This large, cross-sectional study investigated a model of the effects of naturist activity on life satisfaction, mediated by higher body image and self-esteem. Explanations for naturism’s effects on body image (seeing others vs. being seen) were also explored.
Method
Participants and Design
Eight hundred and forty-nine members of the British public were recruited via word-of-mouth and advertisements posted on various Internet forms to take part in this research. Of these participants, 739 (87%) were male, 94 (11%) were female and 16 (.02%) did not identify their gender or reported a non-binary gender. The mean age was 57.19, with a standard deviation of 12.00. Ethnically, 831 (98%) identified as White, 3 (.4%) as East Asian, 2 (.2%) as South Asian, 0 as Black, and 13 (2%) as “other”; 405 (54%) identified themselves as having no religion, 366 (43%) identified as Christian, 0 identified as Muslim, 74 (9%) identified themselves as “other” and 4 (.5%) declined to identify a religion; 702 (83%) identified as heterosexual, 39 (5%) as gay, 94 (11%) as bi-sexual, and 14 (2%) as “other”.
All participants completed the study online using a computer and Qualtrics survey completion software. Participants were told that the study investigated the association between social activities and well-being, but were not made aware of any of the specific hypotheses. All participants indicated whether they had ever taken part in any naturist activities, and if yes, when they began doing so and how frequently they did so. All participants completed measures of body image, self-esteem, and life satisfaction. At the end of the survey participants were thanked and fully debriefed. All those who completed the survey were offered the chance to win £150 in a prize draw. Only participants who completed the questionnaire were included in the prize draw or the analyses.
Measures
To assess participation naturist activities, participants were asked to indicate (yes vs. no) whether they had ever taken part in “in clothes-free activities (i.e., activities in which you have been fully or partially undressed outdoors and/or in the company of other people (other than your close family or romantic partner).” The language of this question deliberately focused on participants’ behaviour and avoided the labels “nudist” and “naturist”. If a participant responded yes to this initial question, they were asked to indicate (using the year and the month) when they had started participating in these activities and how many of these activities they did in a typical year. Subtracting the date at which the participant started taking part in naturist activities from the date at the end of data collection produced a measure of length of time (in years) doing naturist activities.
As both the time spent doing naturist activities and the frequency of these activities were important for this research, the product of the time spent doing naturist activities (in years) and frequency of naturist events per year was used as the index of naturist activity. For example, if a participant reported taking part in naturist activities for 2 years, and typically taking part in 3 naturist activities per, this participant would have an index score of (2 × 3 =) 6. This approach has been successfully used in prior social-psychological research (e.g., Evans-Lacko et al. 2013; Taschler and West 2016; Voci and Hewstone 2003; West and Hewstone 2012; West et al. 2014). Participants who indicated that they had never taken part in any naturist activity were assigned scores of 0 for both the number of years spent taking part in naturist activities and the number of naturist activities per year (thus, they necessarily had index scores of 0 as well).
Positive body image was measured by the 13-item Body Appreciation Scale (Avalos et al. 2005). Using 7-point Likert scales (1 = Strongly Disagree, 7 = Strongly Agree) participants indicated their agreement with the following statements: “I respect my body”, “I do not feel good about my body” (reversed), “On the whole, I am not satisfied with my body” (reversed), “Despite its flaws, I accept my body for what it is”, “I feel that my body has at least some good qualities”, “I take a positive attitude towards my body”, “I am attentive to my body’s needs”, “Despite its imperfections I still like my body”, “My self worth is independent of my body shape or weight”, “I focus a lot energy being concerned with my body shape or weight” (reversed), “My feelings toward my body are positive, for the most part”, “I engage in healthy behaviours to take care of my body”, “I do not allow unrealistic images presented in the media to affect my attitudes toward my body” (α = .84).
Self-esteem was measured with the 10-item Rosenberg Self-Esteem Scale (Rosenberg 1965, more recently used by Ashburn-Nardo et al. 2007). Using 4-point Likert scales (1 = Strongly Disagree, 2 = Disagree, 3 = Agree, 4 = Strongly Agree) participants indicated their agreement with the following statements: “On the whole, I am satisfied with myself”, “At times, I think I am no good at all” (reversed), “I feel that I have a number of good qualities”, “I am able to do things as well as most other people”, “I feel I do not have much to be proud of” (reversed), “I certainly feel useless at times” (reversed), “I feel that I’m a person of worth”, “I wish I could have more respect for myself” (reversed), “All in all, I am inclined to think that I am a failure” (reversed), “I take a positive attitude toward myself” (α = .78).
Overall life satisfaction was measured with the 5-item Satisfaction with Life Scale (Diener et al. 1985). Using 7-point Likert scales (1 = Strongly Disagree, 7 = Strongly Agree) participants indicated their agreement with the following statements: “In most ways my life is close to ideal”, “The conditions of my life are excellent”, “I am satisfied with my life”, “So far, I have gotten the important things I want in life”, “If I could live my life over, I would change almost nothing” (α = .91).
To distract participants from the true purpose of the study, a number of filler items were included concerning relationships between males and females, gender norms and sexual activity in general. The order of presentation of the measures was also counter-balanced across participants. After completing all the measures, participants were thanked and fully debriefed.
