Effects of aerobic and circuit training on fitness and body image among women.

Henry, Ruth N. ; Anshel, Mark H. ; Michael, Timothy 等

Research on body image has been abundant in recent years, partly due to its role in the development of eating disorders, such as anorexia nervosa and bulimia nervosa (Cohen & Petrie, 2005). Body image refers to the accuracy of the perception of the person's bodily size and to the thoughts and feelings associated with the individual's view of the body (Cash, 1989). Traditionally, body image has been viewed as a multidimensional construct comprised of two independent components, perception (i.e., size estimation) and attitudes (i.e., body-related affects and cognitions; Cash, 1989, 1994; Rowe, 1999; Shontz, 1969). While the earlier research addressed the perceptual component of body image (e.g., Thompson, 1995), more recent studies have focused on assessment and treatment of body self-image disturbances (Brown, Cash, & Mikulka, 1990; Grant & Cash, 1995; Koff& Bauman, 1997; Tiggeman & Williamson, 2000).

The collective term referring to body dissatisfaction, discrepancies between actual and perceived body size, and negative affect when comparing one's body to perceived societal norms is called "body image disturbance." The greatest source of body image disturbance is sociocultural (Heinberg, 1995). In many cultures, including the U.S., females feel pressured to achieve a near-impossible degree of thinness and to avoid weight gain and obesity (Striegel-Moore, McAvay, & Rodin, 1986). Messages from the mass media are partly responsible (Van den Buick, 2000).

Van den Buick (2000) reported that among adolescents, the degree of watching television correlated negatively with ideal body image; frequent viewers preferred a thinner ideal body than less frequent television viewers. In another study, Harrison and Cantor (1997) found that the amount of time viewing television significantly predicted overall body dissatisfaction. The print media is not exempt from its influence on body image. Harrison and Cantor also reported a significant relationship between reading of fashion magazines and body dissatisfaction. Thus, the concomitant self-inflicted need to achieve the perceived ideal body type has led to increased body dissatisfaction, especially among adolescents and young adults. Apparently, this issue is not new. For example, Rodin, Silberstein, and Striegel-Moore (1985) predicted that dissatisfaction with one's body may one day become the societal norm, a phenomenon they called normative discontent.

Perhaps the most potentially harmful by-product of negative body image is increased risk of eating disorders (Thompson, 1995). The media has been identified as the culprit in promoting unrealistic ideals. For example, Thomsen, Weber, and Brown (2001) found a strong correlation between frequency of reading women's health and fitness magazines and the use of unhealthy weight practices among high school girls. Because the female adolescent's self-image is usually interpersonally oriented, post pubertal changes in body shape and weight are particularly stressful (Striegel-Moore, Silberstein, & Rodin, 1986). Body shape and weight become critical determinants of self-esteem in adolescence because interpersonal success is increasingly seen as closely linked to physical attractiveness (Brownell, Rodin, & Wilmore, 1992). According to Johnson, Steinberg, and Lewis (1988), thinness has increasingly been associated with a highly valued personal achievement, demonstrating self-control, autonomy, and success. Conversely, obesity, or the absence of weight control can lead to social discrimination and low self-esteem. Thus, the pursuit of thinness is commonly perceived an action or goal in which young women can obtain favorable social responses thereby enhancing self-esteem.

Distorted body image among females, then, is a very common cause of eating disorders. For example, Overdorf (1991) reported that "a little over half (of 102 high school female U.S. athletes) saw themselves as heavy, while in reality only 3% might be classified on the heavy side" (p. 76). Wardle and Foley (1989) found that young women gave negative appraisals of their bodies, tended to be dissatisfied with their appearance, overestimated their body size, and in general, "felt fat." In addition, Maude, Wertheim, Gibbons, and Szmukler (1993) concluded that female high school students displayed higher body image disturbance and dissatisfaction than male students. According to Bruch (1973), disturbances in body image may predispose an individual to relentless dieting or to lack of recognition of the effects of dieting. Typical causes are low self-esteem, perfectionism, and body dissatisfaction (Reel & Gill, 1996).

Although traditional treatment of body image problems has consisted primarily of cognitive-behavioral therapy (Cash, 1995), more recently, exercise has become more popular as a treatment modality (Finkenherg, DiNucci, McCune, & McCune, 1993; Holmes, Chamberlin, & Young, 1994). Researchers have previously documented the positive changes that occur in physical fitness following either aerobic or strength training. In a study of young females, Kraimer, et al., (2001) found that participants participating in bench step aerobics or a combination of bench step aerobics and resistance training improved in V[O.sup.2] peak, resting heart rate, and percent fat. The participants who incorporated resistance training into their regime also improved in muscular strength and endurance. Several studies have examined the effects of exercise training interventions on body image. For example, Koffand Bauman (1997) found that females who participated in step aerobics, weight-training, or running twice per week for six weeks significantly improved appearance evaluation, fitness evaluation, fitness orientation, and body satisfaction as opposed to the control (non-exercise) group. The researchers did not explore the combination of aerobic and strength/circuit training exercise. In a study of women ages 20 to 30 yrs, Skrinar et al. (1986) assessed the effects of intensive endurance training on body consciousness. Participants engaged in an intensive running program for six to eight weeks in addition to other types of moderate sport-related exercise an average of 3.5 hours per day. At the end of the training program, body competence (i.e., effective body functioning) and private body consciousness (i.e., focus on internal bodily sensations) significantly increased, while public body consciousness (i.e., tendency to be concerned about external appearance of the body) did not markedly change. Again, a strength/circuit training condition was not included in this study, however.

