which Berryman and Wilcox (1980) presented to college students for evaluation. The female sex-typed message included intensifying adverbs, tag questions, references to the self, feeling statements, and was 384 words long. The male sex-typed message contained no intensifying adverbs, tag questions or references to the self. However, it did include Obscenities, instances of slang and incorrect grammar, and was 338 words long. After reading one of the two messages, each subject was asked to make evaluations of both the speaker and the message. It was found that students readily attributed the female sex-typed message to a female speaker, and the male sex-typed message to a male speaker. Also, it was found that the female sex—typed message was seen as less commanding and more self-oriented than the male sex-typed message. The authors speculated that the use of tag questions and incomplete assertions may have contributed to this perception. Finally, a trend was noted for the female sex-typed message to be perceived as more compliant than the male sex-typed message. Berryman (1980), in a similar study, used tape recorded dyadic conversations as stimuli to be rated by subjects. Each audiotape consisted of male and female speakers who varied in their use of sex-typed speech forms. The female sex-typed forms included examples of socio-emotional speech, correct pronunciation of -ing word endings, no interruptions, and was 290 words long. The male sex-typed forms included examples of task oriented speech, interruptions, incorrect pronunciation of -ing word endings, and was 582 words. Subjects were asked to rate the speaker on a number of variables which related to four factors: credibility, extroversion, activity, and confidence. The results indicated that female sex-typed speech, regardless of the sex of the speaker, was seen as more credible than male sex-typed speech. Consequently, the author speculated that correct pronunciation and lack of interruptive behavior enhances a speaker's credibility. On the other hand, Berryman found that the extroversion factor was associated with the male sex-typed speech, again regardless of the sex of the speaker. Berryman speculated that interruptions and verbosity contributed to the perceptions of extroversion. The factors of activity and confidence showed only slight differences between the messages, with the female sex-typed message being associated with activity and the male sex-typed message associated with confidence. Note that in this study the female sex-typed message was shorter than the male sex-typed message, while in the Berryman and Wilcox (1980) study the reverse was the case. The subject of verbosity will be discussed in greater detail in a later section. In a related study (Erickson, Lind, Johnson, & O'Barr, 1978), both written and oral stimuli were constructed to vary the use of intensifying adverbs, qualified speech, filled speech, questioning forms, politeness forms, and formal grammar. The authors speculated that these language forms reflect not a female sex-typed speech style, but rather a "powerless“ speech. They found that subjects rating these 'messages considered masculinity and femininity to be associated more with the sex of the speaker than with the type of speech used. In addition, the powerless speech was perceived by subjects as less credible than powerful speech, and powerful speakers as more attractive than powerless speakers. The authors emphasize that the lack of association of powerless speech with femininity is particularly noteworthy when one considers that the language forms associated with female sex-typed speech are the same as those used to define powerless speech. Maxwell (1980) had college students listen to tape recorded female speakers who varied in their use of female sex-typed speech forms. Forms used in this study included tag questions, use of hostile versus non-hostile verbs, extremes in intonation, certain female-preferential adjectives, and intensifying adverbs. Although this was a pilot study in which only eleven subjects participated, some interesting trends were noted. First, it was reported that the recording with the most female sex-typed forms was rated by subjects as motherly, while the recording with no female sex-typed forms was rated as non-motherly. Secondly, a difference in ratings of organization was noted with the female sex-typed recording rated as not organized, and the recording without female sex-typed forms rated as organized. Using a different methodology, Kramer (1977) had subjects complete an 11-point Likert scale for 51 different speech characteristics. Each of the characteristics was rated in terms of female or a male pole. Of the 51 characteristics, 36 significantly differentiated between females and males. Some of the characteristics perceived to be associated with male speech included items such as demanding voice, dominating speech, uses slang, sense of humor, and authoritarian speech, whereas female speech was characterized by such items as gentle speech, gossip, self-revealing speech, talks a lot, and polite speech. In a similar study Edelsky (1976b) presented sex-typed statements to a group of adult subjects and asked them to rate the statements along an adjective scale which was associated with a female and male pole. Predictably, the results indicated the sex-typed statements to be associated with the male and female traits of the scale. A few studies have been conducted to assess subjects' judgments about ideal speech styles. Kramer (1978) conducted a study involving 466 high school and college students in which subjects made ratings of their own speech, a female speaker, a male speaker, and an ideal speaker, on a set of 51 different speech characteristics. She found that the female 10 and male subjects were in close agreement as to the characteristics of an ideal speaker. Secondly, it was found that while the differences were not significant, there was a trend for the perceived characteristics of a male speaker to be more discrepant from the characteristic ideal speaker than for the female speaker. Kramer emphasized that results such as these call into question the assumptions implicit in activities such as assertiveness training that women need to change their speech to conform to male norms. Interestingly, this study was recently replicated, obtaining similar results from a sample of British university students (Giles, Scholes, & Young, 1983). Scott (1980) conducted an intriguing study in which three groups of subjects assigned speech characteristics to either a "competent adult female speaker," a "competent adult male speaker," or a "competent adult speaker, sex unspecified." She reported that the speech characteristics assigned to the competent male speaker differed significantly from those assigned to the competent adult. However, the difference in ratings between the competent female speaker and the competent adult speaker did not reach significance. Thus, the competent female speaker and the competent adult speaker were seen by subjects as more similar in Speech characteristics than were the competent male speaker and the competent adult speaker. Finally, it should be noted that the media contribute to the perpetuation of stereotypes of female and male speech 11 styles. For example, Kramer (1974) undertook an analysis of females' and males' statements in New Yorker Magazine cartoons. She found that females were represented as highly restricted in their use of language. Females spoke considerably less than males, in fewer places, and on fewer tapics. Female and male speakers also differed in their use of exclamations with males using swear words to a much greater extent. Finally, Kramer noted that female speakers were often depicted as using "mommy-talk," consisting of gushy words and adjectives frequently associated with female use. Television also contributes to stereotypes of female and male speech styles. For example, Katzman (1972) reported that on daytime serials males are more likely to discuss business matters and professional relationships, while females are more likely to discuss romance, and family and domestic matters. In addition, females speak less often than males in the television world. Cathey-Calvert (cited in Feshback, Dillman, & Jordan, 1979) analyzed the children's show "Sesame Street," and found that 88% of the speaking characters were male, and that male dialogue totalled approximately 32 minutes, while female dialogue was approximately five minutes. These results are in accord with the general finding that females are underrepresented in television (Feshback, Dillman, & Jordan, 1979). Summarizing the research on perceptions and stereotypes of female and male speech characteristics, it is apparent 12 that people readily assume certain characteristics are associated with one sex or the other. For example, tag questions, qualified statements, intensifying adverbs, politeness, compound requests, and correct speech are perceived as characteristic of female speech. Male speech, on the other hand, is perceived to include more swear words, slang, direct assertions, and imperatives. However, while certain speech characteristics are generally associated with female or male speakers, it is not clear that these associations coincide with subjects' perceptions of masculinity and femininity. Secondly, it appears that male and female sex-typed speech characteristics influence perceptions of the speaker. Male sex-typed speech enhances perceptions of a speaker who is intelligent, assertive, and extroverted. The findings related to perceptions of female sex-typed speech are more complex. On the one hand, female sex-typed speech is perceived as reflecting uncertainty, lacking in command, self-oriented, and compliant. On the other hand, female sex-typed speech is associated with warmth, credibility, and may be closer to perceptions of the ideal speaker than male sex-typed speech. It is possible that these differing perceptions reflect the different speech characteristics manipulated in the studies reviewed. Finally, it is important to remember that stereotypes of female and male speech styles are supported, enhanced and perpetuated by media portrayal. 13 Actual Talk: Linguistic and Conversational Patterns Since most people have clear expectations regarding male and female speech styles, an obvious question is whether or not women and men actually speak in the expected, stereotypic ways. Research studies addressing this question generally use language samples which are analyzed for gender differences in frequency of use of specific language variables or differences in speech styles. Results from these studies are varied and difficult to summarize. Indeed some studies have failed to find the expected gender differences in language. For example, Dubois and Crouch (1975) presented data from a small professional meeting which had been tape recorded. Contrary to the stereotype of womens' use of tag questions, they found that 33 tag questions were spoken, all by men. Crouch and Dubois (1977) again collaborated to examine five specific language variables which according to the literature are used more frequently by women than by men. These variables included tag questions, broken fluency, interjections, garbles, and semantically empty expressions. These particular variables were chosen because, as the authors noted, they have been used to label women's speech as deviant from men's and therefore inferior. The language samples were audiotaped from twelve 50 minute laboratory sessions of a university speech class. When analyzing the data, the authors found no significant differences in the 14 speech of females and males in the use of the five variables. Similarly, Bauman (1976) attempted to identify differential use by the sexes of both tag questions and "qualifying preparatory statements" or preface qualifiers. Preface qualifiers are phrases or statements which precede a declarative and qualify its impact, such as "I may be wrong but," or “You may not agree with me but." Since women's speech is often characterized as uncertain and qualified, it was assumed that female speakers would use tag questions and preface qualifiers more than male speakers. Bauman collected language samples in four separate settings, a graduate linguistics class, a women's discussion group, an office staff meeting, and a party with both women and men present. Unfortunately, the audiotape of the party setting was too garbled to be analyzed. In the remaining three settings, however, Bauman reported that there were no gender differences in the use of the two language variables. Johnson (1980) examined differential use by the sexes of not only tag questions, but questions in general. She analyzed the language behavior of male and female staff at monthly business meetings. She reported that the use of questions was not related to gender, but rather to the role of the speaker as either group participant or group leader. Further, speakers with different roles exhibited different types of questions, such as those used by the group leader to sustain interaction and discussion. 15 A study conducted by Silverman and Zimmer (1976) also failed to find gender differences in language use. The authors were testing Jespersen's (1922) hypothesis that women are more fluent than men due to their smaller and more "central" vocabularies. Ten females and 10 males were audiotaped for three minutes while speaking about a memorable life experience. They found no gender differences in vocabulary use, nor in speaker fluency, determined by counting filled pauses, repetitions, false starts, and unfilled speech. While the above studies failed to find gender differences in the use of language variables, a number of other studies are available which report differences in various aspects of language use. For example, in an early study Wood (1966) had subjects engage in a task in which they were required to describe a photo with enough accuracy that a listener could chose the described photo from a group of similar photos. Gender differences were reported in the descriptions offered, with males using what Wood defined as an empirical style exemplified by references to concrete physical features of the photo. Females' descriptions more often included statements which presented associations to and interpretations of the photo, which Wood labeled a creative, interpretive style. A second early study conducted by Soskin and John (1963) involved the analysis of audiotaped interaction between a husband and wife, recorded over a sixteen hour 16 period. It was reported that the wife produced more expressive and affect statements than the husband, while the husband produced more directive and informational statements. More recently, Nemeth, Endicott, and Wachtler (1976) conducted a study with college students engaging in jury deliberations for mock trial settings. They found that male members of the group contributed more suggestions, opinions, and information than did the females, while the females' comments included a higher percentage of agreement than did the males'. Similarly, Aries (1982) reports that in mixed- sex discussion groups males exhibited significantly more task behavior such as giving opinions and information, while females exhibited expressive behaviors such as reactions to, and support of, male verbalizations. In a related study, Piliavin and Martin (1978) used Bales' categories to analyze the behavior of mixed and same-sex groups. Subjects were college students who participated in four person discussion groups, half of which were same-sex and half mixed-sex. It was reported that same-sex groups behaved in accordance with the authors' expectations. Females'behavior was rated as higher in dramatization, laughter, asking for help, and withdrawal. Males' behavior was rated as higher in presentation of opinions, information, and by disagreements. The authors hypothesized that in mixed-sex groups the sex-typed behaviors of females and males would be accentuated. In 17 contrast to predictions, male and female behavior tended to become less sex-typed in the mixed-sex groups. However, it should be noted that the significant differences found in the same-sex groups were still apparent in the mixed-sex groups, although of lower magnitude. Other researchers have examined gender differences in use of specific vocabulary. For example, Gleser, Gottschalk, and John (1959) collected five minute speech samples from 90 adult subjects. The words used by the subjects were classified according to emotive, cognitive, and perceptive processes. The authors found significant differences in the uses of these categories of words. Female subjects used more words implying feeling and words related to the self. Male subjects, on the other hand, used more words referring to time, space, quantity, and destructive action. It should be noted, however, that the task required of the subjects may have influenced these results. Subjects were asked to report on a memorable life experience. It is likely that women and men talked about different types of experiences, resulting in the use of different vocabularies. Gilley and Summers (1970) looked at the use of hostile verbs in relation to gender differences. Subjects were 100 undergraduates from introductory psychology classes. The procedure involved presenting each subject with a series of cards showing a pronoun and two verbs, one of which was hostile and one neutral. Subjects were to make up a sentence using the given pronoun and one of the verbs. 