1.1. Self-determination theory (SDT)

SDT posits that when executing particular behaviors, an individuals’ level of self-determination–i.e., perceived autonomy–or on the contrary constraint, is a key factor as to whether they will demonstrate more or fewer positive outcomes. The SDT continuum of motivation encompasses intrinsic motivation (i.e., doing something that in itself provides pleasure and satisfaction), various forms of extrinsic motivation (i.e., doing something not for its inherent interest, but to achieve a different purpose), and amotivation. From the most self-determined form to the least, extrinsic motivation can be categorized as identified regulation (i.e., adopting a behavior because it has been identified as a good means to reach a personally valued goal), introjected regulation (i.e., acting out of personal pressure), and external regulation (i.e., acting in relation to external contingencies) (see Fig 1). The last two forms of motivation can be conceptualized as the orientation toward attaining valued outcomes (e.g., to attain self-worth perceptions, or rewards from one’s environment) or avoidance (e.g., of negative affects related to oneself, or punishments from one’s entourage).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. The taxonomy of regulatory styles.

Adapted from Ryan & Deci (2000).

https://doi.org/10.1371/journal.pone.0229046.g001

SDT also proposes hypotheses as to why certain forms of motivation may be preferentially developed by individuals in specific contexts. It assumes that motivation will be more self-determined, and hence outcomes will be more positive, if basic psychological needs are fulfilled. That is, the more individuals experience a great sense of autonomy (i.e., feeling that one has options and is at the origin of his/her actions), competence (i.e., feeling able to master one’s environment) and relatedness (i.e., feeling known and accepted by others), the more motivation will be self-determined [6]. While numerous studies have investigated potential outcomes of motivation in educational domains, previous work on behavioral aspects mostly interrogated persistence or achievement [4]. In the case of PE, no previous research was conducted on SED and there were few attempts to date to examine the relationship between motivational characteristics of students and their PA levels.

1.2. SDT motivation constructs and PA-related behaviors in PE lessons

If numerous studies were conducted to assess the link between PE motivation and general levels of PA in adolescents, there were few attempts to examine the is association using PA levels observed during the PE lessons. The research is even scarcer if one considers only studies that employed an objective assessment of PA, which is currently recommended considering the limitations of self-report scales [7]. Three previous studies used pedometers to account for this variable. Self-determination index was significantly associated with step-count in secondary school Hong-Kong students [8]. A more specific result was reported by Johnson et al. (2017) who observed a significant link between PE enjoyment and step counts in 6^th to 8^th grade US students [9]. On the other hand, no significant associations between motivations (intrinsic motivation, identified regulation, external regulation, amotivation) and PA were found in US students from 6^th to 8^th grade [10]. Two additional studies were conducted using accelerometers. Associations between motivation and moderate-to-vigorous PA were investigated in a sample of Australian 9^th graders [11]. Intrinsic motivation and identified regulation were positively related to PA levels, while external regulation was unrelated, and amotivation was negatively associated. Conversely, no relationships were found between those constructs in a sample of American elementary students [12]. To the best of our knowledge, there were no attempts in previous research to investigate the relationship between the satisfaction of psychological needs and PA levels during PE classes in primary school students. However, this parameter appears to be significantly related to objectively assessed PA levels at this age [13].

1.3. The present study

Using objective devices such as pedometers or accelerometers to measure PA-related behaviors is currently considered the gold standard of assessment. However, there is great variation in terms of methods of measurement of PA and the nature of the activity elements that estimate PA duration. For instance, some authors considered a 20-minute structure session of basketball led by the teacher [9]. Others researchers began recording PA only 20 minutes after students were changed [8]. In another study, total class time (from start of class to end of class) was considered [10]. Noteworthy, as stressed by Gao and colleagues (2013), a typical PE class is classically composed of a warm-up period, instructions, skill practice and game play, and a large percentage of time can be attributed to class management and instruction [14]. As such, children’s PA in PE is intermittent in nature and there are even sequences of lessons that are not deemed to generate movement. The relationships observed between psychological characteristics, such as motivation, and PA-related scores, is thus likely to vary depending on the measurement strategy employed.

In the current study, we explored the relationships between satisfaction of basic psychological needs in PE among students, as well as their motivations in this context, and the levels of PA-related behaviors displayed in PE lessons, assessed objectively with accelerometers. It was hypothesized that higher levels of autonomy, competence, relatedness, intrinsic motivation, and identified motivation, would be positively related to PA and negatively to SED, while introjected regulation, external regulation and amotivation would be negatively related to PA and positively to SED. Moreover, we considered different periods in the PE lesson to investigate these relationships. More precisely, we hypothesized that these relationships would vary throughout the lesson according to which periods could reflect more individual motivation. Specifically, we hypothesized that removing the collective and non-active lesson times period (i.e., sequences during which teachers give instructions or feedback and during which students are expected to stay motionless), would enable us to better capture students’ unique behaviors, thus better relating to motivation per se.

2.1. Procedure and participants

The results presented are part of a large study conducted on motivation and physical activity in Swiss adolescents [15]. During this investigation, 40 PE lessons were filmed. Many different sport and physical activities were practiced (e.g. soccer, rugby, volleyball, badminton, and gymnastics). In the current study, we considered the data provided by students that wore an accelerometer during these 40 videotaped lessons. The sample thus involved 319 students in grade 6 to 8 (47% female; age range = 9–13). Throughout the duration of the entire study, questionnaires assessing motivation and psychological needs in PE were completed during regular classes on five different occasions.

The University of Geneva approved this research. In agreement with the Ethics Committee, all participants were given written informed consent to be signed by their parents prior to participation, and received a written debriefing at the end of the study. The data was obtained and analyzed anonymously.

2.2. Measures

3.1. Basic psychological needs and PA

We first looked at the correlations between the basic psychological needs and PA-related behaviors during the entire lesson. Results indicate that when the total PE time is taken into account, basic needs are unrelated to behaviors, with the exception of perceived competence that was negatively correlated with SED and positively correlated with vigorous PA. These relationships are in line with predictions and previous results; however evidence is scarce considering the number of correlations considered.

The pattern of results was quite different when considering the correlations between psychological needs and PA-related behaviors utilizing PA-related behaviors during only activity periods. Whereas only the competence need was related to PA-related behavior when considering the entire lesson sequence, it appears during activity periods that all three needs were related to PA. With the exception of light PA, we found significant correlations between all three needs and PA-related behaviors. Interestingly, SED was negatively associated with the three needs. Moderate and vigorous PA were positively associated with the three needs. Moreover, the needs of competence and relatedness were more strongly associated to vigorous PA than to moderate PA.

3.2. Self-determined motivation and PA

We looked at the correlations between motivations and PA-related behaviors in the entire lesson. Results indicate that when the total PE time is taken into account, overall self-determined motivations were unrelated to behaviors, with a few exceptions. Among the 32 correlations considered, only 4 were statistically significant. Only two forms of motivations among eight were related to PA behaviors; introjected-avoidance regulation was positively related to light PA, and external-approach regulation was positively related with light and moderate PA and negatively with SED. Vigorous PA was unrelated to any form of motivation.

The pattern of results was quite different looking at the correlations between motivations and PA-related behaviors when only activity sequences were considered. Among the 32 correlations tested, 14 were statistically significant. Five forms of motivations among eight were related to PA behaviors. Intrinsic motivation-stimulation and accomplishment were positively related to moderate and vigorous PA and for introjected-approach and external regulation-approach. Identified regulation was positively related to moderate and vigorous PA. Light PA was unrelated to any motivations.