Pregnancy Should Women Put Up Their Feet or Lace Up Their Running Shoes?: Self-Presentation and the Exercise Stereotype Phenomenon During Pregnancy

Transcription

1 Journal of Sport & Exercise Psychology, 2012, 34, Human Kinetics, Inc. Pregnancy Should Women Put Up Their Feet or Lace Up Their Running Shoes?: Self-Presentation and the Exercise Stereotype Phenomenon During Pregnancy Anca Gaston, Anita Grace Cramp, and Harry Prapavessis University of Western Ontario Little is known about how women who exercise during pregnancy are perceived. The purpose of this study was to investigate whether the positive exercise stereotype (i.e., the general tendency for exercisers to be evaluated more positively than nonexercisers) extends to pregnancy. Adult women (N = 202, mean age = years, SD = 13.46) were randomly assigned to read a description of one of the following pregnant female targets: regular exerciser, active living, excessive exerciser, nonexerciser, or control. Participants then rated the target on 12 personality and 8 physical dimensions. MANOVAs revealed a significant main effect for both physical and personality attributes (p <.05). The regular exerciser and active living target received the most positive ratings on both physical and personality dimensions. Whereas the excessive exerciser received high ratings on most physical characteristics, this target was also perceived as meaner and sadder, and having fewer friends than all other targets. Keywords: exercise, impression formation, exercise psychology, social desirability The benefits of exercise in the general population are well known, and research conducted over the past 25 years has demonstrated that exercise during pregnancy is associated with many benefits for both mother and baby. For example, regular exercise during pregnancy improves or maintains cardiovascular fitness, enhances psychological well-being, helps manage pregnancy-related musculoskeletal conditions, as well as reduces the risk of two serious maternal fetal conditions: gestational diabetes and preeclampsia (American College of Sports Medicine, 2006; Lewis, Avery, Jennings, et al., 2008; Weissgerber, Wolfe, Davies & Mottola, 2006). Given these benefits, Canadian guidelines recommend that in the absence of contraindications, pregnant women should exercise for 30 min at a moderate intensity on 3 4 days of the week (Mottola, in press). Anca Gaston, Anita Grace Cramp, and Harry Prapavessis are with the Department of Kinesiology, University of Western Ontario, London, ON, Canada. 223

2 224 Gaston, Cramp, and Prapavessis Despite such recommendations, many expectant mothers are inactive. The results of two recent reviews demonstrated that exercise levels tended to decrease from pre-pregnancy to pregnancy and that approximately 60% of pregnant women were inactive (Gaston & Cramp, 2011; Poudevigne & O Connor, 2006). Research supports the view that the decline in exercise across pregnancy results not only from physical changes that may make exercise more difficult, but also from a combination of social and psychological changes (Clarke & Gross, 2004). Indeed, the transition to motherhood is complex and central to how women are defined both by themselves and by others (Phoenix, Woollett, & Lloyd, 1991). While pregnancy can be a time of psychological growth and opportunity, mothers-to-be can just as easily find themselves in a fragile, overwhelming, and frightening space (Solchany, Sligar, & Barnard, 2002). As they negotiate new social and psychological roles, pregnant women may thus experience heightened susceptibility to influence from those around them, particularly when it comes to engaging in health behaviors such as exercise (Cannella, 2006; Clarke & Gross, 2004; Thornton et al., 2006). People choose to engage or not engage in exercise for a variety of reasons, including self-presentational motives. Self-presentation refers to people s attempt to monitor and control the impressions others form of them (Leary & Kowalski, 1990) and research has shown that information about people s exercise habits influences the impression formation process. For example, Hodgins (1992) found that compared with people described as inactive and not fit, targets described as fit and regular exercisers were rated more favorable on a variety of personality variables such as self-confidence, happiness, independence, being sociable, and being friendly. In an investigation into the relationship between exercise status and impression formation, Martin, Sinden, and Fleming (2000) developed two bipolar scales: a 12-item personality attributes scale (e.g., lazy works hard, afraid brave, sad happy) and an 8-item physical attributes scale (e.g., ugly good looking, scrawny muscular, sick healthy). Using these measures, they demonstrated that a target described as a regular exerciser received higher ratings on the majority of personality and physical attributes compared with a nonexerciser or control target. This work was later extended to include an active living and excessive exerciser target. Using these same personality and physical attributes scales, Lindwall and Martin Ginis (2006) and Martin Ginis, Latimer and Jung (2003) found that with respect to personality variables, the active living and general exerciser were rated more favorably than the excessive exerciser, nonexerciser, or control. However, results were less consistent with regards to physical ratings. While Martin Ginis et al. reported that the positive ratings received by the active living and regular exerciser on physical attributes also extended to the excessive exerciser, Lindwall and Martin Ginis found that the excessive exerciser was perceived as having a less attractive figure and being sicker than the other two exercise targets. However, all three exercise targets received similar ratings on the remaining physical attributes. The tendency for exercisers to be evaluated more positively than nonexercisers has been referred to as the positive exercise stereotype, and evidence exists that this phenomenon is pervasive and irrespective of participants own exercise status (Rodgers, Hall, Wilson, & Berry, 2009). Several attempts have been made to