18 It was found that the males chose to use the hostile verb rather than the neutral verb significantly more often than did the females. The authors concluded that men are less inhibited in expressing hostility than women. A number of studies have been conducted to examine Lakoff's (1975) hypotheses of certain specific language variables as characteristics of the speech of women. For example, Lapadat and Seesahai (1977) conducted an informal study which provided mixed support of Lakoff's hypotheses. They collected audiotaped samples of college students' conversations in dormitory residence halls and found that, in support of Lakoff, female students used intensifying adverbs, exaggerations, and indirect imperatives more than males. However, contrary to Lakoff's claims, it was found. that males used significantly more tag questions than females. Crosby and Nyquist (1975) conducted a series of studies examining gender differences in the use of tag questions, qualifying words or phrases, politeness phrases, and empty adjectives. In two of the three experimental settings, females used significantly more of these language variables than males. Unfortunately, Crosby and Nyquist neglected to analyze these variables separately. Consequently, one cannot know if the significant differences are a function of one of the variables alone, some variables in combination, or all variables taken together. 19 Hartman (1976) undertook a descriptive study of older adults which provided support for Lakoff's hypothesis that women use tag questions more frequently than men. Interviewing both women and men, Hartman found that women frequently ended sentences with phrases such as "see?," "do you see?," and "you know?." Hartman maintains that these phrases function in the same way as a tag question in that they are a call for validation by the speaker. She also noted that these forms made the speaker sound tentative and unsure of herself. In a study involving same and mixed-sex groups of college students in a problem solving task, McMillan, Clifton, McGrath, and Gale (1977) found firm support for some of the hypothesized differences in women's and men's speech styles. The groups were audiotaped and their language analyzed for four language variables: intensifying adverbs, modal constructions, tag questions, and imperative constructionsin question form. The results indicated that women used all four of these language forms significantly more than men. Analyzing differences between the same and mixed-sex groups, it was found that the men's speech did not change as a function of the group. The women's speech, on the other hand, showed significant differences between the two group situations. Women in the mixed-sex groups used significantly more tag questions, modal constructions, and imperative constructions in question form than did women in the same-sex groups. Women in the same-sex groups used 20 significantly more intensifying adverbs than women in the mixed-sex groups. These findings suggest that women use more qualified speech forms when interacting with men. Similarly, Hirschman (cited in Thorne and Henley, 1975) conducted a small study involving two female and two male college students. Students were audiotaped in dyadic interactions involving all possible pairs of the four students. It was reported that females used a higher percentage of fillers, or filled speech, than males. In addition, Hirschman noted that use of fillers was influenced by the sex of the dyadic partner, as females used fewer fillers in same-sex interactions as compared to mixed-sex interactions. Swacker (1975) looked at the use of qualified speech by subjects who were asked to describe a picture. She found that females tended to qualify their speech when talking about numbers. For example, where a male subject would say "There are six books," a female subject would be more likely to say "There are about six books," thus qualifying the description. Female speech has also been stereotyped as avoiding strong expletives (Lakoff, 1971). In a review of the literature on this tapic Jay (1980) presents empirical support for this stereotype by citing numerous studies which find that men use stronger expletives than women, and use them more often. However, two interesting studies point to the importance of both setting and age 21 for these findings. Oliver and Rubin (1975) conducted a study with women ranging in age from 40 to 55 years. Subjects were given a list of expletives along with descriptions of various settings, and asked to report if they would use the expletive under those conditions. They found that, in general, the more formal the setting, the less likely subjects were to report using an expletive. Baily and Timm (1976) conducted a cross-sectional study of women's and men's use of expletives. The women in the study ranged in age from 17 through 56 years, while the men were aged 19 through 61 years. In general, the results indicated that women used fewer strong expletives than men. However, it was noted that women from ages 31 to 35 years used more strong expletives than any of the other groups, including the age divided groups of men. On the other hand, women in the other age groups used fewer strong expletives than the men. The authors speculated that the greater use of expletives by women aged 31 to 35 may reflect a more relaxed, less traditional attitude about sex roles. Finally, other researchers have looked at gender differences in listening behavior. For example, Coser (1960) reports on behavior at staff meetings. She notes that, as listeners, women laugh more often than men. Indeed of all the witticisms observed, male speakers accounted for 99 out of a total of 103, but as Coser points out."The women often laughed harder" (p. 85). Others have 22 looked at listening responses, such as "mm hmm," "yeah," or "right," which are assumed to be indicative of the listener's attention. Both Dittman (1972) and Hirschman (cited in Thorne and Henley, 1975) report females using these responses more than males. In addition, Hirschman noted that females tend to use listener responses more in same-sex interactions than in mixed—sex interactions. Summary: Language samples. As noted earlier, the studies which examine language samples in attempts to isolate gender differences have yielded complex and varied results. Part of this complexity may result from the fact that researchers have used varied subject populations, diverse settings, and a variety of different dependent measures. In so doing, variables which may interact with gender have been either uncontrolled or ignored. For example, few researchers have yet to examine how variables such as social class, race, age, setting, or topic of conversation may interact with an individual's use of sex-typed speech. With these cautions in mind, however, some tentative conclusions can be drawn. First, males appear to be more task oriented in their speech, to express greater hostility, and to some extent may use stronger expletives than females. Females, on the other hand, appear to use an expressive speech style, include more references to the self, and to use intensifying adverbs, filled speech, and indirect imperatives. There is also limited data to suggest that females' speech may be more qualified than males' speech. 23 As listeners, females tend to laugh and give more listening responses than males. The data regarding tag questions show the most varied results, with three studies finding greater use of tag questions by females, and five studies reporting either no gender differences or greater use by males. Finally, the influence of the sex of the inter- actants on speech styles cannot be overlooked, with two studies indicating that females' speech tend to become more stereotypic in mixed-sex groups, and one study reporting the opposite results. Conversational Patterns. While the above studies have focused on use of language and specific language forms, other researchers have concentrated on variables associated with conversational patterns. These variables include, for example, speaking time, interruptions, and tepic control. Contrary to the common stereotype of the talkative female, there are a number of studies assessing gender differences in speaking time in a variety of settings which find males to be more verbose than females. An early study by Wood (1966) found that when subjects were engaged in a description task, males produced significantly more speech than females. Similarly, Soskin and John (1963), who recorded the verbal behavior of a husband and wife, report that in most situations the husband produced more speech than the wife. 24 In a related study involving couples, Heiss (1962) analyzed interactions between couples who varied in their degree of intimacy from casual dating to committed relationships. Each couple was asked to discuss items of disagreement drawn from a revealed differences task. Heiss reported that males generally spoke more than females, although he noted that with the more intimate couples differences in speech production decreased. More recently, Swacker (1975) conducted a study in which male and female college students were asked to describe artistic drawings. She reported that the 17 male subjects spoke, on the average, for 13 minutes, while the 17 female subjects spoke, on the average, for three minutes. However, as the author pointed out, these results are not entirely accurate since three of the male subjects continued speaking through the end of the 30 minute cassette tape. There is no way of knowing how much longer these subjects would have continued talking had the experimental sessions not been ended. Importantly, when these three high scores were eliminated from the analysis, the results were not significantly affected. A related finding was reported by Eakins and Eakins (1976) concerning verbal behavior at seven different faculty meetings at a southwestern university. They found that males spoke more per turn, took a greater number of speaking turns, interrupted more frequently, and were interrupted less frequently than females. Similarly, Swacker (1976) 25 analyzed question-answer sessions at professional meetings, and found that men's questions were, on the average 52 seconds, while women's questions were, on the average, 23 seconds. In two other studies involving mixed-sex groups, gender differences in participation were noted. Nemeth, Endicott, and Wachtler (1976) in a study involving college students in mock jury deliberations found that not only did males participate more than females, but that males directed their comments toward other males more often than toward females. Aries (1976) compared the verbal behavior of mixed and same-sex discussion groups, and reported that in mixed-sex groups females initiated only 34% of the interaction. In an interesting study on subjects' perceptions of mixed-sex dyadic interaction, Hilpert, Kramer, and Clark (1975) report that women generally perceived the men to be more talkative. When asked to rate the more talkative member of the dyad, women chose their men partners 72% of the time. Men selected themselves as the more talkative partner only 58% of the time. The authors point out that the women's perceptions were closer to reality than the men's, as men generally spoke more than women. However, it should be noted that the greater dominance of males in speaking time is not always a consistent finding. In a study of university committee meetings, Edelsky (1981) identified two types of interaction. The first consisted of a single speaker holding the floor while others 26 listened. The second consisted of simultaneous and fragmented talk by several speakers together. Edelsky reported that in the first type of interaction, males took more turns and longer turns than females. In the second type of interaction, there were no observed gender differences in speaking time. Further, Ickes and Barnes (1977) report that in a waiting room study with unacquainted same-sex dyads, female pairs talked more with each other than did male pairs. Hershey and Werner (1975) compared two types of couples, traditional couples and couples where the wife was associated with the Women's Liberation Movement (WLM). Twenty-eight couples were interviewed by a female experimenter using a revealed differences task. They found that in traditional couples the husbands spoke more than the wives. However, in the WLM couples, the wives produced more speech than the husband. Finally, von Raffler-Engel, Smith, and Cunningham (1978) conducted a small study comparing mixed and same-sex dyads. Subjects were college students discussng a topic of their choice for 15 minutes. They found no gender differences in speaking time. Conversational interruptions have also been examined with regard to gender differences. For example, Zimmerman and West (1975) analyzed data from tape recorded conversations between dyads in informal settings. In these conversations there were 10 female-female dyads, 10 male- male dyads, and 11 mixed-sex dyads. The transcripts of 27 these conversations revealed that in the same-sex dyads, the number of overlaps andinterruptions was fairly equally divided between the first and second speaker. However, dramatic differences appeared in the mixed-sex dyads with the males accounting for 98% of the interruptions and 100% of the overlaps. In a follow-up study West and Zimmerman (1983) gathered language samples from five mixed-sex dyads in an experimental setting. Subjects were unacquainted college students who were asked to get to know each other. Analysis of the audiotaped conversations revealed a total of 28 interruptions, 21 of which were initiated by males. Similarly, Natale, Entin, and Jaffe (1979) report that in language data from mixed and same-sex dyadic interactions, males initiated significantly more interruptions than females. Additionally, both Eakins and Eakins (1976) and McMillan et al. (1977) in the studies discussed earlier found that not only did men initiate most of the observed interruptions in an interaction, but that women were more frequently interrupted than other men. Finally, Fishman (1978) reported findings with regard to control of verbal interaction. Three married couples were audiotaped in their homes for periods ranging from four to fourteen days. The tapes were transcribed and analyzed for gender differences in verbal interaction. Fishman reports several interesting findings. First, women asked questions more frequently than men, using questions 28 to initiate interaction and to facilitate men's conversation. Men, on the other hand, produced twice as many statements as women. Second, men tended to control the tapic of conversation, frequently by failing to respond to female initiated topics or by giving only "minimal response." Third, women used more listener responses than men. Fishman concluded that with regard to attempts at conversational interaction, "women tried more often and yet succeeded less often than men. The men tried less often and seldom failed in their attempts" (p. 404). Summarizing, it is a frequent finding that when women and men interact on a verbal level, men are more in control of the situation. Men more often select the topic for conversation, tend to dominate speaking times, and frequently interrupt women when they try to speak. Development of Sex Differentiated Speech Styles Since it appears that there are differences in at least some areas of adult female and male speech styles, an important question to be addressed is how and when differences develop. One way to address this question is to look at gender differences in mothers' or fathers' speech to their children. Recent research has made it clear that mothers modify their speech styles when interacting with the language learning child (Garnica, 1977; Snow, 1972; Snow, 1977). Such modifications include simplified syntax, repetitions, slower speech, short sentences, high pitch, 29 and rising intonations. However, while