Selected studies have focused on the effects of strength training on body image. For example, Tucker and Maxwell (1992) investigated the effects of weight-training on body image of females participating in a 15-week, 2-days-per-week weight-training program, compared with a non-exercising control group. With pretest scores controlled, the weight-training group had significantly higher posttest scores than controls in both general well-being and body cathexis, defined as the degree of satisfaction a person feels about various parts and processes of the body (Secord & Jurard, 1953). The authors concluded that strength training enhances both body image and overall feelings of well-being. Characteristics of women who experienced the greatest improvements included those who were heavier, shorter, and who were not involved in any other regular physical activity during the program. The researchers surmise that heavier, previously inactive participants were less fit, and therefore had more potential for improvement. In a rare study comparing weight-training with aerobic exercise, Tucker and Mortell (1994) found that middle-aged women who engaged in a home strength training program three times per week for 12 weeks markedly improved body cathexis more than women participating in a walking program of the same frequency and duration. In a more recent study, Ahmed, Hilton, and Pituch (2002) investigated the effects of weight-training on body image in female university students. After 12 weeks of strength training, participants experienced significantly more strength and improved body image despite a slight increase in percent body fat. They also reported improved health and physical fitness.

Williams and Cash (2001) investigated the effects of a six-week circuit weight-training program on college students. Their results showed that even a relatively brief program reduced social physique anxiety and improved appearance evaluation and body dissatisfaction among both males and females. Despite the exclusive use of strength training in this study, the circuit training method they employed may have had aerobic benefits.

Research involving body image changes during an intervention program of both strength and aerobic training programs is rare. Perry et al. (2002) found that high school students participating in a six-month course involving 40-45 minutes per week of vigorous aerobic exercise and 20-30 minutes per week of resistance training did improve in body satisfaction (p<.0009). The researchers did not report the training schedule, that is, the number of days per week, and whether the aerobic and strength training occurred on the same days of the week as opposed to alternative days. The measure of body satisfaction was a change in "real" versus "ideal" figure silhouettes. In addition, the 6-month intervention in this study was lengthy in comparison to more common 8-12 week intervention programs.

Contrary to the results of previous studies, Ford, Puckett, Blessing, and Tucker (1989) found that exercise did not improve body cathexis in college females participating in an 8-week, 3-hours-per-week course in either aerobic dance, jogging for fitness, swimming for fitness, life saving, or weight-training. Although the fitness groups improved in various fitness parameters (i.e., sit-ups and flexiblity tests), none of the activity groups differed from the controls in self-esteem, body cathexis, step test scores, or percent body fat at the time of posttest. Similarly, Anderson, Foster, McGuigan, Seebach, and Porcari (2004) found that self-rated appearance scores in adult men were not altered after 6 weeks of either strength or aerobic exercise.

Although researchers (e.g., Ahmed, Hilton, & Pituch, 2002; Heinberg, 1995; Koff & Bauman, 1997) have found that individuals who exercise have a more positive body image than non-exercisers, others (e.g., Tiggemann & Williamson, 2000) have found that exercisers have worse body image than non-exercisers. In particular, the researchers found that participants who exercised for weight control or improved "tone" had lower body satisfaction, and that exercising for the purposes of health and fitness was related to enhanced body satisfaction. Younger women exercised more for weight control and mood enhancement than did older men and women, who exercised more for health and fitness. These results indicate that exercise motivation may be a mediating factor between exercise and body image.

Strelan, Mehaffey, and Tiggemarm (2003) explored motivates for exercise and its impact on body image among young women. They found that exercise motive was a mediating factor in body image. Individuals who exercised for weight control or to improve body tone and attractiveness did not improve over time in body image or self-esteem. However, exercising for health, fitness, mood, and enjoyment resulted in a positive effect on their body satisfaction and self-esteem. Exercising for appearance was negatively related to body satisfaction (r = -.57), whereas exercising for health fitness reasons or enjoyment/mood was correlated positively with body satisfaction (r = .68 and r = .47, respectively). Exercise for appearance was particularly prevalent in young women, but less so among older women and men.

A descriptive study by Davis et al. (1994) supported the authors' hypothesis that a large percentage of eating-disordered patients had either engaged in excessive exercise (78%) or had been competitive athletes prior to the onset of their disorder (60%). This could provide evidence for the idea that exercise may promote perceptions of an imperfect body for either fitness (physical effectiveness) or appearance (physical attractiveness). Although several researchers have addressed the connection between exercise and body image, most of these studies were descriptive in nature and did not include an exercise intervention. And, apparently the study of combined interval circuit training in changes in body image apparently has not been explored, particularly among females. It is important to determine the exercise regimen that will provide optimal benefits in improving body image and other desirable psychological characteristics.

Thus, the purpose of the current study was to determine the effect of aerobic and strength training on body image among unfit college women. It was hypothesized that aerobic exercise would significantly improve body image, however, a program of interval circuit training would produce even greater improvements in body image than aerobic training alone.

Methods

Participants

The participants (N = 72) were female college students, ranging in ages from 18 to 26 yrs (M = 21.4 yrs., SD = 2.21) who volunteered for the study. Biometric information is included in Table 1. Excluded from the study were students who were currently engaging in vigorous exercise at least twice per week or individuals with self-reported eating disorders.

Materials and Equipment

Body Image Assessment. Before the training period all participants completed the Body Self-Image Questionnaire (BSIQ, Rowe, 1999) to assess nine separate body image components. The BSIQ is a 51-item Likert-type scale with items pertaining to feelings and attitudes about one's body. Responses ranged from 1 (not at all true of myself) to 5 (completely true of myself). A sample item from each subscale and number of items in each subscale is provided in Table 2.

The BSIQ consists of three evaluative subscales: (1) overall appearance evaluation--overall estimation of one's appearance; (2) fatness evaluation-estimation of overall fatness of one's body; (3) health/fitness evaluation--estimation of one's overall physical fitness/health status. The instrument also includes the following subscales which do not involve evaluation: (4) health/fitness influence--the degree to which one's health and fitness influences the way one feels about his or her body; (5) attention to grooming--the amount of time and effort spent in manipulating one's appearance; (6) social dependence--effect of social situations or social acceptance on one's body image; (7) height dissatisfaction--desire to be taller or shorter; (8) negative affect--the negative thoughts and feelings associated with one's body; and (9) investment in ideals--importance of and aspirations to achieve a perfect body. The questionnaire also includes questions concerning amount (days per week) of current participation in vigorous exercise and involvement in sports, as well as demographic questions.

