Tobacco Documents Online

On the Relation Between Family SmoRing Habits and the Smoking Behavior of College Stndents Charles D. Spielberger, Gerard A. Jacobs, Rosario S. Crane and Stephen F. Russell University of South Florida Runaing Head: Family Smoking Habits and Student Smoking Behavior

Abstract This study investigated the relationship between the smoking behavior of college students (603 females; 352 males) and the smoking habits of their parents and older siblings. Students responded to a Smoking Behavior Questionnaire that required them to report whether they were Current Smokers, Occasional Smokers, Ex-Smokers, or Non-Smokers; similar information was obtained about their parents and older siblings. Of the females, 49: were classified as Smokers as compared to only 37S of the males. Students whose parents or older siblings smoked were more likely to bs Smokers themselves; older siblings appeared to have a greater influence on the smoking behavior of younger siblings than did their parents. No differe^.ces were foim d in the smoking habits of the parents or the older siblings of Current, Occa- sional and Ex-Smokers. Contrary to previous iavestigations, there was no evidence of 'same-sea parental modeling of smoking.behavior. On the Relatio: Smoking The controversy on smc in identifying factors that behavior. In reviews of re Wohlford & Giammona, 1969; & Cleary, 1977, 1980), soci habits and peer-group press initiation of smoking, but maintain the smoking habit. Positive relationships smoking behavior of their c Bewley, Bland, Dean, & Poll Horn, Courts, Taylor, & Sol Newman, 1970; Palmer, 1970; one study, which was based to find any relationship be Although an empirical relat: smoking behavior seems firm: tionship reflects environmer Eysenck, 1980). Wohlford (1970) has ea differential impact of the e behavior of their sons and c smoke if their fathers smoke

817 I Smoking Behavior =~ Smoking Behavior 2 On the Relation Between Family Smoking Habits and the Smoking Behavior of College Students bet.reen the smoking behavior of and the smoking habits of their ?onded to a Smokina Behavior t whether they were Current Smokers, 3kers; similar information was alings. Of the females, 49: were 37Z of the males. Students whose a likely cc be Smokers themselves; influence on the smoking behavior 3. No differences were found in 31der siblings of Current, Occa- 3us investigations, there was no smoking.behavior. The controversy on smoking and health has stimulated extensive interest in identifying factors that influence the initiation and maintenance of smoking behavior. In reviews of research in this field (Yatarazzo S;Satarazzo, 1968; Wohl.ford 6 Giammona, 1969; Evans, Henderson, Hill, & Raines, 1979; Leventhal & Cleary, 1977, 1980), social influence variables such as parental smoking habits and peer-group pressures have been repeatedly identified with the initiation of smoking, but relatively little is known about the factors that maintain the smoking habit. z Positive relationships between the smoking habits of parents and the smoking behavior of their children have been reported in eight studies (Baracs, Bewley, Bland, Dean, & Pollard, 1978; Borland & Rudolph, 1975; Clausen, 1968; Horn, Courts, Taylor, 6 Salomon, 1959; Merki, Creswell, Stone, Huffaaa, & Newman, 1970; Palmer, 1970; Salber & Hadsahon, 1961; Wohlford, 1970); only one study, which was based on a very small sample of college students, failed to find any relationship between these variables (Straits & Sechrest, 1963). Although an empirical relationship between parental smoking habits and children's smoking behavior seems firmly established, it is not clear whether chis rela- tionship reflects environmental or constitutional-genetic influences (see Eysenck, 1980). Wohlford (1970) has called attention to the imnortance of examining the differential impact of the smoking habics of fathers and nochers on the smoking behavior of their sons and daughters. Sons were found to be more likely to smoke if their fathers smoked in four studies that examined this relationship 1 11