mothers alter their speech to fit the language ability of their child, it is not clear whether speech is modified as a function of the child's gender. For example, Cohen and Beckwith (1976), looking at maternal language behaviors during the first nine months of life, report no differences in total talk, positive or negative talk, or face-to-face interaction as a function of the babys' gender. Similarly, Phillips (1973) found that with 18 and 28-month-old children, mothers' use of syntax complexity, amount of talking, and use of different types of words were the same for sons and daughters. Golinkoff and Ames (1979) also report parental speech directed toward 19-month-old children to be the same for both girls and boys. Finally, Fraser and Roberts (1975) examined maternal language with children ranging in age from 18 months to six years. Maternal language variables assessed included total words, mean length of utterance, grammatical complexity, and a type-token ratio, assessing use of different types of words. While main effects were noted for age of the child, the authors report that maternal speech to daughters and sons was similar. In contrast, Cherry and Lewis (1976) report that mothers of two-year-old girls talked more, asked more questions, repeated their daughters utterances more often, and used longer utterances than did mothers of two-year-old boys. Similarly, Wootton (1974) analyzed audiorecordings gathered in home settings and found that mothers' language 30 to four-year-old boys included a higher percentage of controlling statements than did mothers' speech to four-year-old girls. Speech directed toward daughters, on the other hand, contained a higher proportion of discussion statements than did speech directed toward sons. Finally, Barnes (1985) found that both mothers and fathers in thirty minute videotaped sessions with their children exhibited significantly more tag questions with their preschool daughters than preschool sons. She also noted that parents used more "test questions," which were defined as questions requiring a specific, correct response, with sons. Questions directed toward daughters, on the other hand, were more conversational in tone. The above studies all examine differences in parental speech when addressed to daughters or sons. Other studies have compared the speech styles of mothers and fathers. For example, Golinkoff and Ames (1979) report that in play situations with both parents present, fathers spoke less and took fewer conversational turns than mothers. On other variables, however, such as length of utterances, directives, questions, and repetitions, mothers' and fathers' speech to their 19-month-old children was the same. In a similar study, Stoneman and Brody (1981) compared language used by parents in dyadic play sessions with their child, and in play sessions with both parents present. They found that fathers in triadic play sessions spoke fewer utterances and took fewer turns than they did in 31 dyadic situations. Mothers, on the other hand, took the same number of turns in both play sessions. In addition, this study replicated Cherry and Lewis' (1976) work, finding that mothers spoke more to daughters than to sons. Berko-Gleason and Greif (1983) report on a series of studies which indicate that fathers are more likely than mothers to use imperatives and threats with their children. Mothers were observed to use indirect imperatives and exhibit more examples of politeness. In a related study, Malone and Guy (1982) compared mothers' and fathers' speech to their three-year-old sons. Language was assessed in the home during ten minute play sessions. They found that fathers used significantly more imperatives than mothers, while mothers used significantly more questions than fathers. Malone and Guy concluded that fathers' speech appeared to be more controlling, while mothers' speech was more child-centered. Greif (1980) conducted a study in which she examined parental use of interruptions and simultaneous speech. Subjects consisted of 16 children, ranging in age from two to five years, who were videotaped in two separate thirty minute play sessions, one with their mother and one with their father. No gender differences were reported in the children's use of interruptions or simultaneous speech with their parents. However, it was found that fathers had a tendency to both interrupt and speak simultaneously more often than mothers. Further, it was reported that both 32 mothers and fathers were more likely to interrupt and speak simultaneously with their daughters than with their sons. Similarly, West and Zimmerman (1977) report that parents are more likely to interrupt children than children their parents. In an analysis of parent-child interaction, they found that 86% of the observed interruptions were initiated by parents. Unfortunately, neither parental sex nor the sex of the child were reported in this study making it impossible to note any patterns of gender differences in interruptions. Nonetheless, West and Zimmerman point out that the finding of parental interruption exemplifies the controlling aspects of interruptive behavior. In a study with slightly older children, aged five to eight years, Noller (1980) videotaped interactions between same-sex parent-child dyads and cross-sex parent-child dyads. Four parental behaviors were assessed: watching, touching, negative talking, and total talking. It was reported that parents in cross-sex dyads engaged in significantly more negative talking than parents in same-sex dyads. A trend was noted for more parental talking within the same-sex dyads. In addition to parents, an important influence for many language learning children is likely to be preschool teachers. Cherry (1975) analyzed the speech of four female preschool teachers as they interacted with the 36 children in their preschool classrooms. She found that teachers used more attention markers in their speech to boys, but were 33 more likely to respond to girls' questions. A trend was noted for teachers to use more imperatives with boys. Finally, Cherry reported that the length of the verbal interaction was not influenced by the child's sex. As noted, Berko-Gleason and Grief (1983) reported differences in mothers' and fathers' speech to their young children. Interestingly, however, they found few differences in the speech of female and male preschool teachers. Female and male teachers produced utterances of approximately the same length, and produced essentially similar repetition rates. There was a nonsignificant trend for male teachers to use more imperatives than female teachers. However, Berko-Gleason and Greif point out that when com- pared to parental speech, male teachers produced fewer imperatives than mothers. Using more global units of measurement, Serbin and O'Leary (1979) report differences in preschool teachers' behavior with girls and boys. Studying fifteen different classrooms, they found that teachers repsond rapidly and loudly to boys' transgressions, while girls are disciplined quietly and away from the groups' attention. Further, teachers' verbalizations to girls depend on the child's location, since girls were most often spoken to when physically close to the teacher. Verbalizations to boys, on the other hand, are independent of the child's location. These results are similar to those reported by Alfgren, Aries, and Olver (1979) who found preschool teachers giving 34 more attention to boys than to girls. As Serbin and O'Leary (1979) point out, it is possible to speculate that differential attention from teachers leads children and adults to view boys' behavior as more interesting and important than girls'. Further Serbin and O'Leary speculate that teachers' verbalization patterns enhance dependency behaviors in girls and independent behaviors in boys. Summarizing the research on adult language with infants and young children, it appears that the language learning environment of boys and girls is similar. ”However, some evidence indicates that tag questions and interruptions are more often directed toward girls, while imperatives are more often directed toward boys. Children may also hear different speech styles from fathers and mothers. Compared to mothers, fathers tend to speak less, and use more imperatives, rough language, and interruptions. Mothers, on the other hand, more often use indirect imperatives, are more polite, and may talk more with their children than fathers. Finally, in the preschool setting, female and male teachers tend to exhibit more similar speech styles than parents. Yet differences are apparent in the amounts of attention directed toward girls and boys, with boys receiving more overall teacher attention. Children's language. There are a number of ways to assess gender differences in children's language. First, there are several studies which have examined children's awareness, rather than use of, sex-typed language. For 35 example, Fillmer and Haswell (1977) presented 28 written sex-typed statements to grade school children asking the children to identify the sex of the speaker. The results indicated that the children typically assigned the statements to a male or female speaker in ways consistent with adult stereotypes. Unfortunately, the results of this study were pooled over all grade levels making it impossible to ascertain whether or not there were developmental differences in the children's responses. As noted earlier, Edelsky (1976a) found developmental differences in children's identification of sex-typed statements. Interestingly, Edelsky (1976c) noted two patterns of development in children's knowledge about sex-typed speech styles. Those statements which conformed to overt rules about speech styles (e.g., "Ladies don't swear" or "Ladies are polite") were identified as sex—typed at earlier ages than were statements which exemplified less obvious differences in speech styles, such as tag questions or indirect imperatives. Additionally, Edelsky observed that the youngest children tended to use topic categories to discriminate between female and male speakers, rather than using the language variables manipulated by the researcher. Garcia-Zamor (cited in Thorne and Henley, l975) conducted a study in which eight preschool children were asked to make judgments about whether an utterance was likely to have been said by a male doll or a female doll. 36 She reported that aggressive and competitive expressions were associated with the male doll, as were expressions related to cars, bright colors, and the word "shit." Expressions associated with the female doll included tag questions, light colors, terms of endearment, and the word "drat." Garcia-Zamor noted that boys were more consistent in their assignations than were girls. Anderson (cited in Thorne, Kramarae, and Henley, 1983) reported that in role play situations children ranging in age from three to seven years portrayed mothers' speech as polite, high pitched, and qualified, while fathers' speech was portrayed as forceful, straightforward, and unqualified. These studies, taken together, indicate that children at relatively young ages begin to incorporate some of the same conceptions and stereotypes with regard to female and male speech styles as adults. As with adults, however, whether the children themselves actually use sex-typed language is another question. Some researchers addressing this question have used a methodology of audiotaping the voices of prepubertal children and asking adults to identify the sex of the child speaker. For example, Edwards, (1977) audiotaped children reading from a prose passage. Adults' accuracy in identifying the sex of the children was approximately 83%. When analyzing the criteria that adults were using to make their identifications, Edwards concluded that roughness of speech as opposed to correct pronunciation 37 was the most important factor. Those children with clear, correct pronunciation were said to be girls, while those children whose speech was more rough were said to be boys. This analysis is congruent with research (Fischer, 1958; Trudgill,1975) indicating that adult females tend to be more precise in their enunciation than adult males. In a related study, Meditch (1975) audiotaped eleven children in an interview situation and in individual free play. Adults judging the sex of these children showed an accuracy rate of approximately 80%. Meditch contends that such a high rate of accuracy in identification indicates that the children are using gender differentiated speech styles. However, since she fails to analyze the children's speech for particular sex-typed language forms, it is impossible to know if the children are using different pitches or intonations, using different styles of interaction, or using specifically different language variables. There are few studies in the literature which have actually used language samples from children to assess gender differences in the use of various different language variables. Early studies in this area focused on differences in the rate of language acquisition. These studies generally indicated that girls' language acquisition is more rapid than boys'. It has been reported that girls produce their first phonemes earlier than boys (Harms and Spiker, 1959; Irwin and Chen, 1946), acquire their first words at an earlier age (Morley, 1957), and by approximately two years 38 of age have a larger vocabulary than boys (Hogan, 1976; Nelson, 1973). Recently, however, findings of gender differences in the rate of language acquisition have been challenged. For example, Klann-Delius (1981), in a thorough review of this literature, concludes that due to conflicting results, deficient methodologies, and the lack of replications, statements about gender differences in language acquisition are unwarranted. In addition to language acquisition, a number of studies have focused on gender differences in speech production in infants and young children. For example, Moss (1967) found that three-week-old girls vocalized more to their mothers than did three-week-old boys. Cherry and Lewis (1976) reported a trend for two-year-old girls in play situations with their mothers to produce more speech than boys. Finally, Brownell and Smith (1973) analyzed the speech of four-year-old children in four conditions, with a teacher and a different number of children in each condition. They found that in all four conditions, the girls produced significantly more speech than the boys. The greater loquacity of young girls as compared to boys, however, is not a consistent finding. Pairing children ranging in age from three to five years in same and mixed-sex play dyads, Garvey and BenDebba (1974) found no gender differences in the amount of Speech produced. Muller (1975) analyzed preschool children's verbal behavior in same-sex play groups and found that boys talked 39 to each other significantly more than girls. Busk (1982) conducted a study in which children from kindergarten, third and fifth grades engaged in a description task with either a male or female adult experimenter. A trend was reported for boys at all grade levels to produce more speech than girls. Further, it was noted that the differences in speech production became greater with increasing age. Finally, Sause (1975) reported greater speech from kindergarten boys than kindergarten girls. It should be noted, however, that this study usedomly a male experimenter, which may have significantly influenced gender differences in children's speech production (Cowan, Weber, Hoddinott, and Klein, 1967). Moving to speech production with older children, Elliot (1978) discusses the difficulties he experienced in trying to elicit discussion from female students in the high school grades. Elliot noted that participation from male students was more readily forthcoming. As noted earlier, Grief (1980) reported no gender differences in children's use of interruptions or overlaps. However, she was analyzing adult-child speech. Esposito (1979) conducted a study which examined interruptions, overlaps, and silences in conversations between children. Subjects were 40 children, ranging in age from approximately three to five years, who were audiotaped in same and mixed-sex dyads during a play session. She found that in mixed-sex dyads boys initiated significantly more 4O interruptions than girls. Further, the interruption rate in mixed-sex dyads was significantly greater than in same-sex dyads. Esposito concluded that young boys interrupted girls in ways similar to adult men interrupting adult women. Haas (1981) examined the language behavior of children aged four, eight, and twelve years as they engaged in same and mixed-sex dyadic conversations. She reported that in same-sex dyads boys talked about