3 Exercise Stereotype Pregnancy 225 understand why regular exercisers seem to be positively stereotyped. According to Sodenstrom (1999), exercise is perceived as a way to achieve physical attractiveness, and physical attractiveness is linked to other valued attributes such as popularity and social competence. Others have argued that because exercise is often perceived as a means of improving one s health, exercisers demonstrate greater personal responsibility over their own lives and are thus more moral than nonexercisers (Brownell, 1991; White, Young & Gillett, 1995). In addition, exercise may benefit from being associated with other healthy lifestyle behaviors such as losing weight and not smoking (White et al.). Furthermore, physical and personality characteristics influence the level of social support an individual receives (Sarason, Sarason, & Shearin, 1986), and social support plays a role in determining an individual s level of activity (Lutz, Linder, & Greenwood, 2005). For example, Chogahara (1999) found that leisuretime energy expenditure was strongly related to positive social influences from family and friends, and a review by Carron, Hausenblas, and Mack (1996) revealed a moderate effect size for family influences on exercise behavior. Research suggests that social support may be especially critical during pregnancy. Collins et al. (1993), for example, found that women who reported higher levels of social support during pregnancy gave birth to babies who scored higher on the APGAR test (a standard tool used to quickly assess a newborn s overall well-being) and experienced less postpartum depression. Given these results, the absence of support for exercise during pregnancy may be particularly detrimental. Using a prospective survey design, Clarke and Gross (2004) examined exercise levels and maternal exercise beliefs across pregnancy and found that 52% of participants reported that they had ceased or reduced their exercise participation in response to advice from family or friends. In addition, 85% of women who had received advice from family and friends reported that they had been discouraged from being active. Despite the benefits of exercise during pregnancy and the potential for pregnant women to be swayed by the impressions of those around them, to the best of our knowledge no study has ever explored whether the positive exercise stereotype extends to pregnancy. Hence, this study represents the first investigation of this phenomenon in a sample of adult women. This population was chosen based on the rationale that pregnant women are more likely to turn to other women when it comes to health advice or social support (Thornton et al., 2006). Using a fivegroup experimental design and drawing on the results of previous studies (Lindwall & Martin Ginis, 2006; Martin Ginis et al., 2003), the following hypotheses were proposed: (a) for personality attributes, the regular exerciser and active living targets will receive more positive ratings compared with the excessive exerciser, nonexerciser, and control targets and (b) for physical attributes, all three exercise targets (regular exerciser, active living, and excessive exerciser) will receive more positive ratings than the nonexerciser and control targets. Although previous findings indicated that the excessive exerciser was not evaluated as positively as the regular exerciser and active living target on all physical attributes, the three exercise targets still received similar scores on the majority of items (Lindwall & Martin Ginis, 2006; Martin Ginis et al., 2003).

4 226 Gaston, Cramp, and Prapavessis Participants Method Two hundred and two adult women from Southwestern Ontario were recruited during the month of July The mean age of the participants was years (SD = 19.79; range = 19 79). All demographic characteristics are presented by group in Table 1. A priori sample size calculation was based on the results obtained in a similar study. Using an undergraduate population and a five-group multivariate design, Lindwall and Martin Ginis (2006) found a significant main effect for target group (η 2 =.26). Based on this large effect size, Cohen (1992) recommends that approximately 42 participants per experimental group are required to be powered at.80 when the alpha is set at.05. Therefore, the current study s five-group design aimed for a sample size of approximately 200 participants. Procedure and Materials Following institutional ethics approval, a mass seeking volunteers was distributed to all the newsgroup subscribers of a public district school board in Ontario, Canada. Individuals on the list included elementary and secondary school teachers, other school staff, board employees as well as some retired teachers and staff. This recruitment strategy was chosen based on the overrepresentation of Canadian women in educational and administrative positions (Statistics Canada, 2010) as well as the potential to reach a wide age distribution. Participants were told that the researchers were interested in the impressions that people form about pregnant women and that the survey would take approximately 10 min of their time. A link to the online questionnaire was provided. Clicking on the link redirected participants to a secured online survey website ( After providing informed consent, participants completed demographic measures, and then read one of five descriptions of a pregnant woman. Due to design limitations associated with the online survey website, a quasi-random group allocation method was used. Specifically, surveymonkey.com did not provide the option of randomly allocating participants to one of the five experimental groups. After completing the demographics questionnaire, participants were asked to pick any number between 1 and 5. Based on the number chosen, skip logic was programmed into the questionnaire design to redirect each participant to the unique target description associated with that particular number (e.g., choosing 1 might result in an individual being directed to the regular exerciser target description whereas choosing 2 might result in the active living target, etc.). To eliminate the risk of group differences based on personal number preferences, a second step was employed. A random-numbers list was generated in advance of the study by the lead investigator using an online randomization program (Urbaniak & Plous, 2011). Every 15 min, the skip logic sequence was updated based on the random numbers list (i.e., between 12:00 and 12:15 all participants who chose 1 might be redirected to the regular exercise target whereas between 12:16 12:30 choosing 1 might redirect participants to the active living target).

