1. Introduction
Worldwide figures suggest that around 20% of males and 27% of females are insufficiently active [
1]. This is a concern for both public health and the economy, with conservative estimates suggesting that the global economic cost of inactivity is US
$53.8b annually [
2]. Recently described as a “miracle cure” [
3], regular physical activity (PA) is well known to improve health [
4] and prevent ill health [
5]. Given the importance of PA, a wealth of literature has sought to identify the determinants of exercise behaviour, with individuals’ beliefs and attitudes found to play a key role [
6,
7]. Beliefs and attitudes have subsequently been targeted in PA interventions with positive effects observed on exercise behaviour and in turn physical and mental health [
8,
9]. Beyond the overall value of exercise, however, PA that is performed in the natural environment (termed green exercise) may have enhanced health benefits [
10,
11,
12]. To better understand the role of green exercise for health and well-being and to contribute to the development of green exercise interventions, therefore, it is vital to elucidate individuals’ beliefs about green exercise. This current paper contributes to these issues by developing and providing initial evidence for the psychometric properties of three questionnaires that assess beliefs about green exercise.
Researchers have increasingly examined the role of urban green spaces (designed and maintained with human input such as parks and domestic gardens) and natural green space (naturally occurring with minimal human input) for PA and health. A growing body of evidence suggests that the natural environment encourages behaviour change by facilitating people to be more active [
13,
14,
15,
16]. Associations have also been found between local green space and mental well-being [
17,
18,
19]. For example, national studies in the UK and the Netherlands have found links between accessible local green space and PA levels [
14,
20,
21]. Further, evidence collected at urban green spaces suggest that people value green spaces for PA and health more so when they view the features and characteristics favourably [
22,
23,
24,
25]. As such, the link between urban green spaces and health is now formally recognised by the World Health Organisation [
26].
Within natural environments and urban green spaces, individuals can participate in green exercise purposefully (as active participation) or incidentally (as functional engagement) [
10]. An example of functional engagement would be walking through a park because it is the shortest route to your destinations. In contrast, if the park was not the quickest route, but was selected because it allowed you to experience nature along your journey, this would be considered active participation. As such, although the experience of nature (i.e., seeing trees and grass) may be similar, the motivation for purposeful and incidental green exercise is different. However, the motivation to do green exercise has not been widely researched [
27,
28]. Instead, a key focus in the literature has been on the psychological benefits of exercising within different environments. For example, comparisons have been made between performing the same exercise in indoor versus outdoor environments [
29,
30,
31,
32] and in urban versus rural outdoor environments [
33,
34,
35,
36]. Evidence indicates that green exercise can elicit psychological improvements over and above that of indoor and urban exercise [
15,
16,
30,
37,
38,
39]. Specifically, acute bouts of green exercise have been shown to facilitate reductions in anxiety [
40], reductions in mood disturbance [
41,
42], and improvements in self-esteem [
43,
44,
45]. Moreover, dose-response relationships have shown that the greatest benefits on mood and self-esteem occur within the first five minutes of green exercise [
42].
Despite the promising evidence for the benefits of green exercise, little is known about how individuals’ thoughts and feelings about green exercise influence participation and subsequent outcomes. This is surprising given the myriad of studies that have demonstrated the importance of understanding the beliefs about PA more generally [
6,
7]. A small number of studies, however, have shown that perceptions of local green space may predict visit frequency more than quantity and proximity [
14,
46,
47,
48]. Additionally, the New Ecological Paradigm [
49] and Nature Relatedness Scale [
50] are tools that have been developed to measure how people feel towards nature, but these have rarely been applied in the context of PA. Nevertheless, one recent study found that nature relatedness was a strong predictor of visit frequency to local green space [
14]. An exploration of green exercise beliefs is needed to better understand how to augment engagement with green exercise behaviours.
The theory of planned behaviour (TPB) is an important framework that has been used to advance understanding of how cognitions influence PA behaviours generally [
7,
51,
52]. Derived from the theory of reasoned action [
53], the TPB assumes that intention to perform a behaviour is best predicted when individuals evaluate the behaviour positively (attitudes), believe peers will support the behaviour (subjective norm), and perceive the behaviour to be within their capabilities (perceived behavioural control; PBC). TPB factors can be assessed directly (e.g., by asking people to report attitudes, norms, and PBC) or indirectly (e.g., by asking people about specific behavioural beliefs and combining the scores with a paired evaluation of the belief) (see
Figure 1). As such, indirect behavioural, normative, and control beliefs combine with evaluations of those beliefs to predict the respective direct measures of attitudes, subjective norms, and PBC. Not only does this enable correlational analyses to establish convergent validity, but also serves to capture the different underlying cognitive processes of each measure [
54]. Despite concerns about the intention-behaviour gap, the TPB [
55] has been the most successful approach in exercise psychology for predicting participation from beliefs [
56]. One meta-analysis revealed that nearly half of variance in PA intentions, and over a quarter of variance in PA behaviours could be explained by beliefs [
7].
As a form of PA, some authors have hypothesised that green exercise can be modelled using the TPB [
15,
57]. Using a systematic review, Calogiuri and Chroni [
15] integrated the green exercise literature with the TPB to propose a schematic model of motivational processes underlying the relationship between natural environments and physical activity behaviours. The evidence collected supports using the TPB framework to explore the green exercise phenomenon. Moreover, empirical evidence—collected using ad hoc TPB questionnaires—has shown that beliefs may predict behaviours such as park visitation [
58], participation in outdoor recreation programs [
59], outdoor walking [
60], and outdoor pool use [
61]. Although these studies have enriched understanding into the role of beliefs on specific green exercise behaviours, the need to create measurement tools for individual studies does not encourage a proliferation of research into green exercise beliefs and also impairs the ability to synthesise evidence across studies. A valid measure of beliefs about green exercise is necessary to deepen our understanding of the relationship between green exercise and health, understand variations in green exercise beliefs, and develop intervention to increases green exercise.