sports and location, while girls talked about school and wishing, and made more references to identify than boys. In mixed-sex dyads, boys' language was characterized as different from girls' by sound effects, direct requests, and emphasis on the topic of sports. Girls in mixed-sex dyads exhibited significantly more laughter and compliance than boys. Unfortunately, Haas neglected to analyze age differences in the language variables studied, making it impossible to note developmental changes. Staley (1981) assessed the use of descriptive language, interpretive-emotive language, reflexive language, and hedges by children aged four, eight, twelve and sixteen years. Descriptive and interpretive-emotive language, as defined by Staley, were similar in style to Woods' (1960) empirical-creative distinction. Reflexive language was defined as language referring to the self, and hedges were defined as any words or phrases which were used to qualify the speaker's remarks. Language samples were gathered by 41 asking subject to describe a series of photos. Results were analyzed to explore both age and gender differences in the use of these language forms. Staley reported that for four-year-olds, females used significantly more interpretive-emotive and reflexive language than males. Eight-year-old males used significantly more descriptive language than eight-year-old females, but in the sixteen- year-old group this difference reversed itself. Further, there were no significant differences reported in the use of hedges. Staley concluded that there were no clear gender differences in the use of sex-typed speech. In a similar study, Busk (1982) looked at qualified speech and filled speech as used by children in kindergarten, third, and fifth grades. Two description tasks, a picture description task and a description of the child's house, were used to elicit language samples. Busk reported that, although the use of qualified speech increased with age, there were no observed gender differences. The use of filled speech, however, exhibited a complex interaction with the kindergarten girls and fifth grade girls who interacted with the male experimenter exhibiting significantly more filled speech. Busk (1983) speculated that kindergarten girls' use of filled speech may have reflected their discomfort with the novelty of interaction with an adult male in the school setting. The use of filled speech by the fifth grade girls, on the other hand, may have reflected the beginning of the use of adult sex-typed Speech patterns. 42 Using an older group of subjects, Poole (1979) compared the speech of sixteen-year-olds as they were individually interviewed by a female interviewer. Among the findings, Poole reported that females used significantly more personal pronouns, and exhibited greater fluency and fewer speech hesitations than males. However, it should be noted that this study is potentially confounded by the lack of a male interviewer. To summarize the research on children's use of sex-typed language, it is clear that adults can be relatively accurate when asked to identify the sex of an unknown child Speaker. However, it is not clear what criteria adults use to make such identifications. With regard to speech production, it may be that as children move from preschool to grade school, boys begin to talk more than girls. Preschool boys also exhibit a tendency to initiate interruptions in ways similar to adult males. The results concerning children's use of language forms associated with sex-typed speech are more difficult to summarize due to the fact that there are very few studies addressing the tOpic. Further, the few studies available have used different methodologies and assessed different dependent measures. However, some tentative conclusions can be drawn. There may be differences in children's topics of conversation, with children preferring sex-typed topics. Boys' language appears to be characterized by more direct requests, while girls make more references to the self and 43 laugh more than boys. The studies by Staley (1981) and Busk (1982) illustrate the complexities involved in assessing both age and gender effects on sex-typed speech, with Staley's data indicating a possible lessening of sex-typed speech with increasing age, and Busk's data suggesting an opposite conclusion. It is clearly premature, therefore, to draw conclusions about the development of sex-typed speech styles. Statement of the Problem As noted earlier, an important question to be addressed by researchers interested in gender differences in the use of language is how and when gender differences develop. However, as the literature review indicates, the research in this area is sparse and lacking in clarity. There are a number of possible theories as to how gender differences in language may be acquired. For example, Lakoff (l975) has speculated that due to the fact that young children's adult contacts are most often female, young children's speech should reflect a female sex-typed speech style. Further, she hypothesizes that during the grade school years, specifically by age ten, girls' and boys' language should diverge. This divergence is thought to be the result of boys dropping female sex-typed language forms from their speech, and adopting a male sex-typed speech style. However, as can be seen, results reported by both Staley (1981) and Busk (1982) fail to support this proposition. 44 An alternate approach is to explore the use of sex-typed speech during the adolescent years. As Katz (1979) points out, the behaviors that are considered sex appropriate in childhood may not be consistent with sex appropriate behaviors in later years. For example, while the game of jacks and wearing make-up are both considered feminine behaviors, they bear little relation to each other and are indeed age specific. In a similar vein, it is possible that while young children are knowledgeable about adult use of sex-typed language, they fail to incorporate such language behavior into their own speech. However, during adolescence individuals adopt a vast repertoire of new behaviors considered appropriate for their sex and future status as adults. When adopting new behaviors, it is likely that the adolescents' ideas about sex-appropriate behaviors exert a strong influence. As Parsons and Bryan (1978) point out, for the adolescent "the influence of gender-role identity on life-style includes beliefs about how one 'should' walk, talk,shake hands, eat, dress, laugh, cry, compete, work... and even think" (p. 9). Thus it is possible to speculate that adolescents will begin to use language variables associated with sex-typed speech as they attempt to talk in ways stereotypically associated with their sex. This study directly tests this possibility by examining adolescents' use of sex-typed language forms and interactional patterns. A number of language variables which have been 45 considered by the literature to be associated with one sex or the other are assessed. These variables include qualified speech, filled speech, intensifying adverbs, and questions, which are frequently said to be associated with female sex-typed speech. Also examined is the use of swear words, or obscenities, typically associated with male sex-typed speech. As noted, the research literature indicates that adult males tend to initiate more interruptions and speak more in mixed-sex interactions than adult females. This research addresses these issues, assessing a number of variables including total words, utterances, turns, and interruptions. Finally, two aspects of listening behavior which appear to be used more by females than males are laughter and supportive listening responses. Each of these behaviors is also examined. In order to sample a range of ages within adolescence, subjects from both high school and college are used. Also, due to the research suggesting that the use of sex-typed language may vary as a function of the sex of subjects' conversational partners, both mixed and same-sex dyads are used in this study. 46 Hypotheses Hypothesis 1: During adolescence gender differences in speech styles will occur. Specifically, females will use qualified speech, filled Speech, intensifying adverbs, and questions more than males. Males will use more swear words than females. Hypothesis 11: During adolescence gender differences in conversational interaction will occur. Specifically, males will tend to dominate the available speaking time and to initiate more interruptions than females in mixed-sex interactions. Hypothesis III: During adolescence gender differences in listening behavior will occur. Specifically, females will laugh and use supportive listening responses more than males. METHODS Subjects. Subjects were recruited from local area high schools and from the university subject pool. Both high school officials and the University Committee for Research Involving Human Subjects approved of the procedures for subjects in this experiment. The original design for this study called for 32 subjects from each grade level, resulting in eight mixed-sex and eight same-sex dyads per grade. However, only a small number of male high school students volunteered for the project and equal NS were not obtainable. Final participation consisted of 16 ninth graders, eight of whom were paired in female-female dyads and eight paired in female-male dyads; 12 eleventh graders, eight of whom were in female-female dyads and four in female-male dyads; 32 first year college students, eight in female-female dyads, eight in male-male dyads, and 16 in female-male dyads (Table 1). All subjects were white, monolingual, native English speakers. Table l. Number of Subjects Per Cell 9th 11th 13th F M F M M Same-sex dyads 8 O 8 0 8 8 Mixed-sex dyads 4 4 2 2 8 8 47 48 Procedure. High school students were contacted at their school, and the study explained to them. Students were allowed by school officials to participate either during their study hall periods,or before or after school. As the study was explained, each student was given an information packet containing a letter to the parent, a consent form to be signed by both parents and student, a scheduling form, and a social network questionnaire (SNQ) (Appendix A). The students were assigned to dyads after they returned the signed consent form, the scheduling form and the SNQ. In order to control for familiarity effects, no subjects were assigned to participate with each other if they indicated a "best friend" status on the SNQ for another volunteer. This restriction did not disallow participation for any subjects. College students indicated willingness to participate in the project on sign-up sheets circulated to the introductory psychology courses at the university. Students received partial course credit for their participation. Prior to scheduling to dyad pairs, subjects were contacted to see if their potential partner was known to them. In no cases did college students know each other prior to participation in the research. To gather language data, subjects were seen in either mixed or same-sex dyads. There were four experimenters, two females and two males, involved in the project. The sex of the experimenter was balanced across mixed and 49 same-sex dyads at each grade level. Subjects were told that they were to engage in a conversation in which they were to gather as much information about their partner as possible. Each subject was given a list of questions to serve as examples of the type of information they could gather. The questions were modified slightly to reflect differences in the life style of college and high school students (Appendices B and C). Additionally, subjects were told that they should feel free to ask questions of each other which were not on the sample sheet. This methodology was chosen in order to obtain a natural conversational flow. Questions were chosen by listening to pilot conversations conducted with high school students, and noting what topics teenagers talked about most. It was assumed that the questions would function as facilitators of normal conversation. Each experimental session began with the following instructions. Note that the instructions for the college students were Slightly modified from the high school instructions. High school instructions: Hi, my name is _____, and I'm from Michigan State. As you know this research is interested in the ways that people talk with each other, so today I want to tape record a conversation between the two of you. Would either of you prefer not to participate? If so, you can leave now. (pause) Okay. So that you'll have something to talk about, I'm going to ask that each of you find out as much information about your partner as possible. Here is a list of questions you might want to ask your partner. I'll 50 give you a few minutes to read them over. (pause) Of course, feel free to ask questions of your partner that are not on this list. As I mentioned, I will be tape recording the conversation. If at any time either of you is uncomfortable, want to leave, or want to have the tape turned off, that's okay. Just let me know. When you have finished talking with each other, I will ask you to fill out a questionnaire to see how much you've learned about your partner. Any questions? Okay. Now I'll start the tape and wait just outside the door. You may talk as long as you like. College instructions: Hi, my name is . This research project is about language and communication. To conduct this type of research, we need language samples, natural conversation between two people. So this evening, I want to tape record a conversation between the two of you. So that you'll have something to talk about, I'm going to ask that each of you find out as much information about your partner as possible. Here is a list of questions you might want to ask your partner. I'll give you a few minutes to read them over. (pause) Of course, feel free to ask questions of you partner that are not on this list. When you have finished talking with each other, I will ask you to fill out a questionnaire to see how much you've learned about your partner. 00 you have any questions? Before we start, read and sign this consent form. If you have any questions about this form, please ask me. As I mentioned, I will be tape recording the conversation. If at any time either of you is uncomfortable, want to leave, or want to have the the tape turned off, that's okay. Just let me know. You may talk as long as you like. The experimenter then started the tape recorder and left the area. The subjects' conversation was limited to thirty minutes due to the length of the cassette. However, subjects were free to end the conversation prior to this time if they wished. 51 When subjects finished talking, the experimenter gave each a questionnaire about their partner to complete (Appendix D). This questionnaire was included because pilot work indicated that those subjects who knew that they would be required to complete a questionnaire about their partner tended to have longer conversations. Thus, the questionnaire was merely a device used to enhance the length of the subjects' conversations. The experimenters' instructions for the questionnaire were: Now that you've had a chance to get to know each other, I would like you to fill out the following questionnaire about your partner. The experimenter read aloud the instructions on the questionnaire and made certain that subjects understood the instructions. Upon completion of the questionnaire, subjects were debriefed and thanked for their participation. Transcription. Audiotapes were transcribed by undergraduate research assistants. Training for transcription was completed by having research assistants work with previously recorded tapes and transcripts until an acceptable level of clarity and accuracy was achieved. Each completed transcript was checked for accuracy once by an undergraduate research assistant and twice by the primary investigator. The transcripts were then typed by the primary investigator and checked one final time for accuracy. Thus, prior to a transcript's availability for coding, it had undergone four complete checks for accuracy. 