5 Table 1 Demographic Characteristics for the Five Treatment Conditions Variable RE AL EE NE CO Statistic p-level (n = 39) (n = 37) (n = 41) (n = 46) (n = 39) Age in years (SD) (15.35) (15.49) (10.98) (11.76) (13.91) F(4, 197) = Range Parental status Has children 48.7% 48.6% 63.4% 63.0% 64.1% χ 2 (4, N = 202) = 4.26 Does not have children 51.3% 51.4% 36.6% 37.0% 35.9% Pregnancy status Currently pregnant 2.6% 0.0% 9.8% 4.4% 5.1% χ 2 (4, N = 202) = Not pregnant 97.4% 100.0% 90.2% 95.6% 94.9% Ethnicity Caucasian 94.9% 86.5% 95.1% 89.1% 91.1% χ 2 (4, N = 202) = Other (African American, Aboriginal, Hispanic, East Indian, or Asian) Marital status 5.1% 13.5% 4.9% 10.9% 8.9% Married/common law 59.0% 67.6% 73.2% 67.4% 74.4% χ 2 (8, N = 202) = 7.74 Single/separated 33.3% 32.4% 26.8% 26.1% 20.5% Other 7.7% 0.0% 0.0% 6.5% 5.1% Annual household income Under $25, % 13.5% 9.8% 4.3% 17.9% χ 2 (28, N = 202) = $25,000-$40, % 10.8% 9.8% 10.9% 10.3% (continued) 227

6 Table 1 (continued) Variable RE AL EE NE CO Statistic p-level (n = 39) (n = 37) (n = 41) (n = 46) (n = 39) $40,000-$60, % 13.5% 22.% 8.7% 20.5% $60,000-$80, % 13.5% 14.6% 19.6% 7.7% $80,000-$100, % 13.5% 19.5% 23.9% 7.7% $100,000-$150, % 27.0% 12.2% 13.0% 12.8% $150, % 2.7% 12.2% 2.2% 7.7% Prefer not to answer 5.1% 5.4% 0.0% 17.4% 15.4% Employment status Employed full time 38.5% 45.9% 51.2% 52.2% 46.2% χ 2 (24, N = 202) = Employed part time 12.8% 10.8% 14.6% 19.6% 15.4% Self-employed 0.0% 2.7% 9.8% 7.7% 7.7% Unemployed/looking for work 5.1% 0.0% 2.7% 4.5% 2.6% Stay at home mother 7.7% 10.8% 9.8% 9.4% 12.8% Student 23.1% 16.2% 7.3% 10.4% 5.1% Retired 12.8% 13.5% 4.9% 10.3% 10.3% Education level achieved Graduate/professional degree 23.1% 43.2% 36.6% 37.0% 30.8% χ 2 (16, N = 202) = Bachelor s 35.9% 35.1% 29.3% 30.4% 33.3% College or technical training 20.5% 18.9% 26.8% 17.4% 23.1% Secondary school diploma 15.4% 2.7% 7.3% 13.0% 7.7% Some secondary school 5.1% 0.0% 0.0% 2.2% 5.1% Self-reported exercise (weekly (23.35) (20.58) (24.73) (22.66) (16.72) F(4, 197) = activity score) (SD) Note: AL = active living; CO = control; EE = excessive exercise; NE = nonexerciser. RE = regular exerciser. 228