818 Smoking Behavior 3 (Banks et al.; 1978; Horn et al., 1959; Salber 6:fac-Mahon, 1961; Wohlford, 1970), and daughters were found to be more 1iiely to smoke if their mothers smoked in three of these studies. Since no relationship was found between the smoking habits of fathers and daughters, nor bet•.aeen mothers and sons, same-sex parental modeling appears to have a stronger impact than genetic factors on children's smoking behavior. Peer group pressure is also widely recognized as a primary factor in the initiation of smoking (e.g., Eysenck, 1980; Watarazzo & w.atarazzo, 1968). Evidence of peer-group influence on the initiation of smoking was reported in six studies that investigated this relationship (Banks et al., 1978; Hill, 1971; Levitt & Edwards, 1970; :Satthews, 1974; :!er'.ci et,al., 1970; Palmer, 1970). Leventhal and Cleary (1980) have recently suggested that peers and parents are both important sources of environmental influence in cigarette smoking, and that older siblings may be even more important than other peers in influenc'_ng adolescents to initiate smoking. Consistent with this view, Banks et al. (1978) found that junior and senior high school students whose siblings smoked were more likely to be smokers themselves. Only one published study could be located in which the relationship be- tween family smoking habits and the maintenance of smoking behavior was inves- tigated. Laoye, Creswell and Stone (1972) found that secondary school students who were re ular smokers were more likel to have arents and friends who g p y smoked than students who had been regular smokers but subsequently stopped. The goals of the present study were to investigate the influence of the smoking habits of parents and older siblings on the initiation and maintenance of smoking behavior for college students. on the basis of previous research findings, positive relationships were expected between family smoking habits and the initiation of et fathers and sons,'and mc behavior was investigate smokers. Qn the basis o smokers were expected to older siblings who smake Sub ects The subjects were 9_ level psychology courses was 19 years. All stude: credit for their particiF comprised primarily of fz than 85 percent of the 24 many of them continue to The students were te sisted of 460 students (1 Spring quarters of the 19 students enrolled in intr sample. Sample II consis during the Fall quarter c cent of the students enro: Construction of the Smoki: T13e SBQ is a 50-item information about student:

Smoking Behavior 3 :facYahon, 1961; Wohlford, to smoke if their mothers :ionship was found between between mothers and sons, Dnger impact than genetic A as a pra--zary factor in the ,azzo & xatarazzo, 1968). - n of smoking was reported p (Banks et al., 1978; Hill, ki et al., 1970; Palmer, 1970). ed that peers and parents are in cigarette smoking, and an other peers in influencing * . which the relationship be- ' smoking behavior was inves- =hat secondary school students parents and friends who ° ;7 but subsequently stopped. '' :igate the influence of the a initiation and maintenance ' basis of previous research :reen family smoking habits Smoking Behavior 4 and the initiation of amoking behavior, and between the smoking habits of fathers and sons, and mothers and daughters. The maintenance of smoking behavior was investigated by comparing ex-snokers with current and occasional smokers. On the basis of the findings reported by Laoye et al. (1972), current smokers were expected to be more likely than ex-smokers to have parents and older siblings who smoked. Method Subiects The subjects were 955 undergraduate students enrolled ir. introductory level psychology courses at a large urban state university; the median age 9 was 19 years. All students volunteered to take part in the study and received credit for their participation. The population served by the university is comprised primarily of families of low to average socio-economic status. More than 85 percent of the 24,000 students at the university are commuters, and many of them continue to reside with their parents. The students were tested over a period of ten months. Sample I con- sisted of 460 students (166 males, 294 females) tested during the Winter and Spring quarters of the 1978-79 academic year; approximately 65 percent of the students enrolled in introductory psychology courses were included in this sample. Sample II consisted of 495 students (186 males, 309 females) tested during the Fall quarter of 1979; this sample included approximately 80 per- cent of the students enrolled in introductory psychology courses. Construction of the Smoking Behavior Questionnaire (SBO) The SBQ is a 50-icem self-report questionnaire designed to elicit specific information about students' smokiag behavior and the smoking habits of their

820 .5 families. In constructing the SBQ, a number of questionnaires used to evaluate smoking habits in previous investigations were carefully reviewed (Clausen, 1968; Horn et al., 1959; Ikard, Green & Horn, 1969; Leventhal & Avis, 1976); items from these instruments were adapted for the present study. A preliminary form of the SBQ was administered to 149 undergraduate students (52 males, 97 females) enrolled in introductory psychology courses. In responding to the questionnaire, the students were asked to indicate if they were interested in meeting in smal], groups to discuss reasons why college students start and continue to smoke: Each student was promised $2.00 for - participating in these discussion sessions. A total of 81 students met with z the investigators in small groups of 7 to 10 students. In order to per=it in-depth discussion of the students' reasons for smaking or not smoking, there were separate groups for current smokers (:h27), ex-smokers (N-17), and non-smokers (N-37). The group discussions were audio tape-recorded . On the basis of an analysis of the responses to the preliminary S3Q and a review of the audio tape, the final set of items for the form of the SBQ that was used in the present study were select % The data presented in this study are based on the first part of the questionnaire, which inquired about the smoking habits of the students, their parents, and their older brothers and sisters.l Procedures The SBQ was administered as part of a larger test battery to groups of 20 to 100 students. In order to encourage participation in the study, most of the group-testing sessions were scheduled immediately after the psychology classes from which the students were recruited. At the begianiag of each tasting session, the students were required to read and then sign a Consent Form, whicy in: tary, and woulc were then inst: as possible. i could learn mo; Prior to admin: subjects: Now, turn tc whether you never smokec "Your Self" time to timF to also chec Current, Occ more than o: older broth, In evaluat initiation of s and c^x-Smokers in the statist: habits of parer the students wF treated as sepa students who re