The BSIQ was selected for this study because it has received rigorous psychometric scrutiny (i.e., exploratory and confirmatory factor analysis), and validated on several samples of normal and divergent groups, including individuals with eating disorders and competitive dancers. Internal reliabilities (Cronbach's alpha) of the nine subscales have ranged from a = .78 (social dependence) to .94 (fatness evaluation) in a study by Rowe (1999). In the present study, alphas ranged from .68 to .92 (Md = .88). These reliabilities compare favorably with other body image instruments (Cash, 2000; Cash & Szymanski, 1995; Franzoi & Shields, 1984). Accurate identification of body self-image among the research participants in these studies supports construct validity of this instrument.

Physical Fitness Testing Protocols

Percent body fat. Body density was calculated from the sum of triceps, suprailium, and thigh skinfolds using the generalized equation generated by Jackson, Pollock, and Ward (1980). Body density was converted to percent body fat using the Siri equation (Siri, 1956).

Step test. The Queens College Step Test (Safrit & Wood, 1995) was used to estimate cardiorespiratory endurance. This test was chosen because it has demonstrated acceptable validity and reliability, with coefficients of .75 and .92, respectively, for college women and an SEE of 2.9 ml/[kg.sup.-1]/[min.sup.-1] (McArdle, Katch, Pechar, Jacobson, and Ruck, 1972). The test protocol for women consists of 3 minutes of stepping on a 16 1/4" bench at a cadence of 22 step cycles per minute after 15 seconds of practicing the tempo with the metronome. Recovery heart rate is recorded via heart rate monitor 15 seconds after the test concludes. V[O.sub.2] max is predicted from the pulse count using the following regression equation (McArdle, Katch, & Katch, 1991): Predicted V[O.sub.2] max = 65.81 - (0.1847 x post-exercise heart rate).

Bench press test. The YMCA Bench Press Test (Golding, Myers, & Sinning, 1982) was the tool used to assess muscular strength and endurance. The protocol requires a female subject to lift a 35-1b. barbell from the chest position to full arm extension at a cadence of 60 lift cycles per minute, with cadence set by a metronome. The test is completed when the participant can no longer lift the bar, or can no longer lift at the specified tempo. This test was chosen because it is a submaximal test; it was believed to be safer for participants since no maximal contractions are required. Validity and reliability data have not been published for the YMCA test; however, submaximal bench press tests in general have been found to have face validity as an endurance measure, and concurrent validity as a strength measure. Submaximal bench press tests have a high correlation with the maximum weight lifted in one repetition (r >.90). Reliability can be expected to be high if examinees are motivated to achieve maximal performance (Safrit & Wood, 1995).

Procedures

All procedures for both testing and training sessions were approved by the university Institutional Review Board. For the pretest, participants completed the BSIQ first. The BSIQ was administered in a group setting, with each of the three groups receiving identical standardized instructions. Within these instructions, participants were told that they would be providing information for a research study, and that they would be asked to answer questions concerning feelings about their body. Because questionnaires were coded by number anonymity was assured. Participants were also assured that they were under no obligation to complete the process if they should decide not to continue.

During the same group session, participants completed an informed consent for both testing and training protocols, as well as a PAR-Q (American College of Sports Medicine, 2000), a physical activity readiness questionnaire in which the participant identifies past or current symptoms that occur during exercise, as well as diagnoses of cardiorespiratory disease. The PAR-Q was employed to identify health problems that would make testing or training unsafe for the participants.

After completion of the BSIQ, each subject participated in the physical fitness evaluation for body composition, cardiorespiratory endurance, and muscular strength/endurance. Each participant performed the three tests in the same order, and tests were explained and administered to each subject by the same tester. Tests were performed individually, except for the step test, in which three participants were tested at once. Each participant performed step and bench press tests on one occasion, whereas skinfolds consisted of the mean of three recorded readings. Room temperature of the testing laboratory was maintained at 70[degrees]F.

At the conclusion of the 12-week training program, all participants were retested for body image and physical fitness, using identical testing procedures as in the pretest. For each participant, order, time of day, and the examiner were consistent on both the pretest and posttest.

Interventions

After pre-test procedures were completed, participants were non-randomly assigned to one of two experimental groups, aerobic exercise (n = 23) or interval circuit training (n = 28), or to a control (non-exercise) group (n = 21). Group assignments were based on their class and work schedules; they did not select nor choose their mode of exercise. The aerobic and circuit training Classes were scheduled at 2:30 p.m. and 3:30 p.m., respectively. The participants did not know which class would meet at which time when they selected their class time, thereby controlling for a selection effect.

Exercise classes met three 50-min. sessions for each of 12 weeks. As reported in the posttest BSIQ, the participants did not engage in vigorous exercise beyond these class periods. The class sessions for both training groups were identical in duration, and the same instructor taught all classes. Finally, the type of music to which all exercisers were exposed was upbeat.

Aerobic Exercise

The aerobic exercise group completed 12 weeks of step aerobics, three times per week. Each session was 50 minutes in duration, with 35 minutes per session devoted entirely to steady state aerobic exercise. The remainder of the class period consisted of warm-up, cool-down, and limited (3-4 minutes) muscular endurance exercises for the abdominals. Participants were urged to adjust exercise intensity to maintain a heart rate of 60-90% of their age-predicted maximum, and recorded their heart rates on a chart at least once per class session to insure that they were exercising at the prescribed intensity. The instructor demonstrated variations in step exercise to accommodate for individual differences in fitness levels; therefore, all participants were able to exercise within the recommended intensity. Although abdominal crunches were included at the end of the step sessions, no other resistance exercises were performed during the classes.