7 Exercise Stereotype Pregnancy 229 Once assigned to a particular condition, participants read one of five descriptions of a pregnant female target. The description featured Jennifer, a fictitious pregnant woman whose age, stage of pregnancy, marital status, and employment situation were all based on mean demographic characteristics from previous research with pregnant women living in the same geographical district (Gaston & Prapavessis, 2009). The information about the targets exercise level was manipulated in each of the five descriptions and the target types were regular exerciser, active living, excessive exercise, nonexerciser, and control. Both the regular exerciser and the active living targets were written in accordance with Canadian guidelines for exercise during pregnancy (Davies et al., 2003). All exercise descriptions were adapted from the Martin Ginis et al. (2003) study examining the positive exerciser stereotype and can be found in Table 2. After reading the description, participants completed a manipulation check to ensure their comprehension of the relevant and controlled information. Subjects were asked to indicate the mother s age (openended), employment status (yes/no), and activity level (very active, moderately active, not at all active, or can t tell). All participants answered these questions Table 2 Target Descriptions for the Five Experimental Conditions Target Type Regular exerciser Description Jennifer is 27 years old and 25 weeks into her first pregnancy. She is of average height and weight. She has brown eyes and straight dark hair that is cut to her shoulders. She and her partner Rob both work outside the home. Jennifer exercises regularly, working out at the gym 3 or 4 days each week. Her exercise program consists of jogging, fitness classes, and some weight training. In her spare time, she listens to music, reads, watches TV, or gets together with her friends. Active living Jennifer lives 3 km away from her job and either walks to work (45 min) or rides her bike (20 min). She never takes the elevator; she always climbs the stairs. Excessive exerciser Nonexerciser Control Jennifer exercises every day for two hours or more, rain or shine. Even when she is feeling sick or injured, she works out by jogging, attending fitness classes and weight training. Jennifer gets upset if for some reason she is unable to exercise. Jennifer is physically inactive and does not participate in exercise activities such as jogging, fitness classes or weight training. Jennifer is 27 years old and 25 weeks into her first pregnancy. She is of average height and weight. She has brown eyes and straight dark hair that is cut to her shoulders. She and her partner Rob both work outside the home. In her spare time, she listens to music, reads, watches TV, or gets together with her friends. Note. The descriptions for the active living, excessive exerciser, and nonexerciser were identical to the description of the regular exerciser except for the sentences in italics, which were replaced by the descriptions provided in the table. No information concerning exercise habits was provided for the control target.

8 230 Gaston, Cramp, and Prapavessis correctly. Participants were then asked to use their imagination to create a vivid image of the target and then to rate her on 12 variables measuring personality attributes and 8 variables measuring physical attributes. Participants were then debriefed about the purpose of the study and given the opportunity to reconsent. Primary Outcome Measures Personality Ratings. Twelve personality dimensions were assessed using bipolar adjectives (see Table 3 for a comprehensive list of items). All items were rated on a 9-point scale and a higher mean score reflects a more positive evaluation (i.e., mean = 1, kind = 9). To remain consistent and produce comparable results, these dimensions were the same as those used in the Martin Ginis et al. studies. According to Martin et al. (2000), these 12 attributes were selected based on evidence that information about an individual s weight and body build influenced judgments of these key attributes. Physical Ratings. The target was also rated on eight physical attributes (see Table 4 for a comprehensive list of items). The same 9-point differential scale was used, and once again, the ratings were the same as those used by Martin et al. (2000) who originally borrowed them from previous work on body build, weight, and judgments about a target s physical attributes. Demographic Measures. Several demographic variables were assessed including age, parental and pregnancy status, marital status, ethnicity, education, employment, and income (see Table 1). In addition, self-reported exercise was also assessed. The Godin Leisure-Time Exercise Questionnaire (GLTEQ; Godin & Shephard, 1985), which yields a total weekly activity score, was used to assess participants activity level. Because previous research indicated that the effect of exercise on impression formation is pervasive and irrespective of raters own exercise status (Rodgers et al., 2009), self-reported exercise was used solely to ensure uniformity between experimental groups. Statistical Analysis The data were checked for normality and linearity and the results were satisfactory. There were no missing values. One-way ANOVA and chi-square procedures were used to ensure homogeneity between treatment groups on demographic characteristics and baseline exercise. Personality and physical dimension ratings were analyzed separately using a multivariate analysis of variance (MANOVA) design. Personality or physical dimension ratings were entered as the dependent variable and target group was entered as the independent variable. Because Box s M tests revealed significant differences in the covariance matrices of the dependent variables (ps <.001), Pillai s trace was used to test for significant multivariate effects. Significant multivariate effects were followed with post hoc analysis for which the alpha was adjusted using Tamhane s T2 tests, an appropriate procedure for comparing groups with unequal variances.