.5 r of questionnaires used to evaluate are carefully reviewed (Clausen, a, 1969; Leventhal & Avis, 1976);_s` or the present study. cistered to 149 undergraduate .., introductory psychology courses. ients were asked to indicate if ., 3ups to discuss reasons why college student was promised $2.00 for A total of 81 students met with . -.~ " . ;;_. ) students. In order to permit . PSS ; for smoking or not smoking, :rs_, ~ ss (:7-27) , ex-smakers (N-17) , ro~ :ns were audio tape-recorded . .~~, )onses to the preliminary S3Q and -~ : items for the form of the SBQ = .cted/ The data presented in this ~ .stionnaire, which inquired about :rents, and their older brotheis% r ts~r'' arger test battery to groups of_Y0. ra ' of icioation in the study, most 4~~ mediataly after the psychology y". 821 Smoking Behavior 6 Form, which informed them that participation in the study was entirely volun- tary, and would consist of responding to several questionnaires. The etudents were then instructed to answer each questior.naire as honestly and accurately as possible. They were also informed that feedback sessions in which they could learn more about the study would be scheduled at the end of the tera.2 Prior to administering the SBQ, the followizg instructions were read to the subjects: Now, turn to the Cigarette Smoking Questionnaire...For Section 1, indicate whether you are a Current Smoker, Occasional Smoker, ax-Smoker, or have never smoked by placing a check in the appropriate space under the column "Your Self". An "b ccasional Smoker" is someone who smokes cigarettes from time to time but not everyday. Whatever your own smoking habits, be sure to also check whether your father, mother or older brother or sister is a Current, Occasional, or Ex-Smoker or has never smoked. (Students with more than one older sibling were asked to provide information about the older brother and/or sister who were the heaviest smokers.) Results In evaluating the relationship between family smoking habits and the initiation of smoking behavior, students classified as Current, Occasional and r".x-Smokers were considered "Smokers", and were treatad as a single group in the statistical analyses. In examining the relation between the smoking habits of parents and older siblings and the maintenance of smoking behavior, the students who were classified as Current, Occasional, and Ex-Smnkars were treated as separate independent groups. In all of the statistical analyses, students who reported they had never smoked were classified as "Non-Smokers".

822 Smoking Behavior 7 Students who indicated they had experimented briefly with cigarettes, but had never become either regular or•accasional smokers were-also included in the Non-Smaker category. The percentages of male and female students classified as Smokers or Non-Smokers in the two samples are reported in Table 1. Since these percen- tages were quite similar, the data for the two samples were combined. For the combined sample, the percentage of female smokers (49'„) was substantially larger than the percentage of male smokers (37Z); the difference between these percen- tages was statistically significant (Y2-12.94, df-1, p<.001). Insert Table 1 about he;e Family Smoking Habits and Students' Smokina Behavior On the basis of the smoking habits of their parents, Smokers and Non- Smokers were assigned to one of the following three categories: (a) Neither parent smoked; (b) Mother or Father smoked, but not both; and (c) Both parents smoked. In Table 2, it can be noted that most students who reported that neither parent smoked were themselves Non-Smokers (females, 64z; males, 717.), and that the percentage of Smokers was higher if mother or father or both parents smoked than if neither parent smoked. In separate 3 x 2 Chi Square analyses, the relationship between parental smoking habics and the students' smoking behavior was statistically significant for females (:C2-11.03, df-2, p <.01), but not for males (%2-3.57, df-2). Insert Table 2 about here In order to c12 and the smoking habi Smokers and Non-Smok in separate 2 x 2 Ch either mother or fat higher percentage of (X--11.33, df-l, p < were themselves Smok the percentage of Sm, (291) who reported t: not for males (35Z) c Surprisingly, tt who reported that onc parents smoked, but t either females (8-.1. for males and femalee sexes were combined a T approached significat If there is aame expected that the per fathers smoked, and a who reported that the of the 414 females wh smoked, 38z were Smok selves Smokers. Sinc approzimateiy the sam