Interval Circuit training

Because the degree of compliance in individual workouts in the weight room is difficult to monitor accurately, this study was designed to facilitate muscular strength/endurance training in a class setting, rather than the traditional individual resistance training approach conducted in a weight room. The interval circuit training group participated in a class that used a circuit of intervals combining aerobic training, anaerobic training, and muscular strength/ endurance training. The circuit involved the use of bench-stepping, jump rope, cardio-boxing with training gloves, partner exercises with 3-kg medicine balls, various lifts with 8-lb. dumbbells, body resistance exercises, and agility drills. Each training session was 50 minutes in duration, including warm-up and cool-down to pre-exercise levels. Although the instructor encouraged participants to push themselves to train at a high level, accommodations for individual fitness levels were made. Although all participants did not exercise at the same absolute intensity, they were urged to attain the same relative intensity, that is, 60-90% of their age-predicted maximum heart rate.

Control Group

The control group was asked to maintain their "usual" exercise level, which was relatively low, during the 12 weeks. Participants reported a mean of 3.38 days of low to moderate aerobic exercise per week for the pretest, and 3.24 days/week for the posttest resulting in no significant change in their estimated V[O.sub.2] max. (p> .05). The pretest mean (exercise days/week) for all three groups was statistically similar (i.e., no significant difference: p =.39). The pretest V[O.sub.2] max for all groups place them at the 50th percentile for 20-29-year-old women.

Protocol adherence was maintained in each of the experimental groups by participants who recorded their heart rate on three occasions during each class session. Heart rate was also monitored by an assistant instructor during class to insure that exercise intensity was within the recommended range to promote cardiovascular fitness (i.e., 65-80% of age-predicted maximum heart rate). Participants whose heart rate was below the prescribed rate at any time during the class were verbally encouraged to increase exercise intensity. Attendance was recorded at each class session.

Data Analysis

Because assignment to groups was nonrandom, it was necessary to show that groups were equal in fitness and body image before the training program began. For each physical fitness and body image variable, pretest scores were analyzed with one-way analyses of variance (ANOVA) to detect pre-existing group differences.

Two 3 (groups) x 2 (time) repeated measures MANOVAs were performed on the three physical fitness variables, and the nine body image variables to detect whether the three groups differed in rates of change in fitness and body image during the 12 weeks. The main effects for group were not of interest; only the main effect for time and the interaction of group and time were analyzed. The Wilkes Lambda statistic, transformed to F, was used to test for significance.

Assuming significant MANOVA results, separate univariate 3 x 2 ANOVAs for each physical fitness and body image variable were performed to determine which variables demonstrated a significant main effect for time or group x time interaction. The familywise a level of .05 was held constant for each univariate ANOVA by using a modified Bonferroni adjustment recommended by Keppel (1991) in cases where the number of comparisons exceeds [df.sub.among]. The adjustment to a for each comparison is calculated as: a per comparison = ([df.sub.among] * familywise [alpha]) / number of comparisons. A modified Tukey post hoe test was performed for each variable with a significant group x time interaction to determine which groups differed from each other at the time of posttest, after adjusting for pretest differences. The post hoc method used for this analysis (Hinkle, Wiersma, & Jurs, 1998) adjusts the Tukey HSD, allowing for use of the pre-training scores as covariates, thereby allowing pairwise comparisons of adjusted means (1). For those variables with a significant main effect for time, with no significant group x time interaction, a post hoe test was performed to determine which groups improved significantly over time.

Results

The rate of compliance in both experimental groups was high, with 90.2% compliance in the aerobics/strength circuit group and 91.2% compliance in the aerobics group. Compliance was defined as the number of classes attended dived by the total number of classes (n = 35) x 100. Control participants reported on the BSIQ that they did not increase in exercise frequency during the 12 weeks.

Physical Fitness Variables

Pretest ANOVA for each physical fitness variable confirmed that no significant differences existed among the three groups in any of three variables (p>.05) before the training program began. Table 3 depicts the pretest and posttest means and standard deviations for each variable. The 3 x 2 repeated measures MANOVA for physical fitness variables yielded a significant main effect for time (F(3,64) = 39.33, p<.001) and for time x group interaction (F(6,128) = 12.99, p<.001). As a follow-up to the significant interaction, univariate ANOVA was performed on each of the three fitness variables, with the significance level adjusted to ct = .033 to control for type I error. Univariate ANOVAs on the three fitness variables also yielded significant time x group interactions (Table 4) for all variables, indicating that the there was a significant difference among the three groups in amount of change during the training program. Tukey post hoc analyses indicated that the aerobics/strength training group had the greatest improvement in percent body fat (aerobics/strength < aerobics < control), with significant posttest differences between the interval circuit training group and control group (Q = 9.33, p<.001), between the interval circuit training group and the aerobics group (Q = 4.29, p = .013), and between the aerobics group and the control group (Q = 4.92, p = .004).

For the VO2 max results, the interval circuit training group was significantly higher than the control group at posttest (Q = 4.98; p = .002). No other pairwise comparisons were significant for VO2 max. In strength, the interval circuit group performed a significantly greater number of bench press repetitions at posttest than both the control group (Q = 7.00; p<.001) and the aerobics group (Q = 4.12, p = .016). No other pairwise comparisons were significant for the bench press. In the post hoc analyses, adjusted means were used as posttest means, taking into account the effect of pretest scores. (1)

Body Image Variables

ANOVA conducted on pretest scores revealed that no significant differences (p>.05) existed among the three groups in any of the nine body image variables. Table 3 presents the pretest and posttest scores for each body image subscale. A 3 (group) x 2 (time) MANOVA with repeated measures on the last factor yielded a significant main effect for time (F (9,61) = 3.54, p = .001) and for time x group interaction (F(18,122) = 2.02; p = .013). The group x time interaction indicates that the three groups differed on changes in body image during the training program. Univariate analyses of variance were then performed to detect which physical fitness variables were responsible for the significant interaction in the overall MANOVA. A modified Bonferroni adjustment held the familywise type I error rate constant (.05) by adjusting the per comparison criterion to ct = .011. In the follow-up univariate analyses (Table 5), the interaction for group x time was significant for the body image variables of health/fitness evaluation (F(2,69) = 5.08, p = .009), and negative affect (F(1,69) = 6.67; p = .002). Tukey pairwise comparisons were computed to detect the location of these differences. The aerobics/strength training group had lower posttest scores than the control group for negative affect (Q = 4.11, p = .013), and better posttest scores than the aerobics group for health/fitness evaluation (Q = 3.91, p = .019).