9 Table 3 Descriptive Statistics for Personality Attribute Ratings by Experimental Group Personality Attributes RE (n = 39) AL (n = 37) EE (n = 41) NE (n = 46) CO (n = 39) Mean kind* 6.38ab (.98) 6.43ab (1.28) 5.83a (1.36) 6.46ab (1.15) 6.61b (1.02) Few many friends* 6.90a (1.64) 6.68a (1.20) 5.76b (1.51) 6.30ab (1.52) 6.51ab (1.28) Lazy works hard* 7.38a (.91) 7.08ac (1.28) 7.41a (1.44) 5.80b (1.20) 6.30bc (1.10) Afraid brave* 6.61a (1.22) 6.51a (1.48) 6.71a (1.75) 5.60b (1.22) 5.64b (.87) Unintelligent intelligent 7.12a (1.16) 6.95a (1.40) 6.78a (1.26) 6.46a (1.38) 6.46a (1.10) Sloppy neat* 6.87ab (1.40) 6.54abc (1.24) 7.17a (1.28) 6.24bc (1.39) 5.82c (1.30) Sad happy* 7.41a (1.14) 7.27a (1.22) 6.17b (1.54) 6.37b (1.25) 6.38b (1.21) Lacking self-confidence confident* 7.44a (1.14) 7.48a (1.10) 6.05b (2.07) 6.08b (1.49) 6.26b (.99) Lacking self-control has self-control* 7.51a (1.17) 7.48a (1.10) 7.73a (1.43) 6.00b (1.41) 5.97b (1.04) Unsociable social 7.28a (1.26) 7.19a (.94) 6.39a (1.56) 6.76a (1.43) 6.85a (1.48) Dependent independent* 7.08 (1.42)ab 7.35 (1.21)a 7.24 (1.55)a 6.26 (1.54)b 6.28 (1.26)b Not friendly friendly* 7.28 (1.26)a 6.81 (1.35)ab 6.27 (1.34)b 6.52 (1.35)ab 6.89 (1.02)ab Note. Standard deviations are in parentheses. *Indicates that post hoc analysis revealed significant differences between groups for that variable at the p <.05 level. Values in the same row that do not share a common subscript are significantly different at the p <.05 level. Higher mean score reflects a score closer to the anchor on the right side of each attribute pair. All ratings are on a 9-point scale. AL = active living; CO = control; EE = excessive exercise; NE = nonexerciser; RE = regular exerciser. 231

10 Table 4 Descriptive Statistics for Physical Attribute Ratings by Experimental Group Physical Attributes RE (n = 39) AL (n = 37) EE (n = 41) NE (n = 46) CO (n = 39) Ugly good looking 6.64a (1.44) 6.57a (1.37) 6.71a (1.19) 6.09a (1.11) 6.05a (1.05) Sexually unattractive sexually attractive* 6.74a (1.27) 6.62ac (1.27) 6.85a (1.06) 5.67b (1.40) 5.92bc (.90) Underweight overweight* 5.05ab (1.38) 5.11a (.74) 4.24b (1.61) 5.83c (.93) 5.41ac (.78) Scrawny muscular* 6.72a (.99) 6.00b (.92) 7.32a (1.15) 5.13c (1.00) 4.95c (.60) Sick healthy* 7.33a (1.13) 7.24a (1.23) 7.05a (1.38) 5.85b (1.35) 5.60b (.94) Has unattractive figure has attractive figure* 7.03a (1.22) 6.76a (1.28) 7.27a (1.22) 5.63b (1.29) 5.67b (.92) Unfit fit* 7.59a (1.02) 7.16a (1.14) 8.39b (.97) 4.36c (1.63) 4.97c (1.11) Physically weak physically strong* 7.33ab (1.19) 7.10a (1.24) 8.04b (1.12) 5.00c (1.61) 5.18c (.94) Note. Standard deviations are in parentheses. *Indicates that post hoc analysis revealed significant differences between groups for that variable at the p <.05 level. Values in the same row that do not share a common subscript are significantly different at p <.05 level. Higher mean score reflects a score closer to the anchor on the right side of each attribute pair. All ratings are on a 9-point scale. AL = active living; CO = control; EE = excessive exercise; NE = nonexerciser; RE = regular exerciser. 232