52 This level of scrutiny allows one to be reasonably confident of a high level of accuracy on all transcripts. Coding. A group of six undergraduate research assistants worked on the project as coders, receiving approximately twenty hours of group training on the coding scheme. Each coder was responsible for coding a total of ten of the thirty conversations, so that each conversation was double coded. In addition, the primary investigator coded all thirty conversations. Coders worked on individual transcripts and indicated on the transcript each instance of a variable of interest. When each conversation had been coded by two coders and the primary investigator, the primary investigator compared them for accuracy. The primary investigator was responsible for resolving any discrepancies between the three codings. Errors found were most often errors of omission rather than errors of interpretation. In other words, a common error was the miscount of the laughter-giggle category, or the omission of one or two fillers. Upon resolution of discrepancies among the three separate codings, it is fair to assume that the transcripts were as close to perfectly coded as possible. Dependent measures. The following variables were used to assess sex-typed speech. 1. Qualified speech referred to words or phrases indicative of uncertainty. There were four categories of qualified speech. d) 53 Qualifying words were defined as any word such as "maybe" or "probably" which appeared to reflect uncertainty on the part of the speaker. Preface qualifiers were defined as a phrase preceding an utterance which functioned to qualify that utterance. Typical preface qualifiers are phrases such as,"I may be wrong but...", or “I'm not sure about this but...." Declarative questions were defined as declarative statements which were intonated as a question. Transcribers were responsible for noting question intonation on transcripts by the use of question mark punctuation. For example, a declarative question was written, "You belong to J.A.?" Tag questions were defined as declarative statements followed by a question such as, "You are nine-years-old, aren't you?" Two types of tag questions are most commonly used. Formal tag questions, such as the one above, use a correct verb form in the tag. Informal tag questions usually employ a single word as the tag, such as,"You are nine-years- old, right?" Since both formal and informal 54 tag questions function to qualify a direct assertion, they were combined into a single category in this analysis. 2. Filled speech was defined as words or sounds indicative of hesitancy. Filled speech often functions to maintain the floor while not actively contributing to the conversation. For example, "uhm" or "err" are considered filled speech. 3. Intensifying adverbs are adverbs such as "so," "very," or "quite." Typical use of this variable is, “It is so cold today." 4. Questions were defined as the total questions used, regardless of the type or form of the question. 5. Swearing was defined as any word commonly thought of as swearing or vulgarity. In order for a word to be coded as swearing, all three coders were required to independently agree upon its placement in this category. All of the above variables were analyzed in terms of their rate of use rather than absolute frequency. This was done to correct for the fact that talkative subjects can exhibit a greater frequency of use, while still exhibiting the same rate of use as less talkative subjects. Rate of use was determined by dividing each variable by an appropriate measure of speech production. For example, qualifying words and intensifying adverbs were divided by 55 total words used, while preface qualifiers were divided by utterances. The following variables were used to assess conversational interaction. 1. The total words used by each speaker were counted. Filled speech sounds, since they occurred within a Speaker's turn, were counted as words. Supportive listening responses, since they were considered listening behavior and not speaking behavior, were not counted as words. Utterances were defined as an instance of one speaker's speech bounded either by the other speaker or by a pause of one second or more. Turns were defined as all utterances of one speaker until the other speaker speaks. Mean length of utterances was determined by dividing utterances by words. Mean length of turns was determined by dividing turns by utterances. Interruptions were defined as a violation of the speaker's turn by more than one word. Interruptions were distinguished from overlaps which are occurrences of simultaneous speech just at the end of a speaker's turn but which do not necessarily "step on" or violate the speaker's turn. The behavior. I. 56 following variables were used to assess listening Laughter was counted only when exhibited by the listener. A speaker's laughter was not counted in this category. Supportive listening was defined as words or sounds indicative of active listening. For example, "uh huh," "yeah," or "right" when uttered by the listener are considered supportive listening responses. A rate of use for supportive listening was obtained by dividing each subject's supportive listening responses by their partner's number of utterances, so that supportive listening was analyzed by rate of use rather than absolute frequency. RESULTS Due to the fact that conducting large numbers of F tests can produce spuriously significant results, alpha was set at .01. However, since this research is exploratory in nature, it appeared likely that potentially interesting results would be ignored with such a strict significance level. Thus, results ranging in significance level from above .01 to .10 have been reported as trends in the data. It is assumed that the knowledge of such trends will be of value to other researchers interested in this area of investigation. Hypothesis I: During adolescence gender differences in speech styles will occur. Speci- fically, females will use qualified speech, filled speech, intensifying adverbs, and questions more than males. Males will use more swear words than females. The dependent variables used to assess sex—typed speech were analyzed in a 2(sex) x 2(dyad) X 3(grade) analysis of variance design. There were no significant main effects or interaction effects for the use of qualifying words, preface qualifiers, declarative questions, filled speech, or swear words. The means for each of the variables assessed are presented in values of absolute frequency in Table 2. There was a significant main effect in the use of intensifying adverbs for grade (F = 5.156, p < .009) 57 58 Table 2. Frequency of Use of Sex-typed Language Variables anlifying words Sex Dyad Grade Preface qualifiers Sex Dyad Grade Declarative questions Sex Dyad Grade Tag questions Sex Dyad Grade Female 16.08 Same-sex 21.41 9th 13.00 Female .47 Same-sex .59 9th .31 Female 13.42 Same-sex 13.91 9th 10.37 Female 1.58 Same-sex 2.03 9th 1.25 Male 25.23 Mixed-sex 17.18 11th 16.17 Male .50 Mixed-sex .36 11th .42 Male 14.73 Mixed-sex 13.89 11th 12.17 Male 1.45 Mixed-sex .96 11th 2.92 13th 23.88 13th .59 13th 16.31 13th 1.16 (table continues) Filled speech Sex Dyad Grade Intensifying adverbs Sex Dyad Grade Questions Sex Dyad Grade Swear words Sex Dyad Grade 59 Female 53.29 Same-sex 65.97 9th 37.50 Female 9.95 Same—sex 10.59 9th 3.94 Female 40.61 Same-sex 43.13 9th 30.88 Female 3.53 Same-sex 4.88 9th 1.81 Male 62.45 Mixed-sex 46.00 11th 47.75 Male 10.91 Mixed-sex 9.96 11th 13.33 Male 44.27 Mixed—sex 40.61 11th 45.92 Male 4.09 Mixed-sex 2.43 11th 5.50 13th 69.56 13th 12.34 13th 46.00 13th 4.03 60 indicating that the use of intensifying adverbs increased with age (Table 3). Subsequent analysis using the Scheffe method of comparison between means indicated that both eleventh graders and college students used more intensifying adverbs than ninth graders (p < .05). There were no interaction effects for the use of intensifying adverbs. Table 3. Mean Rate of Use of Intensifying Adverbs Sex Female Male .01 .00 N.S. Dyad Same-sex Mixed-sex .00 .OO N.S. Grade 9th 11th 13th .00 .01 .01 p < .009 Analysis of the use of tag questions revealed two trends of interest. First, a trend was observed for more tag questions to be used in same-sex interactions than in mixed- sex interactions (F = 3.411, p < .071). Second, a trend was observed for grade (F = 2.955, p < .061). Inspection of the means using the Scheffe method of comparison between means indicated that eleventh graders used more tag questions than either ninth graders or college students (p < .10) (Table 4). Analysis of transcripts with a specific focus on the use of tag questions led to speculation that the trends in the use of tag questions were the function of one 61 Table 4. Mean Rate of Use for Tag Questions Sex Female Male .04 .03 N.S. Dyad Same-sex Mixed-sex .04 .02 E < .071 Grade 9th 11th 13th .03 .06 .02 p < .061 particular eleventh grade female-female dyad who used a large number of tag questions. Thus, an analysis was conducted omitting the data from this particular dyad (Table 5). While the trend for grade disappeared, the trend for dyad remained (F = 3.033, p < .088), indicating that more tag questions were used in same-sex dyads than in mixed-sex dyads. Table 5. Mean Rate of Use of Tag Questions Without 11th Grade Dyad Sex Female Male .06 .07 N.S. Dyad Same-sex Mixed-sex .08 .05 E < .088 Grade 9th 11th 13th .06 .10 .06 N.S. A further analysis of the use of tag questions by college students only revealed a two-way interaction between sex and dyad (F = 3.307, p < .080) indicating that in 62 same-sex dyads males used more tag questions than females (Table 6 and Figure 1). Table 6. Mean Rate of Use for Tag Questions by College Students, Sex by Dyad Interaction Females Same-sex dyads Mixed-sex dyads .01 .02 Males .05 .02 .05 - .04 P tag questions .03 - questions Males .02 L Females .01 A Same-sex Mixed-sex dyads dyads Figure l. Sex by dyad interaction for college students' use of tag questions. 63 There was a significant two-way interaction between sex and dyad participation in the use of questions (F = 7.006, p < .001) (Figure 2 and Table 7). An investigation of the means of this interaction using a one-way analysis of variance revealed that in mixed-sex dyads females initiated more questions than males (F = 3.906, E < .058). Further, females in mixed-sex dyads initiated more questions than females in same sex-dyads (F = 4.064, E < .051). Analyses of male subjects in same-sex dyads failed to reach significance, most likely due to the small Ns involved. i—u—I m0 ll Female questions utterance ’14 H 0" I I I I I Male 0-0 N I Same-sex Mixed-sex dyad dyad Figure 2. Sex by dyad interaction for rate of questions. Table 7. Mean Rate of Questions for Sex by Dyad Interaction Females Same-sex dyads Mixed-sex dyad .13 .17 B < .051 Males .17 .13 N.S. N.S. E< .058 64 Hypothesis 11: During adolescence gender differences in conversational interaction will occur. Specifically, males will tend to dominate the available speaking time and to initiate more interruptions than females in mixed-sex interactions. A 2(sex) X 2(dyad) X 3(grade) analysis of variance was conducted with the variables used to assess speech production. A main effect for grade (F = 4.169, p < .021) indicated that the number of words used tended to increase with age. An analysis of the mean word production for each grade using the Scheffe method of comparison between means indicated that college students used more words than ninth graders (p < .05). Also. a trend was observed for males to use more words than females (F = 3.482, E < .068) (Table 8). There were no interaction effects for word production. Table 8. Mean Number of Words Used Sex Female Male 1724.29 2278.77 2 < .068 Dyad Same-sex Mixed-sex 2041.41 1798.68 N.S. Grade 9th 11th 13th 1237.38 2081.50 .2215.00 3 < .021 Analysis of utterances revealed two trends of interest. First, more utterances were produced in same-sex dyads than in mixed-sex dyads (F = 3.976, p < .052). Second, the number of utterances tended to increase with age (F = 2.468, p < .095). An analysis of the means for each grade using 65 the Scheffe method of comparison between means indicated that college students produced more utterances than ninth graders (p < .10) (Table 9). There were no interaction effects for utterances. Table 9. Mean Number of Utterances Sex Female Male 304.45 319.55 N.S. Dyad Same-sex Mixed-sex 341.53 273.93 p < .052 Grade 9th 11th 13th 243.56 315.83 341.00 p < .095 A main effect for turns indicated that turn taking tended to increase with age (F = 4.035, p < .024). Inspection of the means using the Scheffe method for com- parison between means indicated that this effect was due to a difference in turn taking between ninth graders and college students (2 < .05) (Table 10). Table 10. Mean Number of Turns Sex Female Male 122.79 151.55 N.S. Dyad Same-sex Mixed-sex 132.03 134.82 N.S. Grade 9th 11th 13th 96.69 144.83 147.34 p < .024 66 Analysis of turns also revealed a two-way interaction between sex and dyad participation (F = 3.971, p < .052). A one-way analysis of the means of this interaction indicated that in same-sex dyadic interactions, male subjects took significantly more turns than female subjects (F = 8.252, p < .007). In mixed-sex interactions the number of turns for females and males was essentially the same (Table 11 and Figure 3). Table 11. Means for Turn Taking Interaction Females Same-sex dyads Mixed-sex dyads 115.71 134.93 N.S. Males 181.00 134.71 N.S. B < .007 N.S. 190 - 180 1- 170 r 160 b Number of 150 L Turns 140 _ Males 130 ' Females 120 - 110 - 100 - Same-sex Mixed-sex dyad dyad Figure 3. Sex by dyad interaction, turns. 67 Analysis of words per utterance yielded two significant main effects. First, a main effect for sex (F = 8.875, p < .004) indicated that males used significantly more words per utterance than females. A second main effect for grade (F = 4.588, p < .015) indicated that the number of words per utterance increased with age. An analysis of the means for grade using the Scheffe method of comparison between means indicated that college students used more words per utterance than did ninth graders (p < .05) (Table 12). There were no interaction effects in the number of words per utterance. Table 12. Mean Number of Words Per Utterance Sex Female Male 5.56 6.85 p < .004 Dyad Same-sex Mixed-sex 5.88 6.22 N.S. Grade 9th 11th 13th 5.05 6.34 6.42 p < .015 Analysis of utterances per turn resulted in a main effect for dyad, indicating a trend for more utterances per turn in same-sex dyads than in mixed-sex dyads (F = 3.533, B < .066) (Table 13). 68 Table 13. Mean Utterances Per Turn Sex Female Male 2.39 2.16 N.S. Dyad Same-sex Mixed-sex 2.52 2.07 p < .066 Grade 9th 11th 13th 2.32 2.32 2.30 N.S. There was also a significant (F = 8.345. E < .006) two-way interaction in utterances per turn between sex and dyad participation (Figure 4 and Table 14). A one-way analysis of the means of this interaction revealed that in same-sex dyadic conversations female subjects used more utterances per turn than did male subjects (F = 3.350, p < .077), while in mixed-sex dyadic conversations males used more utterances per turn than females (F = 5.028, E < .034). Further, females in mixed-sex dyads used significantly fewer utterances per turn than did females in same-sex dyads (F = 7.731, p < .01). Finally, it must be noted that analysis of interruptions was not possible. This was due to the fact that as transcribers worked, there was a large amount of drift away from the original definition of an interruption. With such lack of agreement, or low reliability, on the identification of interruptions, analysis became meaningless. 