For overall appearance evaluation and health/fitness influence, group x time interaction was not significant (p> .05), however, the main effect for time was significant (F(1,69) = 6.78, p = .011; F(1,69) = 10.52, p = .002, respectively). Post hoc analyses of these variables indicated that the interval circuit group improved significantly in overall appearance evaluation and health/fitness influence, while the aerobics group improved in health/fitness influence.

Discussion

The purpose of this study was to determine the effect of aerobic training and interval circuit training on changes in body image among unfit college women. It was hypothesized that the group experiencing the combined aerobic and strength conditioning program would result in significantly superior body image and selected fitness parameters as opposed to the aerobic only and control groups. The results supported our predictions.

With respect to fitness variables, the interval circuit group had a greater VO2 max at the end of the training program than either of the other two groups. The success of the circuit training program in the present study was likely due to the interval training utilized for the combination treatment group. Elite endurance athletes have trained using intervals for years, and often attribute their success to interval training (McArdle, Katch, & Katch, 1991). Because intense exercise intervals are interspersed with rest intervals or intervals of less intense exercise, a larger amount of total work can be accomplished than with steady-state aerobic exercise that is performed continually.

Although the interval circuit training group demonstrated the greatest improvement in strength, the aerobics group also showed a modest increase in strength. The strength increase in the aerobics group was likely due to the continuous exaggerated upper body movements utilized in the aerobic step classes, because other upper-body exercises during the aerobics classes were minimal. The participants did not hold hand weights during the step training, and no pushups or other resistance exercises were included in the class sessions.

The interval circuit training group had a lower percent body fat at posttest than either of the other two groups, and the aerobics group had a lower percent fat than the control group. While these results indicate that both training programs were effective in lowing percent body fat, the combination program was the most effective. This result is not surprising, however, because the increased muscle mass that doubtless accompanied the increase in strength would have altered body composition in the direction of more muscle mass and less fat.

The results of this study were in agreement with other previous findings. For example, Tucker and Maxwell (1992) found the body cathexis scores of females participating in strength training to be greater than those of a control group. Ahmed, Hilton, and Pituch (2002) also found strength training to be beneficial in improving body image in young females. Results of the current study lend support to these findings in that the aerobic plus strength group had a significantly better positive body image scores in several of the body image subscales than controls. Tucker and Mortell (1994) also found that weight training improved body image more effectively than walking.

Although the aerobic exercise in the current study was more vigorous than walking, the interval circuit training was still more effective than the aerobics-only program for improving body image. It should be noted that the inclusion of strength training for this group actually decreased the number of minutes dedicated to aerobic training, with positive results for both physical fitness and body image.

The findings of the present study are also important because the strength training program was introduced into the training regimen without adding extra time to the overall exercise program. Time devoted to strength training replaced time for aerobic training, thereby using an interval training format. Since most exercisers are eager to get the best fitness program possible for limited time investment, the evidence that a combination program can improve both fitness and body image is noteworthy.

It would be reasonable to suggest that the body image of the participants in the current study may have improved because they felt that they were taking positive steps to reshape their bodies. However, changes in body image were accompanied by measurable improvements in fitness. Whether exercise produces increased self-efficacy, thereby improving all aspects of self-esteem, including body image is open to speculation. While the results of this study do not prove that an increase in physical fitness causes body image to improve, the present results do reveal that the improvements attained after the exercise program were both psychological and physical.

Exercise improves female body self-image; however, a woman's ideal body image continues to shift toward a thinner standard (Davis, 1997). Thus, even improved body image due to training may be transient. The woman is briefly more satisfied with her body until she "raises the standard" and may become less satisfied with her body again. Further research is needed that follows female body image over a period of years.

Davis and Fox (1993) explored the association between excessive exercise and weight preoccupation in women, theorizing that those who exercise excessively share characteristics with those who are preoccupied with their weight. The results did not support this theory. However, they found that excessive exercisers reported greater body satisfaction and were less emotionally reactive (i.e., neurotic) than non-exercisers, while weight preoccupation was associated with less body satisfaction and more neuroticism. Our results indicated that women who engage in an extensive, but "normal" exercise regimens do not experience low body image.

There were selected limitations in this study. For example, participants were not assigned to exercise groups on a strictly random basis. It was imperative that the student's class schedule and the exercise program schedule were compatible; therefore, participants were assigned to groups based on their class schedules. This non-random assignment strategy may have compromised internal validity.

Nevertheless, pretest ANOVAs confirmed similar levels of physical fitness and body image prior to the intervention. A final limitation was the control was not fully sedentary. They engaged in either a non-aerobic or low intense exercise program. However, significant group differences found in our results confirmed that exercise of greater intensity resulting in improved cardiovascular fitness results in desired changes in body image.

As indicated earlier, although several studies have compared changes in body image after different types of aerobic training programs and/or strength training programs, the effect of the combination of aerobic and strength training on body image has been virtually ignored by researchers, at least among college women. Researchers (e.g., Skrinar, et al., 1986; Tucker & Maxwell, 1992; Koff & Bauman, 1997) have documented that changes in physical fitness are related to improvements in body image.