11 Exercise Stereotype Pregnancy 233 Group Equivalency Results Results of one-way ANOVA and chi-square procedures indicated that there were no significant differences between groups (ps >.05, see Table 1) and confirmed that our randomization strategy was successful. As a result, it was deemed unnecessary to use demographic variables as covariates in the subsequent analyses. Personality Ratings The main effect for target type was significant, Pillai s Bartlett trace V = 0.84, F(48, 756) = 4.2, p <.001, η 2 =.21. Post hoc analyses revealed significant differences between groups on 10 of the 12 personality attributes (p <.05, see Table 3). Specifically, the regular exerciser and active living targets were rated significantly higher than the nonexerciser and control target on six items and higher than the excessive exerciser on four items. The positive exercise stereotype did extend to the excessive exerciser for several attributes, as this target was found to be significantly harder working, braver, neater, having self-control, and independent than the nonexerciser or control targets. However, the excessive exerciser was also rated as significantly meaner, having fewer friends, and less friendly than at least one other target. In general, the nonexerciser and control targets received the lowest ratings, a few of which extended to the excessive exerciser (sloppy, sad, and lacking selfconfidence). The only exception was kindness, an attribute on which the control group was rated significantly higher than the excessive exerciser. Physical Ratings The main effect for target type was significant, Pillai s Bartlett trace V = 0.88, F(32, 772) = 6.81, p <.001, η 2 =.22. Post hoc analyses revealed significant differences between groups on seven of the eight physical attributes (p <.05, see Table 4). These differences revealed that the positive exercise stereotype did extend to the excessive exerciser for all of these attributes. In general, the three exerciser targets (regular exerciser, active living, and excessive exerciser) received significantly higher ratings on most variables compared with the nonexerciser and control targets. In addition, the excessive exerciser was rated as fitter than both the regular exerciser and active living targets and more physically strong and underweight than the active living target. Discussion This study examined whether the positive exercise stereotype extends to pregnancy. Although all targets were positively evaluated (i.e., all mean scores were above the midpoint of the scale), several significant group differences emerged indicating that pregnant women derive additional self-presentational benefits from exercise. Beyond this generalized conclusion, our findings merit discussion within the context of previous work and healthy exercise behavior during pregnancy.

12 234 Gaston, Cramp, and Prapavessis With regard to personality attributes, it was encouraging to find that both the regular exerciser and the active living target consistently received the highest ratings on the majority of these attributes. Both of these targets engaged in recommended levels of activity during pregnancy and these results indicate that, in general, women have a positive perception of exercise during pregnancy. More specifically, the targets who engaged in healthy levels of activity were perceived more positively than nonexercisers or a target whose exercise status is unknown. Given the adverse consequences of inactivity and the potential for pregnant women to be swayed by the opinions of friends and family, these findings are encouraging. Contrary to our hypothesis, the excessive exerciser did not score significantly lower than the regular exerciser or active living target on all personality attributes. Although the excessive exerciser did receive the lowest score on two personality attributes and shared three other low scores with the nonexercise and control targets, this target still scored higher than the nonexerciser and control targets on five attributes. Although these mixed results provide partial evidence that the positive exercise stereotype extends during pregnancy to excessive exercisers, it is important to note this target was rated positively only on attributes related to being efficient, in control, and having a strong work ethic (neat, brave, having self-control, hard working, and being independent). Conversely, the excessive exerciser had the lowest ratings on interpersonal characteristics and attributes associated with personal satisfaction. Specifically, this target was perceived to be the meanest, to have the fewest friends, to be the saddest, least self-confident, and least friendly. Taken together, these results indicate that our participants perceived the excessive exercise target to be more driven and determined, but at the same time personally unhappy and socially inept. Although these findings are largely in line with those of previous studies describing a nonpregnant female target (Lindwall & Martin Ginis, 2006; Martin Ginis et al. 2003), several differences emerge when individual attributes are examined. For example, while we found that the excessive exerciser was rated as more independent and braver than the nonexerciser and control targets, these three targets received similar ratings in the Lindwall and Martin Ginis and Martin Ginis et al. studies. The pregnant status of our target combined with participant differences may, at least in part, account for these differences. For example, it is well known that exercise becomes increasingly difficult during pregnancy, and engaging in the level of exercise equal to that of the excessive exerciser would be especially difficult and out of the ordinary. Taken together, these factors may have led to our participants tendency to rate this target as brave and independent. In addition, it should not be overlooked that both Lindwall and Martin Ginis and Martin Ginis et al. used an undergraduate population while the adult women in our study ranged from 19 to 79 years of age. It is likely that given the older age of the majority of our participants, engaging in such high levels of activity would seem particularly difficult and nontraditional. Excessive exercise during pregnancy is not recommended, and these findings are difficult to interpret without taking into account each individual s unique value system. For this reason, it is likely that these confounding results have different implications for each woman. For women who may already have a strong exerciser identity and who wish to appear independent and driven, self-presentational motives may lead them to excessive exercise behavior during pregnancy. On the other hand,