The aim of the current article was to develop and provide initial evidence of the validity of three questionnaires that assess global beliefs about green exercise amongst the general population. The research drew upon comprehensive guidelines for developing TPB questionnaires [
63], which provided recommendations for overall structure, item wording, and scoring criteria, and a recent a systematic review that identified a 16-point criteria for assessing the quality of TPB questionnaires [
64]. These criteria emphasise the importance of methodological rigour, such as the inclusion of an elicitation study, developing both indirect and direct measures, and establishing content validity. As such, the current research was divided into two distinct studies (see
Figure 2). Study 1 focused on the elicitation and content analysis of salient beliefs about green exercise, and Study 2 focused on the development and validation of the three questionnaires that assess indirect beliefs, direct beliefs, and intention to perform green exercise respectively.
4. General Discussion
The aim of the current study was to develop and provide initial evidence of the validity of three questionnaires that assess individuals’ beliefs about green exercise. Using the TPB as a theoretical framework and drawing upon established guidelines [
63,
64], tools were developed to assess direct and indirect measures of attitudes, subjective norms and PBC, and intention to perform green exercise. Evidence was provided for the factorial validity, composite reliability, and parallel-form reliability for each of the three questionnaires. Consistent with previous studies [
15,
58,
59,
60,
61], our findings support the theoretical structure of the TPB in relation to performing PA. To date though, no other instruments exist that focus explicitly on beliefs about green exercise. As such, the current findings offer a novel suite of measurement instruments that could be used to provide important insight into the role of individuals’ beliefs in green exercise, thereby contributing to the development of theory and effective interventions.
For content validity, salient beliefs were established through an elicitation study (Study 1). Although valuable for development of TPB questionnaires [
62,
64], this step is often overlooked within PA research [
66]. The salient beliefs captured in the elicitation study informed the development of the BAGE-ID and provide confidence that the items reflect the key cognitions about green exercise in the general population. Consistent with previous research, the advantages and likeable features of green exercise were the most prevalent [
77,
78]. Interestingly, however, weather—as a disadvantage or dislikeable feature of green exercise—was the most reported salient belief. This may be because the temperate oceanic climate of the UK is not conducive to green exercise throughout the year. Similarly, previous research has found that climatic conditions have an important impact on PA levels across different populations [
79,
80,
81].
Evidence was provided for the factorial validity of the three measures across two samples (refinement and validation). Specifically, following modifications in the refinement sample, the BAGE-ID, BAGE, and INT-GE had good model fits and all factors loadings were significant in both the refinement and validation samples. Of note, however, is that the item relating to weather in the BAGE-ID was removed during the modification process. Although weather was identified as dislikeable feature of green exercise in Study 1, climatic conditions were also listed as a facilitating factor. These contrasting views may partly explain why the factor loading of weather on PBC was not as strong as other items in Study 2. Instead—and congruent with previous research [
14,
23,
47]—items relating to the size, facilities, and safety of local green space may be more reflective of PBC. Overall though, the confirmatory factor analyses indicated that the refined questionnaires had a good factorial validity and model fits were comparable to other questionnaires based on the TPB [
82,
83].
The current findings provide support for using the TPB as a model to explore green exercise, and the proposed relationships between indirect and direct measures of attitudes and subjective norms. Similarly, previous research has successfully employed the TPB to explore the relationship between PA and nature [
15,
57]. Consistent with Calogiuri and Chroni [
15], we found evidence that the motivational processes associated with green exercise can be modelled from beliefs to intentions. Indeed, previous research has demonstrated that TPB factors can predict intention to engage in outdoor recreation programmes [
59] and to visit state parks [
58]. The current findings extend the literature by demonstrating that both indirect and direct measures of attitudes, subjective norms, and PBC do predict intention to perform green exercise. Beyond the empirical support for the TPB, these findings could underpin the development of evidence-based interventions to promote intention to perform green exercise.
Previous research has made an important contribution in demonstrating that green exercise has important psychological and health benefits [
30,
37], and that individuals’ beliefs can facilitate the benefits of PA [
8,
9]. The questionnaires developed in this study complement this research and provide tools to explore the role of beliefs in green exercise behaviours and outcomes. The consistent use of the three questionnaires will facilitate attempts to synthesise research findings and enable researchers to address theoretically interesting questions, such as which beliefs are the most salient predictors of green exercise behaviours and under what conditions? In total, the questionnaires contain 39 items, all written in the English language with scaled multiple-choice responses, and can be completed in less than 15 min. As such, researchers and health professionals could utilise the tools to assess beliefs in a variety of settings, including field studies, laboratory-based experiments, and applied interventions.
Key strengths of the present research were the use of an elicitation study, the development of questionnaires that assess both indirect and direct measures of attitudes, subjective norms and PBC about green exercise, and the ability to replicate the observed factor structures in two samples. According to the assessment criteria for TPB questionnaire development [
64], such steps would enable the current study to achieve Grade A. Despite these strengths, some limitations should be noted. With regards to methodology, the randomisation of participants to the refinement and validation samples in Study 2 led to significant between-group differences for age and gender. In the future, it might be useful to consider stratified random sampling to control for demographics prior to doing confirmatory factor analyses. The correlational nature of Study 2 also limits the ability to infer causality in the relationships between indirect, direct, and intention measures. Further, although the findings demonstrate that beliefs predict the intention to perform green exercise, the relationship with subsequent behaviour was not explored. Future research should therefore explore whether intentions mediate the relationship between beliefs and green exercise behaviour.