In summary, then, participants who trained for 12 weeks using a combination of aerobic, anaerobic, and strength circuit training improved in several components of body image and on selected features of physical fitness. Greater improvements were made by the combination group than by those who trained only aerobically. The control group did not improve during the 12 weeks in body image or in physical fitness. Taken together, it appears that exercise has a positive effect on the body image of college-aged women. Although aerobic exercise has been found to be beneficial in improving some components of body image, apparently a circuit training program of aerobic, anaerobic, and strength intervals may elicit more positive changes in both physical and psychological parameters than no regular exercise or aerobic exercise only.

Future research is needed comparing the effect of aerobics/strength circuit training, as opposed to strength training only on changes in body image as a function of age and gender. In particular, intervention studies are needed to determine the optimal exercise program duration, and to ascertain whether exercise-induced improvement in body image has long-term as opposed to short-term benefits.

References

Ahmed, C., Hilton, W., and Pituch, K. (2002). Relations of strength training to body image among a sample of female university students. Journal of Strength and Conditioning Research, 16, 645-648.

American College of Sports Medicine (2000). Guidelines for exercise testing and prescription (6th ed.). Philadelphia, PA: Lippincott, Williams, & Wilkins.

Anderson, M. L., Foster, C., McGuigan, M. R., Seebach, El, & Porcari, J. P. (2004) Training vs. body image: does training improve subjective appearance ratings? Journal of Strength and Conditioning Research, 18, 255-259.

Brown, T.A., Cash, T. F., & Mikulka, P.J. (1990). Attitudinal body-image assessment: Factor analysis of the Body-Self Relations Questionnaire. Journal of Personality Assessment, 55, 135-144.

Cash, T. F. (1989). Body-image affect: Gestalt versus summing the parts. Perceptual and Motor Skills, 69, 17-18

Cash, T. F. (1994). Body-image attitudes: Evaluation, investment, and affect. Perceptual and Motor Skills, 78, 1168-1170.

Cash, T. F. (1995). What do you see when you look in the mirror? Helping yourself to a positive body image. New York, NY: Bantam Books.

Cash, T. F. (2000). The MBSRQ Users' Manual (3rd revision). (Available from Old Dominion University, Norfolk, VA23529-0267).

Cash, T. F., & Szyraanski, M. L. (1995). The development and validation of the Body-image Ideals Questionnaire. Journal of Personality Assessment, 63, 466-477.

Cohen, D., & Petrie, T. (2005). An examination of psychosocial correlates of disordered eating among undergraduate women. Sex Roles: A Journal of Research, 52, 29-42.

Davis, C. (1997). Body image, exercise, and eating disorders. In K.R. Fox (Ed.), The physical self: From motivation to well-being (lap 143-174). Champaign, IL: Human Kinetics.

Davis, C., & Fox., J. (1993). Excess exercise and weight preoccupation in women. Addictive Behaviors, 18, 201-211.

Davis, C., Kennedy, S. H., Ravelski, E., & Dionne, M. (1994). The role of physical activity in the development and maintenance of eating disorders. Psychological Medicine, 24, 957-967.

Finkenberg, M. E., DiNucci, J. M., McCune, S. L., & McCune, E. D. (1993). Body esteem and enrollment in classes with different levels of physical activity. Perceptual and Motor Skills, 76, 783-792.

Ford, H. T., Puckett, J. R., Blessing, D. L., & Tucker, L. A. (1989). Effects of selected physical activities on health-related fitness and psychological well-being. Psychological Reports, 74, 203-208.

Franzoi, S. L., & Shields, S. A. (1984). The Body Esteem Scale: Multidimensional structure and sex differences in a college population. Journal of Personality Assessment, 48, 173-178.

Golding, L. A., Myers, C. R., & Sinning, W. E. (1982). The Y's way to physical fitness. Chicago, IL: National Board of the YMCA.

Harrison, K. and Cantor, J. (1997). The relationship between media consumption and eating disorders. Journal of Communication, 7, 40-67

Heinberg, L. J. (1995). Theories of body image disturbance. In J. K. Thompson (Ed.), Body image, eating disorders and obesity: An integrative guide for assessment and treatment. Washington, D.C.: American Psychological Association.

Hinkle, D. E., Wiersma, W., & Jurs, S. G. (1998). Applied statistics for the behavioral sciences. Boston, MA: Houston Mifflin.

Holmes, T., Chamberlin, P., & Young, M. (1994). Relations of exercise to body image and sexual desirability among a sample of university students. Psychological Reports, 74, 920-922.

Jackson, A. S., Pollock, M. L., & Ward, A. (1980). Generalized equations for predicting body density of women. Medicine and Science in Sports, 12, 175-182.

Johnson, A., Steinberg, R., & Lewis, W. (1988). Bulimia. In K. Clark., R. Parr. & W. Castelli (Eds.), Evaluation and management of eating disorders, anorexia, bulimia and obesity (pp. 177-190). Champaign, Illinois: Life Enhancement Publications.

Keppell, G. (1991). Design and analysis: A researcher's handbook. Englewood Cliffs, N.J.: Simon and Schuster.

Koff, E. & Bauman, C. L. (1997). Effects of wellness, fitness, and sport skills programs on body image and lifestyle behaviors. Perceptual and Motor Skills, 84, 555-562.

Kraimer, W. J., Keuning, M., Ratamess, N. A., Volek, J.S., McCormich, M., Bush, J. A., Nindl,, B. C., Gordon, S. E., Mazzetti, S. A., Newton, R. U., Gomex, A L, Wickham, R. B., Rubin, M. R., and Hakkinen, K. (2001). Resistance training combined with bench-step aerobics enhances women's health profile. Medicine and Science in Sports and Exercise, 33, 259-269.

McArdle, W. D., Katch, F. L., Pechar, G. S., Jacobson, L, & Puck, S. (1972). Reliability and interrelationships between maximal oxygen intake, physical work capacity and step-test scores in college women. Medicine and Science in Sports, 4, 182-186.

McArdle, W. D., Katch, V. L., & Katch, F. I., (1991). Exercise physiology: Energy, nutrition, and human performance (3rd ed). Philadelphia, PA: Lea & Febiger.