13 Exercise Stereotype Pregnancy 235 self-presentational motives may help those women who value being perceived as personally fulfilled and a good friend to avoid unhealthy levels of activity. Furthermore, while previous research has produced mixed results with respect to physical attributes, our findings are consistent with Martin Ginis et al. (2003) and indicate that a positive exercise stereotype does extend to an excessive exercise target. This may not be surprising given the fact that several of the dimensions were directly related to physical fitness (fitness, strength, muscularity, and body weight). This tenet is further supported by the fact that the excessive exerciser was perceived as fitter than the regular exerciser and active living targets and stronger and more muscular than the active living target. However, it is encouraging to note that for the majority of physical attributes, the excessive exerciser was rated similarly to the regular exerciser and active living targets. This indicates that pregnant women can enjoy a host of self-presentation benefits by engaging in healthy levels of exercise. It is also worth noting that almost all mean group scores were above the midpoint of the scale (i.e., greater than 5) for both personality and physical attributes, indicating that all targets were perceived in a positive manner by our participants. Furthermore, although the nonexerciser was perceived less favorably than the three exercise targets on the majority of attributes, no significant differences emerged between the nonexerciser and control targets. In contrast, both Lindwall and Martin Ginis (2006) and Martin Ginis et al. (2003) reported that the nonexerciser was rated significantly lower than the control target with respect to several personality and physical attributes. It appears, therefore, that inactivity during pregnancy does not elicit a terribly negative evaluation. One possible reason for this may be that our participants were ignorant of the importance of exercise during pregnancy. Another possibility is that having experienced pregnancy themselves (i.e., approximately half reported having children), many of our participants may have been reluctant to judge any target harshly. While this is the first study to examine the positive exercise stereotype in the context of pregnancy, several strengths and limitations need to be acknowledged. With regards to our sample, our goal was to obtain a sample of women which would adequately represent the diversity of the female friends and relatives that most pregnant women likely surround themselves with. Although we did succeed as far as age, income, and maternal experiences are concerned, over 90% of our participants were white. Although this is not surprising given the geographical area in which the study was conducted, it is likely that perceptions surrounding health behaviors during pregnancy are culturally dependent. In addition, our sample consisted primarily of highly educated professionals working within the educational sector. It is unknown how sampling such a narrow segment of the population may have biased our results. Future research may wish to include women from a more diverse cultural and socioeconomic background. In addition, our study did not include men. Given the role that husbands/partners play in influencing pregnant women s exercise behavior (Symons Downs & Ulbrecht, 2006), future studies may wish to investigate whether the same results are obtained using a sample of men. Finally, these findings have several real-world implications. Given the importance of regular exercise during pregnancy, it was encouraging to find that the targets described as meeting guidelines or engaging in active living consistently received higher ratings on both physical and personality attributes compared with the nonexerciser and control targets. In light of historically conservative attitudes

14 236 Gaston, Cramp, and Prapavessis toward prenatal exercise, our results represent a significant and highly positive shift in thinking. However, the failure for the nonexerciser to be rated more negatively than the control target indicates that being inactive during pregnancy was not perceived all that negatively. Had our participants been more aware of the protective benefits of exercise during pregnancy, it is likely that that this target would have received the lowest ratings. This finding indicates that continued efforts to increase awareness of the physical and psychological benefits of exercise during pregnancy are necessary. In addition, prenatal health professionals need to remain cognizant of the relationship between social support and health behaviors and encourage women to develop a strong and supportive network of friends and family. References American College of Sports Medicine. (2006). Impact of physical activity during pregnancy and postpartum on chronic disease risk: Roundtable Consensus Statement. Medicine & Science in Sports & Exercise, 38, Brownell, K.D. (1991). Personal responsibility and control over our bodies. When expectations exceeds reality. Health Psychology, 10, Cannella, B.L. (2006). Mediators of the relationship between social support and positive health practices in pregnant women. Nursing Research, 55(6), Carron, A.V., Hausenblas, H.A., & Mack, D. (1996). Social influence and exercise: A metaanalysis. Journal of Sport & Exercise Psychology, 18(1), Chogahara, M. (1999). A multidimensional scale for assessing positive and negative social influences on physical activity in older adults. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 54B(6), S356 S367. Clarke, P., & Gross, H. (2004). Women s behaviour, beliefs and information sources about physical exercise in pregnancy. Midwifery, 20, Cohen, J. (1992). A power primer. Psychological Bulletin, 112, Collins, N.L., Dunkel-Schetter, C., Lobel, M., & Scrimshaw, S.C. (1993). Social support in pregnancy: Psychosocial correlates of birth outcomes and postpartum depression. Journal of Personality and Social Psychology, 65(6), Davies, G.A., Wolfe, L.A., Mottola, M.F., & MacKinnon, C. (2003). Joint SOGC/CSEP clinical practice guideline: exercise in pregnancy and the postpartum period. Canadian Journal of Applied Physiology, 28, Gaston, A., & Cramp, A. (2011). Exercise during pregnancy: A review of patterns and determinants. Journal of Science and Medicine in Sport, 14, Gaston, A., & Prapavessis, H. (2009). Maternal-fetal disease information as a source of exercise motivation during pregnancy. Health Psychology, 28, Godin, G., & Shephard, R. (1985). A simple method to assess exercise behavior in the community. Canadian Journal of Applied Sport Sciences, 10, Hodgins, M. (1992). A person-perception study of the healthy body-healthy mind stereotype. The Irish Journal of Psychology, 13, Leary, M.R., & Kowalski, R.M. (1990). Impression management: A literature review and two-component model. Psychological Bulletin, 107, Lindwall, M., & Martin Ginis, K. (2006). Moving towards a favorable image: The selfpresentational benefits of exercise and physical activity. Scandinavian Journal of Psychology, 47, Lewis, B., Avery, M., Jennings, E., Sherwood, N., Martinson, B., & Crain, A.L. (2008). The effect of exercise during pregnancy on maternal outcomes: Practical implications for practice. American Journal of Lifestyle Medicine, 2,