Results
Preliminary Analyses
Descriptive statistics can be seen in Table 1. In total 861 people started the survey. Most of the scales used in the survey contained some reversed items. For example, the Rosenberg Self-Esteem Scale contained the items “On the whole I am satisfied with myself” (non-reversed item) and “At times I think I am no good at all” (reversed item). If a participant gave the same response to the reversed and non-reversed items throughout the scale, this was marked as an incorrect completion and this participant’s data were discarded. Twelve participants (10 men and 2 women) were discarded for this reason, leaving 849 participants for analysis.
Table 1 Descriptive statistics of all variables used in Study 1
Overall, most participants (805 or 95%) indicated that they had taken part in naturist activity at least once in their lives; participants also indicated that they took part in a mean of 36.75 naturist events per year. However, this number may be misleadingly high. Fifty per cent of participants took part in 17 or fewer naturist activities per year, suggesting that these participants were not ardent naturists, but rather people who would occasionally take part in clothing-optional activities. Furthermore, the phrasing of the question, which asked participants whether they had ever “been fully or partially undressed outdoors and/or in the company of other people (other than [their] close family or romantic partner)”, was very inclusive, and may have increased the number of positive responses. For example, each topless trip to a beach would count as a separate activity. For those who had spent time taking part in naturist activities, overall they indicated that they started taking part a mean of 23.65 years ago, a value close to the median (24.83).
Mediation Analyses
Differences in the scales used to measure each of the variables could make statistical relationships between them difficult to interpret. For example, self-esteem was measured with a 4-point scale, body image and life satisfaction with 7-point scales, and the index of naturist activity had a minimum of 0 (0 naturist activities per year × 0 years) and a maximum of 21,900 (365 naturist activities per year × 60 years). To increase clarity of presentation, all variables were normalised so that they had a mean of 0 and a standard deviation of 1 (as previously done by Fryer Jr. and Levitt 2004; Strand 2012). These standardized variables are used in the subsequent mediation analyses and correlations between these standardized variables can be seen in Table 2.
Table 2 Correlations between standardised variables used in Study 1
It was hypothesised that naturist activity would predict higher life satisfaction and that this relationship would be mediated in turn by more positive body image and higher self-esteem. These relationships were investigated using PROCCESS macros (Model 6, with a 95% confidence interval based on 10,000 bootstrap samples; Hayes 2009). Age, ethnicity, religion and sexual orientation were included as covariates in these analyses to rule out possible alternative explanations for our findings (and also, in the case of age, because of a possible confound of the measure of naturist activity, which incorporated a measure of time in years). The effects of gender are investigated more explicitly below. Only statistically significant relationships are reported.
The hypothesised mediated relationships were supported by the data; see Fig. 1. Naturist activity predicted more positive body image (b = .08, p = .03), which in turn predicted higher self-esteem (b = .50, p < .0001), and greater life satisfaction (b = .08, p = .01). Higher self-esteem also predicted greater life satisfaction (b = .56, p < .0001). The total indirect effect of naturist activity on life satisfaction was positive and significant; LLCI = .0051, ULCI = .083, point estimate = .038. All other indirect effects of naturist activity on life satisfaction can be seen in Table 3. This model accounted for 37% of the variance in participants’ life satisfaction scores.
Table 3 Mediators of the indirect relationship between naturist activity and life satisfaction in Study 1
Testing the Reversed Model
While the data fit the proposed model well, it is necessary to test the reversed causal relationships. While it is possible that taking part in naturist activity causes improvements in body image, self-esteem, and life satisfaction, it is equally possible that people who are happier with themselves and their bodies are consequently more likely to take part in naturist activities. This reversed model—in which life satisfaction predicted naturist activity, mediated by higher self-esteem and more positive body image—was tested in the same manner as the proposed model: using PROCCESS macros (Model 6, with a 95% confidence interval based on 10,000 bootstrap samples), and including age, ethnicity, religion and sexual orientation as covariates.
The data did not fit the reversed model as well as the original, proposed model. Life satisfaction predicted both higher self-esteem (b = .60, p < .0001) and more positive body image (b = .09, p = .01). Self-esteem also predicted more positive body image (b = .45, p < .0001), but body image failed to predict participation in naturist activities (b = .07, p = .10), as did self-esteem (b = .05, p = .27) and life satisfaction (b = −.04, p = .29). Furthermore, while the total indirect effect of life satisfaction on naturist activities was positive and significant (LLCI = .01, ULCI = 11, point estimate = .06), none of the specific mediated pathways was significant (LLCI = −.007, ULCI = .07, point estimate = .03; LLCI = −.003, ULCI = .05, point estimate = .02; and LLCI = −.0002, ULCI = .02, point estimate = .006).
Effects of Gender
This sample has a large ratio of male participants (739) to female participants (94). Furthermore, the men reported more positive body image than did the women; M = 5.71 versus M = 5.45, t (831) = 2.94, p = .003. It was thus important to verify that the findings applied to women as well as to men. To this end, participants’ gender was investigated as a potential moderator of the relationship between naturist activity and body image using PROCCESS macros (Model 1, 95% C.I. 10,000 bootstrap samples). This moderation was not significant (b
interaction
= .08, p = .29). Furthermore, when only female participants were selected, naturist activity continued to significantly predict positive body image (b = .15, p = .02).