Smoking Behavior 7 -iefly with cigarettes, but mokers were•also included in :s classified as Smokers or Table 1. Since these percen- samples were combined. For the :rs (49S) was substantially larger difference between these percen- r parents, Smokers and Non- hree categories: (a) Neither not both; and (c) Both parents students who reported that rs (females, 645; males, 71L), f mother or father or both In separate 3 x 2 Chi Square icing habits and the students' for females (:&11.03, df-2, iere 823 Smoking Behavior 8 In order to clarify the relationship between students' smoking behavior and the smoking habits of their parents, the percentages of female and male Smokers and Non-Smokers who reported that "Neither Parent Smoked" were compared in separate 2 x 2 Chi Square analyses with the percentages who reported that either mother or father smoked, or that both parents smoked. A significantly higher percentage of the females who reported that mother or father smokad (%2-11.33, df-l, p< .001), or that both oarents smoked (X2-6.75, df-1, p,-.01), were themselves Smokers. For males who reported that mother or father smoked, the percentage of Smokers (44%) was also significantly higher than for males (297.) who reported that neither pareat smoked (%2a3.98, df-1, p<.05), but a not for males (35%) who reported that both Darents smoked (:C2•0.86, df-1). Surprisingly, the percentage of Smokers was slightly higher for students who reported that one parent smoked than for those who reported that both parents smoked, but these differences were not statistically significant for either females (%-'-1.17, df-l) or males (3--1.95, df=1). Since the trends for males and females in this comparison were similar, the data for the two sexes were combined and evaluated in a 2 x 2 Chi Square analysis, which - approached significanca (%2-3.06, df-1, p <.10). If there is same-sex modeling of parental smoking habits, it would be expected that the percentage of smokers would be greater among sons whose fathers smoked, and among daughters whose mothers smoked. Of the 336 females who reported that their mothers smoked, 52% were Smokers, as compared with 51: of the 414 females whose fathers smoked. For the 259 males whose fathers smoked, 38S were Smokers; 39% of the 188 males whose mothers smoked were them- selves Smokers. Since the percentage of females and males who smoked was approximately the same, irrespective of whether their mathers or fathers

824 Smoking Behavior 9 smoked, there appears to be little evidence of same-se= modeling by children of the smoking behavior of their parents. Of the 955 students who participated in the present study, 398 females and 223 males reported that they had older siblings, and 177 females and 80 males reported that they had both older brothers and older sisters. For the students with older siblings, it can be seen in Table 3 that the percentage of Smokers and Non-Smokers was approximately the same as for the total sample (compare colusai 1 in Table 3 with the data for the combined sample in Table 1). Insert Table 3 about here 9 Students with older siblings were assigned to one of the following three categories: (a) Neither older brother nor older sister smoked; (b) Older brother or older sister smoked, but not both; (c) Older brother and older _ sister both smoked. The percentage of Smokers and Non-Smokers in each of these categories is reported in Table 3, in which it may be noted that the percentage of Smokers was highest for students with older brothers and sisters who smoked and lowest for those whose siblings were non-smokers. More than 70Z of the students of both sexes whose older siblings were non-smokers were themselves Non-Smokers. Differences in the percentages of Smokers and Non- Smokers in the three categories were evaluated in 2 x 3 Chi Square analyses, and found to be significant for both females (X'-43.39, df-2, p<.001) and males (%-.9.96, df-2, p <.01). The relationship between students' smoking behavior and the smoking habits of their older siblings was further evaluated in 2 x 2 Chi Square analyses. The percentage of Smokers was significantly higher for students whose older-brothers or si brothers and sisters who s: were non-smokers Column 3 Males, %2•7.86, df 1, P <.( p <.001; Males, %2-5.63, d: in the "brother and sister "brother or sister smoked" significant for either femz In order to evaluate t brothers and older sisters of their younger siblings, students whose older siblin older sisters smoked, the p higher than the percentage whereas the difference in t Smokers for the 231 student (%2-0.73, df-1). Thus, old smoking behavior of their y whereas older brothers appe behavior of younger sibling. In the preceediag anal: influence on the smoking be: parents. The combined infl siblings on the smoking beh. present study is examined i.• percentages of Smokers and :