15 Exercise Stereotype Pregnancy 237 Lutz, R., Linder, D., & Greenwood, M. (2005). Social influences and physical activity. In the Association of Exercise Physiology Study Guide (1-28). Retrieved from exercisephysiologists.com/asep-epcresourcemanual/index.html Martin Ginis, K.A., Latimer, A.E., & Jung, M.E. (2003). No pain no gain? Examining the generalizability of the exercise stereotype to moderately active and excessive active targets. Social Behavior and Personality, 31, Martin, K.A., Sinden, A.R., & Fleming, J.C. (2000). Inactivity may be hazardous to your image: The effects of exercise participation on impression formation. Journal of Sport & Exercise Psychology, 22, Mottola, M.F. (in press). American College of Sports Medicine, Guidelines for Exercise Testing and Prescription, 9th edition. Chapter 8, Exercise Prescription for Healthy Populations and Special Considerations (In Press). (Deborah Riebe, Associate Ed.). Philadelphia, PA: Lippincott, Williams & Wilkins. Phoenix, A., Woollett, A., & Lloyd, E. (1991). Motherhood. Meanings, practices, and ideologies. London: Sage Publications. Poudevigne, M.S., & O Connor, P.J. (2006). A review of physical activity patterns in pregnant women and their relationship to psychological health. Sports Medicine (Auckland, N.Z.), 36, Rodgers, W.M., Hall, C.R., Wilson, P.M., & Berry, T.R. (2009). Do nonexercisers also share the positive exercise stereotype?: An elicitation and comparison of beliefs about exercisers. Journal of Sport & Exercise Psychology, 31, Sarason, I.G., Sarason, B.R., & Shearin, E.N. (1986). Social support as an individual difference variable: Its stability, origins, and relational aspects. Journal of Personality and Social Psychology, 50(4), Sodenstrom, T. (1999). Gymkulturens logik: Om samverkan mellan kropp, gym och samhälle. [The Logic of Gym Culture: On the interplay between the body, gyms, and society]. Unpublished doctoral dissertation, Umeå University, Umeå. Solchany, J., Sligar, K., & Barnard, K. (2002). Promoting Maternal Role Attainment an Attachment During Pregnancy: The Parent-Child Communication Coaching Program. In J.M. Maldonado-Duran (Ed.), Infant and Toddler Mental Health (pp ). Washington, DC: American Psychiatric Press. Statistics Canada. (2010). Women in Non-traditional Occupations and Fields of Study. Symons Downs, D., & Ulbrecht, J.S. (2006). Understanding exercise beliefs and behaviors in women with gestational diabetes mellitus. Diabetes Care, 29, Thornton, P.L., Kieffer, E.C., Salabarria-Pena, Y., Odoms-Young, A., Willis, S.K., Kim, H., et al. (2006). Weight, diet, and physical activity-related beliefs and practices among pregnant and postpartum Latino women: The role of social support. Maternal and Child Health Journal, 10, Urbaniak, G., & Plous, S. (2011). Research Randomizer. Retrieved April 1, 2008 from Weissgerber, T.L, Wolfe, L.A., Davies, G.A.L., & Mottola, M.F. (2006). Exercise in the prevention and treatment of maternal-fetal disease: A review of the literature. Applied Physiology, Nutrition, and Metabolism, 31, White, P., Young, K., & Gillett, J. (1995). Bodywork as a moral imperative: Some critical notes on health and fitness. Society and Leisure, 18, Manuscript submitted: October 4, 2011 Revision accepted: January 15, 2012