Overdorf, V. G. (1991). Eating related problems in female athletes. In D.R. Black (Ed.), Eating disorders among athletes (pp. 67-85). Reston, VA: Association s of the American Alliance for Health, Physical Education, Recreation and Dance.

Perry, A. C., Rosenblatt, E. S., Kempner, L., Feldman, B. B., Paolercio, M. A., & Van Bemden, A. L. (2002). The effects of an exercise physiology program on physical fitness variables, body satisfaction, and physiology knowledge. Journal of Strength and Conditioning Research, 16, 219-226.

Rodin, J., Silberstein, L. R., & Striegel-Moore, R. H. (1985). Women and weight: A normative discontent. In T. B. Sonderegger (Ed.), Psychology and gender: Nebraska Symposium on Motivation (pp. 267-307). Lincoln, NE: University of Nebraska Press.

Rowe, D. A. (1999). Development of the body self-image questionnaire. Measurement in Physical Education and Exercise Science, 3, 223-247.

Safrit, J. M., & Wood, T. M. (1995). Measurement in physical education and exercise science (3rd ed.). St. Louis, MO: Mosby.

Salem, S. K., & Elovson, A. C. (1993, August). Importance of ideal body image, self-esteem and depression in females. Paper presented at the Annual Convention of the American Psychological Association, Toronto, Ontario, Canada.

Schuler, P. B., Broxon-Hutcherson, A., Philipp, S. F., Ryan, S., Isosaari, R. M., & Robinson, D. (2004). Body-shape perception in older adults and motivations for exercise. Perceptual and Motor Skills, 98, 1251-1260.

Secord, P. F., & Jourard, S. M. (1953). The appraisal of body-cathexis: Body cathexis and the self. Journal of Consulting Psychology, 17, 343-347.

Shontz, F. C. (1969). Perceptual and cognitive aspects of body experience. New York, NY: Academic Press.

Sift, W. E. (1956). Gross composition of the body. In J. H. Lawrence & C. A. Tobias (Eds.), Advances in biological and medical physics (Vol. 4, pp. 239-280). New York, NY: Academic Press.

Skrinar, G. S., Bullen, B. A., Cheek, J. Mm, McArthur, J. W., & Vaughan, L. K. (1986). Effects of endurance training on body-consciousness in women. Perceptual and Motor Skills, 62, 483-490.

Strelan, P., Mehaffey, & Tiggemann, M. (2003). Self-objectification and esterrm in young women: the mediating role of reasons for exercise. Sex Roles: A Journal of Research, 48, 89-95.

Striegel-Moore, R., McAvay, G., & Rodin, J. (1986). Psychological and behavior correlates of feeling fat in women. International Journal of Eating Disorders, 5, 935-947.

Thompson, J. K. (1995). Body image, eating disorders, and obesity: An integrative guide for assessment and treatment. Washington, D.C.: American Psychological Association.

Thomsen, S. R., Weber, M. M., & Brown, L. B. (2001). The relationship between health and fitness magazine reading and eating-disordered weight-loss methods among high school girls. American Journal of Health Education, 32, 133-138.

Tiggemann, M., & Williamson, S. (2000). The effect of exercise on body satisfaction and self-esteem as a function of gender and age. Sex Roles: A Journal of Research, 44, 119-129.

Tucker, L. A., & Maxwell, K. (1992). Effects of weight-training on the emotional well-being and body image of females: Predictors of greatest benefit. American Journal of Health Promotion, 6, 338-344.

Tucker, L. A., & Mortell, R. (1994). Comparison of the effects of walking and weight-training program on body image in middle-aged women: An experimental study. American Journal of Health Promotion, 8, 34-42.

Van den Buick, J. (2000). Is television bad for your health? Behavior and body image of the adolescent "couch potato." Journal of Youth and Adolescence, 29, 273-281.

Williams, P. A., & Cash, T. F. (2001). Effects of a circuit weight-training program on the body images of college students. The International Journal of Eating Disorders, 30, 75-82.

Wylie, R. C. (1974). The self-concept: A review of methodological considerations and measuring instruments (Rev. ed.). Lincoln, NE: University of Nebraska Press.

Ruth N. Henry

Lipscomb University

Mark H. Anshel

Middle Tennessee State University

Timothy Michael

Western Michigan University

Address correspondence To: Ruth N. Henry, D.A., Department of Kinesiology, Lipscomb University, Nahsville, TN 37204 E-mail: [email protected]

Table 1. Descriptive data and demographics.

 Aerobics Interval Circuit
 (n=23) * (n=28) *

 Mean [+ or -] SD Mean [+ or -] SD

Age (yrs) 19.4 [+ or -] 1.2 19.1 [+ or -] 1.6
Weight (kg) 63.8 [+ or -] 13.2 61.3 [+ or -] 7.3
Height (in) 65.6 [+ or -] 2.2 64.6 [+ or -] 1.9
Body mass index 22.9 [+ or -] 4.3 22.8 [+ or -] 2.5

(kg/m2)

 Control
 (n=21) *

 Mean [+ or -] SD

Age (yrs) 20.1 [+ or -] 1.6
Weight (kg) 63.7 [+ or -] 11.3
Height (in) 66.2 [+ or -] 3.5
Body mass index 22.4 [+ or -] 2.8

(kg/m2)

* One subject withdrew from each group. Sample sizes reflect intact
groups.

Table 2. Example items from the Body Self-Image Questionnaire

Subscale # of Items Example Item

OAE 6 I look good in clothes
FE 9 Parts of my body are fat
HFE 7 My body is strong
HFI 6 I watch carefully what I eat, to maintain
 a health body.
AG 4 I'm usually well dressed
SD 3 I compare my body to people I'm close to
 (friends, relatives, etc.)
HD 3 I wish I were a different height
NA 6 Most days I feel bad about my body
II 7 I particularly notice how much body fat
 other people have

OAE = overall appearance evaluation; FE = fatness evaluation;
AG = attention to grooming; HFE = health/fitness evaluation;
HFI = health/fitness influence; SD = social dependence;
HD = height dissatisfaction; NA = negative affect;
II = investment in ideals

Table 3. Pretest and posttest means and standard deviations for
physical fatness and body image variables.