69 Table 14. Mean Utterances Per Turn Interaction Females Same-sex dyads Mixed-sex dyads 2.70 1.85 p < .01 Males 1.95 2.28 N.S. E < .077 E < .034 3.0 - 2.5 - Male utterances turn 2'0 ' Female 1.5 - 1.0 - Same-sex Mixed-sex dyad dyad Figure 4. Sex by dyad interaction, utterances per turn. Hypothesis III: During adolescence gender differences in listening behavior will occur. Specifically, females will laugh and use supportive listening responses more than males. A 2(sex) X 2(dyad) X 3(grade) analysis of variance was conducted with laughter and supportive listening responses as dependent variables. There was a main effect for sex (F = 5.515, p < .023) indicating a trend for females to 70 exhibit more laughter than males (Table 15). There were no interaction effects in laughter. Table 15. Mean Number of Occurrences of Laughter Sex Female Male 34.08 17.05 p < .023 Dyad Same-sex Mixed-sex 32.66 22.32 N.S. Grade 9th 11th 13th 29.06 35.25 24.44 N.S. There was a significant main effect indicating that supportive listening responses increased with age (F = 5.423, B < .007) (Table 16). A comparison of means using the Scheffe method of comparison between means indicated that college students used more supportive listening responses than did ninth graders (p < .05). There was also a trend to indicate that females used more supportive listening responses than males (F = 3.312, p < .075). There were no interaction effects in supportive listening responses. Table 16. Mean Rate of Supportive Listening Response Sex Female Male .07 .05 p < .075 Dyad Same-sex Mixed-sex .07 .06 N.S. Grade 9th 11th 13th .04 .06 .08 p_ < .007 DISCUSSION Sex-typed Speech As noted in the literature review, there is little data available indicating clear patterns of gender differences in young children's or preadolescents' use of sex-typed speech. There is even less data on use of sex- typed speech by adolescents. However, it is reasonable to speculate that as adolescents experiment with sex-role behaviors they perceive to be typical of adults, they will adopt what they think to be sex-typed speech styles. Therefore, Hypothesis I focused on the use of sex-typed language by adolescents. It was hypothesized that female subjects would use more qualified and filled speech, and more intensifying adverbs and questions than male subjects. Male subjects, on the other hand, were expected to use more swear words than female subjects. These expectations were not fully supported by the data. To understand the obtained results, each of the variables will be separately considered. Qualified speech. Women's speech is often stereotyped as uncertain or tentative, with this lack of certainty presumably revealed by the use of a number of different language variables. This study assessed the use of four variables which can function to lessen the impact of a 71 72 direct assertion: qualifiying words, preface qualifications, tag questions, and declarative questions. Contrary to expectations, there were no main effect gender differences in the use of any of these four variables. Explanation for the lack of gender differences can be found in an examination of the frequencies of use of these variables, , and in the ways that they were used by speakers. h Use of qualifying words by both females and males represented approximately one percent of the total words «A used. Similarly, preface qualifiers were essentially nonexistent in the subjects' speech. The tepic of conversation engaged in by subjects may be responsible for such a low frequency of use. As noted by Busk (1982), conversational topic can have a powerful influence on the presence or absence of qualified speech. In this experimental setting, the questions used by subjects as conversational tools elicited straightforward rather than qualified responses. For example, few subjects responded to their partner's questions with remarks such as, "I have about three brothers," or "I'm not sure but I think I have three brothers." There were some conversational questions, however, to which a qualified response was not improbable. For example, it would not be unreasonable to expect subjects to make statements such as "I kinda like the Rolling Stones," or "This may sound stupid, but I'd like to be a tree surgeon." However, these types of statements simply did 73 not occur. When answering questions about topics such as favorite television shows, music groups, teachers, or even questions about career plans, few subjects paused to reflect on answers or qualify their remarks. Rather, subjects responded to such questions with statements such as "My favorite show is Double Trouble," or "My favorite teacher WW is Mrs. H.". As noted, there were no gender differences in the use [Tm—M of tag questions. As with qualifiying words and preface qualifiers, the frequency of use of tag questions was small. Tag questions ranged from two to six percent of the total questions used. Indeed, there were many subjects who used no tag questions at all. However, trends in the data suggested that tag questions were used more in same-sex dyads and by eleventh grade subjects. This somewhat puzzling result led to an examination of each individual's use of tag questions. While most individuals used from zero to three tag questions, there was one eleventh grade female dyad pair which used a total of 15 tag questions, with one partner using ten tag questions and the other partner using five. While analysis of the data without this pair led to an elimination of the trend for grade, there still remained a trend indicating that same-sex dyads used more tag questions than mixed-sex dyads. Thus, an analysis of the data from college students was undertaken to examine the differences between same and mixed-sex dyads in the use 74 of tag questions. An interaction effect revealed that in same-sex dyads college males used more tag questions than college females, while in mixed-sex interactions the use of tag questions by college females and males was the same. Since this result was contrary to expectations, the transcripts were examined to determine how tag questions were used. This examination revealed that tag questions were used most often in sequences where the speaker was checking to see if the listener understood the speaker's utterances. For example, in one college male dyad, a speaker explained the Operation of the Michigan State University cyclotron while using tag questions such as, "There's this glass thing that splits the beam, right?" or "They have these different computer probes around it, right?" In each instance the speaker was not expressing his own uncertainty, but rather using the tag question to make certain that the listener understood his remarks. This use of tag questions is contrary to Lakoff's (1975) claim that tag questions are a reflection of the speaker's uncertainty. The final variable reflecting qualified speech to be analyzed was declarative questions. Contrary to the low frequencies observed for the other variables reflecting qualified speech, declarative questions accounted for 32 percent of the questions asked by female subjects, and 33 percent of the questions asked by male subjects. Examination 75 of the transcripts revealed that in most instances declarative questions were used in a question-response- declarative question-response sequence, as in the following example from a conversation between college students. Speaker A: where do you live? Speaker B: over in Brody Speaker A: you live in Brody complex? Speaker 8: yeah in Butterfield Hall This example illustrates that rather than expressing uncertainty, the declarative question was used by speakers as a probe to elicit further response. The fact that both declarative questions and tag questions were used in a manner which did not reflect qualified speech, illustrates that the assumed function of a language variable may not always be correct. Indeed, it may be that some of the variables which have been used to label women's speech as qualified instead reflect different functions. For example, it has been suggested that qualified words may reflect sensitivity or awareness of another speaker's perspective (Busk, 1982; McMillan et al., 1977), and that tag questions can be used to initiate and maintain interaction (Fishman, 1978; Johnson, 1980). Summarizing, there were no main effect gender differences in the use of any of the variables hypothesized to reflect qualified speech. In same-sex dyads college males used more tag questions than college females, while one particular eleventh grade pair appeared to be responsible for the 76 observed grade effect. The subjects' use of both tag questions and declarative questions calls into question the assumed function of these language variables. College males used tag questions to check the listener's understanding, while declarative questions were used by both female and male subjects, not to qualify assertions, but rather to elicit further elaboration from a conversational partner. Filled speech. Filled speech is thought to indicate a speaker's hesitancy and maintain a conversational turn while the speaker does not actively contribute to the conversation. As women's speech is commonly stereotyped as hesitant and lacking in assertion, it was hypothesized that female subjects would use more filled speech than male subjects. However, there were no gender differences in the use of filled speech. Again, as with qualified speech, an explanation for this result may involve the frequency of use of filled speech. For both female and male subjects, filled speech represented only three percent of their total speech. Such a low frequency of filled speech is indicative of the fact that there was little in this conversational interaction to elicit filled speech. As noted, few speakers paused to reflect on answers prior to speaking, and thus filled speech was generally not used. Intensifyingadverbs. It was hypothesized that female subjects would use intensifying adverbs more than male subjects. Stereotypically this language variable is 77 consistently associated with female use, and two empirical studies indicate greater actual use of intensifying adverbs by females. There were, however, no gender differences in the use of intensifying adverbs. Rather, a significant main effect for grade level showed that eleventh grade students and college students used intensifying adverbs more than ninth graders. However, as the means for this variable show, the actual use of intensifying adverbs was very low, so that while significant, this result cannot be considered highly meaningful. Thus, it is most prudent to conclude that intensifying adverbs were simply not used by subjects in this experimental setting. Questions. There were gender differences in the use of questions. An interaction effect revealed that in same-sex dyads, the college male pairs asked more questions than female pairs. In mixed-sex dyads, however, females asked more questions of their partners than did males. This pattern highlights interesting differences in the interaction styles of same and mixed-sex dyad pairs. In same-sex pairs, the greater use of questions by the college male subjects is indicative of an interaction which was characterized by question-response sequences. The female pairs, on the other hand, appeared to reject the question- answer style of interaction and adopt a more conversational style. Looking at transcripts of female pairs, there were many instances where one speaker's comments appeared to build upon or flow from the previous speaker. The following 78 is an example in which two female college students are discussing summer plans. This example has been edited slightly to make it both shorter and more readable. Speaker A: Speaker Speaker Speaker Speaker Speaker Speaker Speaker Speaker Speaker A: B: going to stay here and take about// 12 creditS/lget a job/II'll be living in Cedar Villagel/and hopefully I'll have time to have fun go to school in the summer//I couldn't do that I I want to get out fast yeah you sound like escape that will be/lweirdl/I can't wait till this term's out I know/lthat's what I'm thinking once I get towards the end of the term I'll be out/II'm not staying here this summer if they're easy classes though it shouldn't be bad summer it's not gonna be a math class oh that's good AS noted, this type of conversational interaction was not as readily apparent in the conversations between college males. Also interesting is the finding that in mixed-sex dyads, females initiated more questions than males. This reflects an interaction in which female subjects are asking questions, and male subjects answering. This finding mirrors that of Fishman's (1978) in which females in mixed-sex interactions do the "work" of maintaining the conversational interaction. 79 Swear words. It was hypothesized that males would use more swear words than females. However, there were no observed gender differences in the use of swear words. This finding is undoubtedly a function of both the experimental setting and the topic of conversation. First, most students learn at relatively young ages that swearing is inappropriate language in the classroom. While subjects were not specifically in a classroom for the experimental interaction, an academic environment still surrounded them. It is likely that this environment suppressed the use of swear words. In addition, it is likely that the formality of the experiment itself, along with the tape recorder, further inhibited swearing. Second, swear words are used most often to express strong emotion, such as anger or disappointment. It was rare for these conversations to elicit such emotion, further suppressing the use of swear words. It needs to be pointed out, however, that some swearing was used. Female subjects used an average of 3.5 swear words, while male subjects used an average of 4 swear words. Further, an interesting nonsignificant trend was observed for same-sex dyads to use twice as many swear words as mixed-sex dyads, suggesting that the presence of an opposite sex partner suppressed swearing of both males and females. Summarizing, few of the expected gender differences in the use of sex-typed language were found in this data. 80 This result has been partially discussed by examining the frequency and function of the various different language forms. However, the finding of no gender differences warrants further discussion. First of all, as Thorne (1983) points out, gender may be of less importance in some situations than in others. In this particular situation, a research experiment, the role of "subject" may have superceded female and male roles, and thus mitigated potential gender differences (Eagly, 1983). In line with thisreasoning, Trudgill (1983) states that "the larger and more inflexible the difference between the social roles of men and women in a particular community, the larger and more rigid the linguistic differences tend to be" (p. 88). This being the case, it is possible to speculate that for these subjects, the role of student is a more salient role than that of female or male. Thus, as a group of high school and college students, rather than as a group of females and males, few gender differences were exhibited in their speech. An alternate possibility is that today's adolescents are less sex-typed than in previous years (Staley, 1982). While sex-typed language is only one aspect of sex-role behavior, the female subjects in this study were neither uncertain, tentative, nor hesitant in their speech styles. This failure to conform to stereotypic female speech styles may reflect a lessening of sex-role proscriptions in today's adolescents. Indeed, Trudgill (1983) has observed that 81 the "beginning of a move away from sex-role stereotyping probably explain the fact that linguistic differences between younger men and women now appear to be smaller than in the case of older people" (p. 95). On the other hand, it is interesting to speculate that the language of today's adolescents may be as sex~typed as "older speakers," but that the specific language forms used by female or male Speakers are changing. For example, Lakoff (1975) and others (Jespersen, 1922; Key, 1975; Reik, 1954) have claimed that females use specific adjectives such as "divine" or "lovely.'I However, it is rare to hear this type of vocabulary used by younger speakers. In this study, neither female nor male Speakers used many intensifying adverbs, which have often been associated with adult female speakers. Thus, an interesting research question to be addressed in the future is what, if any, speech forms are gender Specific for adolescents, and how these forms may be different from the sex-typed speech used by adults. Finally, it must be considered that the failure to find gender differences may have resulted from the small proportion of male subjects participating in the study. As noted earlier, attempts were made to provide equal numbers of male and female subjects. However, there was an extremely low rate of participation by high school males, with only six individuals volunteering to participate. It is possible that this lack of male participation suppressed the emergence of gender differences in the use of language. 82 Speech Production and Interruptions As noted in the introduction, the research literature on speech production with young children reveals mixed results, with some studies showing girls to be more talkative than boys, and some studies showing the reverse. The literature involving adults, on the other hand, indicates that in many instances males tend to be more loquacious than females. Further, most of the available literature indicates that male speakers tend to initiate more interruptions than female speakers. Hypothesis II dealt with gender differences in the use of available speaking time and interruptions in adolescents' conversations. It was hypothesized that male subjects would initiate more interruptions and speak more than female subjects, and that this difference would be most pronounced in mixed-sex interactions. A number of different variables were used to assess speech production as a function of age, gender, and dyad participation. First, it was clear that grade had an important impact on speech production. College students used more words and utterances, and took more turns in these conversational interactions than did ninth graders. In addition, as measured by the mean length of utterance, college students' utterances were typically longer than ninth graders'. Eleventh grade subjects occupied a middle ground, typically being significantly different neither from the ninth graders nor the college students. 