Different Levels of Naturist Activity
It is notable that the proportion of participants in this sample who reported ever taking part in any clothes-free activity is much higher than the proportion reported in previous research (Carr-Gomm 2012; Ipsos-Mori 2011). Even accounting for the inclusive language of the survey, people who frequently take part in naturist activities are likely overrepresented in this sample, which raises questions about the generalizability of the findings to the broader public who do not frequently take part in naturist activities.
Moderation Analyses
To test whether these findings also applied to more representative participants (i.e., who did not frequently take part in naturist activities), all 849 participants were ranked from least participation in naturist activities (rank = 1) to most participation in naturist activities (rank = 849), and this rank was used as a moderator of the relationship between naturist activity and body image using PROCCESS macros (Model 1, 95% C.I. 10,000 bootstrap samples).
This moderation was significant (b
interaction
= −.004, p = .0001). Furthermore, the pattern of results suggested that the association between naturism and positive body image was stronger for more representative participant samples. Examination of the moderated analysis revealed that the relationship between naturist activity and body image was strongest for participants who took part in the least naturist activities (i.e., those with the lowest rank numbers; b = 1.38, p < .0001), weaker for those in the middle of the distribution (b = .87, p < .0001) and weakest for participants who took part in the most naturist activities (i.e., those with the highest rank numbers; b = .36, p = .0001).
Median Split Analyses
This pattern was also reflected in the correlations between naturist activity and body image, and the indirect effect of naturism on life satisfaction via body image and self-esteem. A median split divided participants into those who took part in fewer naturist activities and those who took part in more naturist activities. The subset of participants who took part in fewer naturist activities consisted of 365 men, 47 women, and 11 of unidentified or non-binary gender (mean age = 55.31). In this subset the mean number of naturist activities per year was 9.91, and the median was 6.00; the mean number of years since they first started taking part in naturist activities was 13.84 and the median was 9.08. When using only these participants (n = 423), the correlation between naturist activity and body image was somewhat stronger (r = .12, p = .02), and naturist activity continued to directly predict more positive body image (b = .13, p = .01), and to indirectly predict higher life satisfaction via higher body image and self-esteem (LLCI = .01, ULCI = .08, point estimate = .04).
Using this subset (i.e., those who took part in fewer naturist activities), I also found further support for the specific proposed model of relationships between variables. Using this subset of the participants, the reversed model (i.e., in which life satisfaction predicted naturist activity via self-esteem and body image) fit the data poorly despite the stronger correlation between naturist activity and body image. Body image weakly predicted participation in naturist activities (b = .004, p = .01), but the total indirect effect of life satisfaction on naturist activities was not significant (LLCI = −.002, ULCI = .002, point estimate < .0001).
By contrast, when using only the subset of participants whose naturist activity scores fell above the median (n = 424), the crucial relationships were no longer significant. The correlation between naturist activity and body image was not significant (r = .02, p = .65). In the mediation model naturist activity did not directly predict more positive body image (b = .01, p = .82), nor did it indirectly predict higher life satisfaction via higher body image and self-esteem (LLCI = −.03, ULCI = .03, point estimate = −.002). In sum, both the moderation analyses and the median split analyses found that the relationship between naturism and positive body image was strongest for those participants who took part in fewer naturist activities. Indeed, it appears that the relationship no longer exists for individuals who pass a certain threshold number of naturist activities. This is similar to other psychological relationships like the association between income and subjective well-being which is strongest for poorer individuals, and weakens or ceases to apply above a certain level of income (Sengupta et al. 2012).
Seeing or Being Seen?
The final analyses investigated whether seeing others naked or being seen naked by others was more strongly related to positive body image. Two independent raters, unaware of the hypotheses, coded the free responses of the participants who had previously taken part in at least one clothes-free activity (n = 805). These raters quantitatively coded (i.e., 0 = no, 1 = yes) whether or not the participant had indicated benefiting from seeing a variety of other naked people (e.g., as one participant said, “Improves own body image, seeing others with normal, not perfect bodies”). The raters also quantitatively coded (i.e., 0 = no, 1 = yes) whether or not the participant had indicated benefiting from being seen naked by others without negative judgements (e.g., as one participant said, “no one judges you on your body shape which gives me more confidence”). All disagreements between raters were managed through discussion.
‘Seeing others’ scores and ‘being seen’ scores were used to predict participants’ body image with PROCCESS macros (Model 1, 95% C.I. 10,000 bootstrap samples). As before, age, gender, ethnicity, religion and sexual orientation were included as covariates. Seeing others naked predicted more positive body image (b = .41, p = .008), though being seen by others did not predict more positive body image (b = .15, p = .16). There was also an interaction between seeing others and being seen (b
interaction
= −.53, p = .04). However, being seen naked by others did not significantly predict positive body image whether participants reported benefiting from seeing others naked (b = −.38, p = .11) or not (b = .15, p = .16).