 Aerobics

 Pretest Posttest
 Mean (SD) Mean (SD)

%Fat 23.54 (6.7) 22.09 (5.6)
[VO.sub.2max] 35.77 (4.7) 37.01 (4.1)
Strength 20.82 (9.2) 26.00 (10.8)
OAE 19.65 (5.7) 20.26 (5.5)
FE 29.91 (8.7) 29.35 (9.3)
AG 16.35 (2.4) 16.43 (2.5)
HFE 24.74 (4.8) 24.43 (5.5)
HFI 23.83 (3.5) 25.00 (3.5)
SD 11.43 (2.1) 11.43 (2.5)
HD 8.35 (4.2) 8.52 (4.0)
NA 18.17 (7.4) 16.52 (7.1)
II 29.70 (4.3) 28.65 (5.0)

 Interval Circuit

 Pretest Posttest
 Mean (SD) Mean (SD)

%Fat 23.02 (5.0) 20.19 (4.7)
[VO.sub.2max] 35.12 (2.9) 37.40 (3.0)
Strength 24.25 (10.2) 33.50 (10.5)
OAE 18.93 (5.3) 21.50 (4.5)
FE 29.21 (9.4) 26.61 (9.9)
AG 15.89 (2.3) 16.21 (2.2)
HFE 22.32 (4.9) 25.68 (4.8)
HFI 23.29 (3.2) 24.96 (3.1)
SD 11.43 (2.1) 10.82 (2.4)
HD 6.89 (3.8) 5.93 (3.2)
NA 17.93 (6.8) 14.29 (6.0)
II 27.93 (5.2) 27.39 (4.7)

 Control

 Pretest Posttest
 Mean (SD) Mean (SD)

%Fat 22.28 (6.1) 22.86 (5.5)
[VO.sub.2max] 36.48 (3.7) 36.30 (3.02)
Strength 26.05 (9.9) 27.10 (11.1)
OAE 21.05 (4.8) 20.95 (3.9)
FE 24.48 (10.1) 25.67 (11.3)
AG 16.14 (2.7) 15.10 (2.9)
HFE 23.76 (5.1) 24.71 (6.8)
HFI 23.38 (5.2) 23.52 (3.7)
SD 10.00 (4.0) 10.10 (3.7)
HD 6.33 (3.9) 6.00 (3.9)
NA 13.29 (7.6) 14.14 (7.1)
II 26.43 (72.) 25.90 (7.4)

[V.sub.O2max] = estimated [V.sub.O2max]; Bench = number of bench press
repetitions; OAE overall appearance evaluation; FE = fatness
evaluation; AG = attention to grooming; HFE = health/fitness
evaluation; HFI = health/fitness influence; SD = social dependence;
HD = height dissatisfaction; NA = negative affect;
II = investment in ideals

Table 4. Results of MANOVA, Univariate ANOVAs, and pairwise comparisons
for physical fitness variables.

 Main Effect (Time) Interaction (Group x Time)

 F p F p
MANOVA
 39.33 <.001 12.99 <.001

Univariate
ANOVAs

Fat 24.08 <.001 * 17.39 <.001 *
VO2 max 10.11 .002 * 7.30 .001 *
Strength 52.63 <.001 * 13.30 <.001 *

 Pairwise Comparisons (v)

 A vs. I/C A vs. C I/C vs. C
MANOVA

Univariate
ANOVAs

Fat .013 (1) .004 (1) <.001 (1)
VO2 max .217 .202 .002 (2)
Strength .016 (3) .125 <.001 (3)

A = Aerobics group; I/C = Interval circuit group; C = Control group
% Fat = percent body fat; [VO.sub.2max] = estimated [VO.sub.2max];
Strength number of bench press repetitions

* significant with modified Bonferroni adjustment (p<.033)

(1 2 3) Significant pairwise comparisons following significant group
x time interaction:

(1) Interval circuit < aerobics < control (p<.05)

(2) Interval circuit > control (p<.05)

(3) Interval circuit > aerobics and control (p<.05)

Table 5. Results of MANOVA, Univariate ANOVAs, and Pairwise Comparisons
for Body Image Variables.

 Main Effect (Time) Interaction (Group x Time)

 F p F p

MANOVA 3.54 .001 2.02 .013

Univariate
ANOVAs

OAE 6.78 .011 (1) 4.33 .017
FE 1.51 .224 4.27 .018
AG .82 .368 3.11 .051
HFE 724 .009 (1) 5.08 .009 *
HFI 10.52 .002 (1,2) 2.13 .126
SD .35 .555 .63 .536
HD 2.49 .118 2.08 .133
NA 8.53 .005 (1,2) 6.68 .002 *
II 3.26 .075 0.19 .825

 Pairwise Comparison (p)

 A vs. I/C A vs. C I/C vs. C

MANOVA

Univariate
ANOVAs

OAE
FE
AG
HFE .0193 (3) .665 .180
HFI
SD
HD
NA .167 .561 .013 (4)
II

A = Aerobics group; I/C = Interval circuit group; C = Control group
OE = overall appearance evaluation; FE = fatness evaluation;
AG = attention to grooming; HFE = health/fitness evaluation;
HFI = health/fitness influence; SD = social dependence; HD = height
dissatisfaction; NA = negative affect; II = investment in ideals
* significant interaction with modified Bonferroni adjustment (p<.011)

Simple main effects for time:

(1) Interval circuit group improvement (p<.001)

(2) Aerobics group improvement (p<.05)

Significant pairwise comparisons in variables with significant group
x time interaction:

(3) Interval circuit > aerobics (p = .019)

(4) Interval circuit < control (p = .013)