83 This result apparently reflects the greater ease experienced by college students in talking with an unfamiliar peer. The college students participating in this study were all first year students, who over the course of the academic year had experienced many opportunities for interactions similar to the experimental setting. High school students, on the other hand, were not likely to be as practiced in this type of interaction. Support for the hypothesis that males would assume more of the available speaking time was evidenced by the finding that males used more words than females, and that males' utterances were significantly longer than females' utterances. Interaction effects indicated that in same-sex dyads, college male pairs took significantly more turns than did female pairs. This result is clarified by looking at subjects' use of utterances per turn. An interaction effect for this variable indicated that in same-sex dyads, female pairs used more utterances per turn than male pairs. This finding, together with the finding that college male dyad pairs took more turns than female pairs, shows that while female pairs tend to take fewer turns than male pairs, their turns tend to be longer. In mixed- sex dyads, on the other hand, female subjects' turns tend to become shorter, while male subjects' turns become longer. This pattern of turn taking is most likely the result of subjects' use of questions. As noted earlier, 84 female dyad pairs asked fewer questions than male pairs, while females in mixed-sex interactions asked more questions than did males in mixed-sex interactions. Often, asking a question of one's partner results in a relatively short turn, frequently of only one utterance. In same-sex dyads, college males exhibited a question- response interaction style which resulted in many turns per speaker. Female pairs, on the other hand, asked fewer questions of each other and consequently had longer turns. In mixed-sex dyads, this pattern of turn taking and questions reversed itself for female subjects. In these dyads, female subjects asked the questions, while male subjects answered. Thus, the females had short "question" turns, while the males had longer "response" turns. As noted, these findings suggest that in mixed-sex interactions, it is the female who often assumes the responsibility for maintaining the conversation (Fishman, 1978). It was reported in the results section that interruptions were not analyzed due to lack of agreement by transcribers on the definition of interruptions. This occurrence needs further discussion. As noted, transcribers received considerable training in the mechanics of transcription, which included learning to recognize interruptions based on the research definition. After training, it appeared that transcribers understood the definition and were working accordingly. However, as 85 they began working, there slowly occurred considerable drift away from the research definition of interruptions. Examination of this phenomenon revealed that transcribers began to consider interruptions in a more subjective sense than the operational definition. For example, if a speaker spoke within milliseconds of a partner's turn and at the same time failed to acknowledge the partner's completed turn, the transcribers would often record an interruption. This is illustrated in the following example from a conversation between high school students. Speaker A: 'we have a dog//this is the first pet that I remember having/II remember hearing stories from my parents Speaker 8: I'm going to soccer practice after school Note that Speaker 8 did not violate Speaker A's turn by speaking before she was finished. However, Speaker 8 responded quickly and with little regard for Speaker A's statement. This type of interaction was often considered an interruption by the transcribers. On the other hand, transcribers often failed to consider overlaps of as many as five words to be interruptions, if the overlap acknowledged the first speaker's turn. This drift away from the research definition of interruptions to a more subjective "feeling" definition of interruptions occurred with the majority of the transcribers. This behavior calls into question the meaningfulness of the research definition and suggests 86 that a closer look at the "felt" sense of interruptions is warranted. This is clearly a question which needs to be addressed in future research in interruptions. Summarizing the findings with regard to Hypothesis 11, college students' conversations were considerably longer than high school students', most likely reflecting the greater ease of college students with the task of talking to an unfamiliar peer. Second, results supported the hypothesis that males would tend to monopolize the available speaking time, with males using more words, and longer utterances than females. Third, the results regarding turn taking and length of turns clearly mirrors the question-answer patterns characteristic of the different dyad types. Finally, there may be reason to question the meaningfulness of the traditional research definition of interruptions. Listening Behaviors As McGhee (1979) points out, there is an expectation in American culture for males to be the ones to initiate humor, while females are expected to respond. Consequently, it was hypothesized that females would exhibit more laughter than males. This hypothesis was supported by the data with female subjects exhibiting twice as much laughter as male Subjects. However, before concluding that female subjects were laughing in response to male initiated humor, it is instructive to examine the overall means obtained for this variable. 87 Inspection of the means reveals that there was more laughter in same-sex dyads than in mixed-sex dyads. This being the case, it is not prudent to assume that female subjects in this study were laughing in response to male witticisms. Rather, it may be that laughter was functioning as an expression of fellowship between like pairs. A more detailed analysis and coding of the transcripts would be necessary to test such a hypothesis. However, it is interesting to note that while conducting this research, the experimenters commented that female dyads appeared to have more "fun" with the experiment than male dyads. In fact, the experimenters observed that several of the college women exchanged telephone numbers in anticipation of seeing each other again. This expression of friendship was not observed with the male dyads. Such behavior further exemplifies that for male subjects the experiment was a task to be completed, while for female subjects the experiment was an interesting affective experience. Hypothesis III also predicted that female subjects would exhibit more supportive listening responses than male subjects. This prediction was partially supported by the data with a trend indicating that females used more supportive listening responses than males. However, the variable most predictive of supportive listening was age, with college students exhibiting more supportive listening than high school students. This finding is in 88 accordance with Dittman's (1972) research in which he .found significant age differences in supportive listening responses. Conclusion There are several major points to be emphasized from this study. First, while the finding of no significant differences is frequently considered unimportant in psychological research, in gender difference research the finding of no differences is highly interesting. In this particular case, it suggests that female adolescents' language is neither qualified nor hesitant. As noted, such a finding may reflect a relaxation of traditional sex- role standards by today's adolescents. It further illustrates that there may be more similarities than differences in adolescents' use of language. Second, while gender differences in the use of specific language variables were not readily evident, there were gender differences in the structure and style of interaction. Female subjects in this study took it upon themselves to maintain conversation with males by shortening their turns and asking more questions. Differences between female and male dyads further indicated differences in interaction styles. Males appeared more concerned with adhering to the experimental task, and maintained an interview "question- response" interaction. Females, on the other hand, exhibited a conversational style, which was more relaxed and less 89 task oriented than males' interactions. This finding mirrors the instrumental-expressive distinction frequently reported in studies of female and male interaction, and suggests that females and males may approach the activity of "talk" with different intentions and objectives. It is possible that males perceive talk to be a means to an end, while females perceive talk to be an end in itself. Third, it is important to remember when considering the differences between same and mixed-sex dyads, that there were no male-male dyads at the high school level. Thus, conclusions about same-sex male dyads are based on data from college students only. Clearly further research with younger adolescents is needed. While the literature on use of sex-typed language and interaction patterns has been rapidly expanding in the past ten years, little attention has yet been directed toward children or adolescents. The findings from this study suggest many research questions in need of exploration. HOpefully, as researchers interested in language and gender begin to turn their attention to the many variables, including age, which may interact with gender, some of these questions will be addressed. APPENDICES APPENDIX A 90 Appendix A MICHIGAN STATE UNIVERSITY Department of Psychology East Lansing, MI 48824 Psychology Research Building Dear Student and Parents, I am a graduate student in the Deve10pmental Psychology program at Michigan State University and am currently conducting research for my doctoral dissertation on individual differences in the use of language. I am contacting you to ask that you and your parents consider your participation in my current research project. Should you participate, you will be asked to engage in a conversation with a student from your school. To provide you with something to talk about, you will be asked to find out specific information about your conversational partner. Since your partner will also be trying to find out about you, you may be asked questions about such things as school classes, TV shows, or extracurricular activities. We use this conversational method in order to obtain language samples which we later analyze for differences in linguistic variables. Therefore, we will be recording the conversation. The conversational interaction will take place either during the school day or directly after school. We will attempt to arrange schedules so that students do not miss academic classes. As a student, I think you will find this to be a fun experience. It will give you the chance to talk with someone from your school who you don't yet know well, and find out about a new person. If you would like to be able to participate in this research, both student and parent need to sign the enclosed consent form. The student will need to fill out the Social Network Questionnaire, which we will be using to make decisions about each student's conversational partner. Also, we will need information about your schedule on the Schedule Information form. Please return the consent form, the Social Network Questionnaire, and the Schedule Information form to your school. Thank you for your cooperation and participation. Sincerely, Kristine Busk 91 Michigan State University Department of Psychology Consent Form 1. I have freely consented to take part in a scientific study being conducted by Kristine Busk, under the supervision of Dr. Elaine Donelson. 2. The study has been explained to me and I understand the explanation that has been given and what my participation will involve. 3. I understand that I am free to discontinue participation at any time without penalty. 4. I understand that the results of the study will be treated in strict confidence and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. 5. I understand that participation in the study does not guarantee any beneficial results to me. 6. I understand that, at my request, I can receive additional explanation of the study after my participation is completed. Signature of student Date Signature of parent Date Signature of parent Student‘s school Student's grade 92 Schedule Information If you want to participate in this research, please help us with scheduling by providing the following information. 1. Do you have a study period during which you could participate? What time does it meet? 2. Would you be willing to participate during your lunch period? What time would that be? 3. What days would you be able to stay after school (8 hour to an hour) to participate in the research? 4. Are there any days of the week when you would be able to participate before school begins? Which days? We will call you as soon as our schedule is complete to let you know when your participation will be. Name Grade Phone Number 93 Name School Grade Social Network Questionnaire Instructions: To assign each of you a conversation partner, we need to know about your friends and social relationships. We will try to pair you with someone you don't know very well, and someone you would not be uncomfortable talking to. When naming people, give both first and last names, or last initial, if known. If you cannot think of names for a particular question, leave spaces blank. Your responses will be confidential. 1. Whom do you consider to be your one best friend of the same sex in your grade at school? 2. Whom do you consider to be your one best friend of the other sex in your grade at school? 3. Name three other people of each sex in your grade at school whom you know well. Girls Bgy§_ 1. 1. 2. 2. 3. 3 4. Name two people of each sex in your grade with whom you spend the most time outside of school. Girls Boys I. . 2. 2. 5. Name two people of each sex in your grade with whom you spend the most time at school. Girls Boys 1. . 2. 2. APPENDIX B 94 Appendix B Conversational Questions - High School Here are some questions you might want to consider asking your partner during your conversation. Of course, feel free to ask questions that are not included on this list. What is your favorite class? Why do you like that class? What is your least favorite class? What don't you like about that class? 00 you belong to any school clubs? What are they? How much TV do you usually watch, per day or per week? What is your favorite TV Show? What do you like about that Show? Where do you live? How many brothers and sisters do you have? Are they older or younger than you? When's your birthday? 00 you have any pets? What kind? Do you take care of them? Do you date anybody? Do you have any regular jobs around the house? 00 you get an allowance? Have you thought about what kind of job, or career, you want to have? What kind of sports do you like? What do you usually do after school? What do you usually do on weekends? APPENDIX C 95 Appendix C Conversational Questions - College Here are some questions you might want to consider asking your partner during your conversation. Of course, feel free to ask questions that are not included on this list. What's your major? What type of job do you want after school? Where do you live (campus)? What was your high school like? What are your favorite classes? What did you do over spring break? What kind of music do you like? What movies have you seen lately? Do you have a job? Do you get along with your roommates? What is your favorite TV Show? What do you like about that Show? How many brothers and sisters do you have? Are they older or younger than you? What kind of sports do you like? What do you usually do on weekends? What is your favorite restaurant or bar? What kind of books or magazines do you like to read? Are you involved in any clubs or school organizations? What are you going to do this summer? APPENDIX D 97 Post Conversation Questionnaire College Instructions: This questionnaire is designed to find out how much you learned about your partner in your conversation. Answer the following questions as best you can. In answering these questions assume that you knew nothing about your partner prior to today's conversation. Answer these questions only on the basis of what you learned today, and not on any prior knowledge. For example, if you happen to know that your partner has two brothers, but neglected to ask about that in your conversation, you would not be able to answer question 3. 1. What is your partner's name? What is your partner's favorite class? DON Does your partner have any brothers or sisters? 4. What is your partner's favorite TV Show? 5. Is your partner involved in any clubs or school organizations? 6. What is your partner going to do this summer? 7. Where does your partner live? 8. Does your partner get along with his/her roommates? 9. What kind of job, or career, does your partner want to have? 10. Does your partner have a job? 11. What kind of sports does your partner like? 12. What other information did you learn about your partner? REFERENCES REFERENCES Alfgren, S.H., Aries, E.J., & Olver, R.R. (1979). Sex differences in the interaction of adults and preschool children. Psychological Reports, 44, 115-118. Aries, E. (1976). Interaction patterns and themes of male, female and mixed groups. Small Group Behavior, 1, 7-18. Aries, E.J. (1982). Verbal and nonverbal behavior in Single-sex and mixed-sex groups: Are traditional sex roles changing? Psychological Reports, 51, 127-134. Bailey, L.A., & Timm, L.A. (1976). More on women's - and men's - expletives. Anthropological Lipguistics, 48, 438-449. Barnes, C.L. (1985). Comparison of mothers' and fathers' speech to their preschool children. Unpublished doctoral dissertation, Michigan State University, East Lansing, MI. Bauman, M. (1976). Two features of "women's speech?". In B.L. Dubois and I.M. Crouch (Eds.), The sociology of the language of American women. San Antonio, TX: Trinity University Press. Berko-Gleason, J., 3 Greif, E.B. (1983). Men's speech to young children. In B. Thorne, C. Kramarae, and N. Henley (Eds.), Language, gender and society (pp. 140-150). Rowley, MA: Newbury House Publishers. Bernard, J. (1972). The sexpgame. New York: Atheneum. Berryman, C.L. (1980). Attitudes toward male and female sex- appropriate and sex-inapprOpriate language. In C.L. Berryman and V.A. Eman (Eds.), Communication, language, and sex: Proceedings of the First Annual Conference Ipp.ll95-216). Rowley, MA: Newbury House Publishers. Berryman, C.L., & Wilcox, J.R. (1980). Attitudes toward male and female speech: Experiments on the effects of sex-typical language. The Western Journal of Speech Communication, 44, 50-59. Brownell, W., 8 Smith, D.R. (1973). Communication patterns, sex, and length of verbalization in speech of four-year-old children. Speech Monographs, 49, 310-316. 98 99 Busk, K. (1982). Use of female sex-marked speech as a function of sex, age, experimenter's sex, and situational content. Unpublished masters thesis, Michigan State University, East Lansing, MI. Busk, K. (1983, April). Children's use of sex-marked speech. Paper presented at the 11th Annual Michigan Womenrs Studies Association Conference, Kalamazoo, MI. Cherry, L. (1975). Teacher-child verbal interaction: An approach to the study of sex differences. In B. Thorne and N. Henley (Eds.), Language and sex: Difference and dominance (pp. 172-183). Rowley, MA: Newbury HousegPublishers. Cherry, L., & Lewis, M. (1976). Mothers and two-year-olds: A study of sex-differentiated aspects of verbal interaction. Developmental Psychology, 12, 278-282. Cohen, S.E., & Beckwith, L. (1976). Maternal language in infancy. Developmental Psychology, 12, 371-372. Coser, R.L. (1960). Laughter among colleagues: A study of the social functions of humor among the staff of a mental hospital. Psychiatry, 22, 81-95. Cowan, P.A., Weber, J., Hoddinott, B.A., & Klein, T. (1967). Mean length of a spoken response as a function of stimulus, experimenter, and subject. Child Development, 22, 191-203. Crosby, F., & Nyquist, L. (1975). The female register: An empirical study of Lakoff's hypotheses. Language in Society, 6, 313-322. Crouch, I.M., & Dubois, B.L. (1977). Interpersonal comnunication in the classroom: Which sex's speech is inferior? Journal of the Linguistics Association of the Southwest, 2, 129-141. Dittmann, A.T. (1972). Developmental factors in conversational behavior. Journal of Communication, 22, 404-423. Dubois, B.L., & Crouch, I.M. (1975). The question of tag questions in women's speech: They don't really use more of them, do they? Language in Society, 4, 289-294. Eagly, A.E. (1983). Gender and social influence: A social psychological analysis. American Psychologist, 22, 971-981. Eakins, B., & Eakins, G. (1976). Verbal turn-taking and exchanges in faculty dialogue. In B.L. Dubois and I.M. Crouch (Eds.), The Sociology of the Lan uages of American Women. San Antonio, TX: TrinityfiUniversity Rress. 100 Edlesky, C. (1976a). The acquisition of communicative competence: Recognition of linguistic correlates of sex roles. Merrill- Palmer Quarterly, 22, 47-59. Edlesky, C. (1976b). Subjective reactions to sex-linked language. The Journal of Social Psychology, 92, 97-104. Edlesky, C. (1976c). Recognizing sex-linked language. Language Arts, so, 746-752. _““‘"‘ Edlesky, C. (1981). Who's got the floor? Language in Society, 19, 383-421. Edwards, J.R. (1977). Social class differences and the identification of sex in children's speech. Journal of Child Languajg, o, 121-127. Elliott, J. (1978). Sex role constraints on freedom of discussion: A neglected reality of the classroom. The New Era, pp, 147-155. Erickson, 8., Lind, E.A., Johnson, B.C., & O'Barr, W.M. (1978). Speech style and impression formation in a roust setting: The effects of "powerful" and "powerless" speech. Journal of Experimental Social Psychology, 14, 266-279. Esposito, A. (1979). Sex differences in children's conversation. Language and Speech, 22, 213-220. Feshbach, N.D., Dillman, A.S., & Jordan, T.S. (1979). Portrait of a female on television: Some possible effects on children. In C.B. Kopp (Ed.), Becoming female: Perspectives on development (pp. 363-386). New York: Plenum Press. Fillmer, H.T., & Haswell, L. (1977). Sex-role stereotyping in English use. Sex Roles, 2, 257-263. Fischer, J.L. (1958). Social influences on the choice of a linguistic variant. Word, 14, 47-56. Fishman, P.M. (1978). Interaction: The work women do. Social Problems , gs, 397-406. Fraser, C., & Roberts, N. (1975). Mothers' speech to children of four different ages. Journal of Psycholinguistic Research, 4, 9-16. Garnica, O.K. (1977). Some prossodic and paralinguistic features of speech to young children. In C.E. Snow and C.A. Ferguson (Eds.), Talking_to children: Language input and acquisition (pp. 63-88). London: Cambridge University Press. 101 Garvey, C., 8 BenDebba, M. (1974). Effects of age, sex, and partner on children's dyadic speech. Child Development, 42, 1159-1161. Giles, H., Scholes, J., 8 Young, L. (1983). Stereotypes of male and female speech: A British study. Central States Speech Journal, 24, 255-256. Gilley, H.M., 8 Summers, C.S. (1970). Sex differences on the use of hostile verbs. Journal of Psychology, 12, 33-37. Gleser, G.C., Gottschalk, L.A., 8 John, W. (1959). The relationship of sex and intelligence to choice of words: A normative study of verbal behavior. Journal of Clinical Psychology, 12, 182-191. Golinkoff, R.M., 8 Ames, G.J. (1979). A comparison of fathers' and mothers' speech with their young children. Child Development, 50, 28-32. Greif, E.B. (1980). Sex differences in parent-child conversations. Women's Studies International Quarter1y, 2, 253-258. Haas, A. (1981). Partner influences on sex-associated Spoken language of children. Sex Roles, 1, 925-935. Harms, I., 8 Spiker, C. (1959). Factors associated with the performance of young children on intelligence scales and tests of speech development. Journal of Genetic Psychology, 24, 3-22. Hartman, M. (1976). A descriptive study of the language of men and women born in Maine around 1900 as it reflects the Lakoff hypotheses in "Language and Woman's Place." In B.L. Dubois and I.M. Crouch (Eds.), The sociology of the languages of American women. San Antonio, TX: Trinity University Press. Heiss, J.S. (1962). Degree of intimacy and male-female interaction. Sociometry, 22, 197-208. Hershey, S., 8 Werner, E. (1975). Dominance in marital decision making in women's liberation and non-women's liberation families. Family Process, 14, 223-233. Hilpert, F., Kramer, C., 8 Clark, R.A. (1975). Participants' perceptions of self and partner in mixed sex dyads. Central States Speech Journal, 22, 52-56. Hogan, 0. (1976). Sex differences in language development. Paper presented at the meeting of the Midwestern Psychological Association, Chicago. Ickes, W., 8 Barnes, R.D. (1977). The role of sex and self-monitoring in unstructured dyadic interactions. Journal of Personality and Social Psychology, 22, 315-330. 102 Irwin, O.C., 8 Chen, H.P. (1946). Development of speech during infancy: Curve of phonemic types. Journal of Experimental Psychology, 22, 431-436. Jay, T.B. (1980). Sex roles and dirty word usage: A review of the literature and a reply to Haas. Psychological Bulletin, 22, 614-621. Jespersen, O. (1922). Language, its nature, development, and origin. London: George Allen and Unwim Limited. Johnson, J.L. (1980). Questions and role responsibility in four professional meetings. Anthropological Linguistics, 22, 66-76. Katz, P.A. (1979). The development of female identity. Sex Roles, 5, 155-178. Katzman, N. (1972). Television soap operas: What's been going on anyway? Public Opinion Quarterly, 22, 200-212. Key, M.R. (1972). Linguistic behavior of male and female. Linguistics, 22, 15-31. Key, M.R. (l975). Male/female language. Metuchen, NJ: The Scarecrow Press. Klann-Delius, G. (1981). Sex and language acquisition - Is there any influence? Journal of Pragmatics, 2, 1-25. Kramer, C. (1974). Folk-linguistics. Psychology Today, 2, 82-85. Kramer, C. (1977). Perceptions of female and male speech. Language and Speech, 22(2), 151-161. Kramer, C. (1978). Women's and men's ratings of their own and ideal speech. Communication_9uarterly, 22(2), 2-11. Lakoff, R. (1973). Language and woman's place. Language in Society, 2, 45-79. Lakoff, R. (1975). Language and woman's pjace. New York: Harper and Row. Lapadat, J., 8 Seesahai, M. (1977). Male versus female codes in informal contexts. Sociolinguistics Newsletter, 2(3), 7-8. Malone, M.J., 8 Guy, R.F. (1982). A comparison of mothers' and fathers' speech to their 3-year-old sons. Journal of Esycholinguistic Research, 2, 599-608. 103 Maxwell, E. (1980). Reactions to women's speech variation. In C.L. Berryman and V.E. Eman (Eds.), Communication, lapguage and sex: Proceedings of the first annual conference (pp. 217-228). Rowley, MA: Newbury House Publishers. McGhee, P.E. (1979). The role of laughter and humor in growing up female. In C.B. Kopp (Ed.), Becoming female (pp. 183-206). New York: Plenum Press. McMillan, J.R., Clifton, A.K., McGrath, 0., 8 Gale, W.S. (1977). Women's language: Uncertainty or interpersonal sensitivity and emotionality. Sex Roles, 2, 545-559. Meditch, A. (1975). The development of sex specific speech patterns in young children. Anthropological Linguistics, 12, 421-433. Morley, M.E. (1957). The development and disorder of speech in young children. London: Livingstone. Moss, H.A. (1967). Sex, age, and state as determinants of mother- infant interaction. Merrill-Palmer Quarterly, 113, 19-36. Muller, E. (1972). The maintenance of verbal exchange between young children. Child Development, 42, 930-938. Natale, M., Entin, E., 8 Jaffe, J. (1979). Vocal interruptions in dyadic communication as a function of speech and social anxiety. Journal of Personality and Social Psychology, 22, 865-878. Nelson, K. (1973). Structure and strategy in learning to talk. Monographs of the Society for Research in Child Development, 22, Nemeth, C., Endicott, J., 8 Wachtler, J. (1976). From the '50's to the '70's: Women in jury deliberations. Sociometry,‘22, 293-304. Newcombe, N., 8 Arnkoff, 0.8. (1979). Effects of speech style and sex of speaker on person perception. Journal of Personality and Social Psychology, 21, 1293-1303. Noller, P. (1980). Cross-gender effect in two-child families. DevelOpmental Psychology, 12, 159-160. Oliver, M.M., 8 Rubin, J. (1975). The use of expletives by some American women. Anthropo1ogical Linguistics, 12, 191-197. Parsons, E.C. (1913). The old fashioned woman: Primitive fancies about the sex. New York: G.P. Putnam and Sons. Parsons, J.E., 8 Bryan, J. (1978). Adolescence: Gateway to androgyny. Michigan Occasional Papers in Women's Studies, 2, 1-25. 104 Phillips, J.R. (1973). Syntax and vocabulary of mother's speech to young children: Age and sex comparisons. Child Development, 44, 182-185. Piliavin, J.A., 8 Martin, R.R. (1978). The effects of the sex composition of groups on style of social interaction. Sex Roles, 4, 281-296. Poole, M.E. (1979). Social class, sex and linguistic coding. Language and Speech, 22, 49-67. Reik, T. (1954). Men and women speak different languages. Psychoanalysis, 2, 3-15. Sause, E.F. (1976). Computer content analysis of sex differences in the language of children. Journal of Psycholinguistic Research, 5, 311-324. Scott, K.P. (1980). Perceptions of communication competence: What's good for the goose is not good for the gander. In C. Kramarae (Ed.), The voices and words of women and men (pp. 199-208). Oxford: Pergamon Press. Serbin, L.A., 8 O'Leary, K.D. (1979). How nursery schools teach girls to shut up. In J.H. Williams (Ed.), Psychology of women: Selected readings (pp. 183-187). New York: W.W. Norton and Company. Siegler, D.M., 8 Siegler, R.S. (1976). Stereotypes of males' and females' speech. Psychological Reports, 22, 167-170. Silverman, E., 8 Zimmer, C.H. (1976). The fluency of women's speech. In B.L. Dubois and I.M. Crouch (Eds.), The sociolo of the languages of American women. San Antonio, TX: r1n1ty University Press. Snow, C.E. (1972). Mothers' speech to children learning language. Child Development, 42, 549-565. Snow, C.E. (1977). Mothers' Speech research: From in ut to interaction. In C.E. Snow and C.A. Ferguson (Eds.§, Talkin to children: Language ipput and acquisition (pp. 31-50). London: Cambridge university Press. Soskin, W.F., 8 John, V.P. (1963). The study of spontaneous talk. In R. Barker (Ed.), The stream of behavior (pp. 228-281). New York: Appleton-Century-Crofts. Staley, C.M. (1982). Sex-related differences in the style of children's language. Journal of Psycholinguistic Research, 11, 141-158. . 105 Stoneman, Z., 8 Brody, G.H. (1981). Two's company, three makes a difference: An examination of mothers' and fathers' speech to their children. Child Development, 22, 705-707. Swacker, M. (1975). The sex of the speaker as a sociolinguistic variable. In B. Thorne and N. Henley (Eds.), Language and sex: Difference and dominance (pp. 76-83). Rowley, MA} NeWbury House Publishers. Swacker, M. (1976). Women's behavior at learned and professional conferences. In B.L. Dubois and I.M. Crouch (Eds.), 122 sociology of the languages of American women. San Antonio, TX: Trinity University Press. Thorne, B. (1983). An analysis of gender and social groupings. In L. Richardson and V. Taylor (Eds.), Feminist frontiers: Rethinking sex, gender,_and society (pp. 61-63). Reading, MA: Addison-Wesley Publishing. Thorne, B., 8 Henley, N. (Eds.). (1975). Language and sex: Difference and dominance. Rowley, MA: Newbury House Publishers. Thorne, 8., Kramarae, C., 8 Henley, N. (1983). Language, gender, and society. Rowley, MA: Newbury House Publishers. Trudgill, P. (1975). Sex, covert prestige, and linguistic change in the urban British English of Norwich. In B. Thorne and N. Henley (Eds.), Language and sex: Difference and dominance (pp. 88-104). Rowley, MA: Newbury House Publishers. Trudgill, P. (1983). Sociolinguistics: An introduction to language and society. New York: Penguin Books. Von Raffler-Engel, W., Smith, A.L., 8 Cunningham, R.D. (1978, August). Patterns of dominance in same-sex and other-sex gyadic interaction. Paper presented at the IXth World Congress of Soéiology, University of Uppsala, Sweden. West, C., 8 Zimmerman, D.H. (1977). Women's place in everyday talk: Reflections on parent-child interaction. Social Problems, 24, 521-529. West, C., 8 Zimmerman, D.H. (1983). Small insults: A study of interruptions in cross-sex conversations between unacquainted persons. In B. Thorne, C. Kramarae, and N. Henley (Eds.), Language, gender and society (pp. 103-118). Rowley, MA: Newbury House Publishers. Wood, M.M. (1966). The influence of sex and knowledge of communication effectiveness on spontaneous Speech. Word, 22, 112-137. "" 106 Wootton, A.J. (1974). Talk in the homes of young children. Sociology, 2, 277-295. Zimmerman, D.H., 8 West, C. (1975). Sex-roles, interruptions, and silences in conversation. In B. Thorne and N. Henley (Eds.), Language and sex: Difference and dominance (pp. 105-129). Rowley, MA: Newbury House Publishers. "I111111111111111111“