key: cord-321036-nqg9tz4s
authors: Mascret, Nicolas
title: Confinement during Covid-19 outbreak modifies athletes’ self-based goals
date: 2020-09-07
journal: Psychol Sport Exerc
DOI: 10.1016/j.psychsport.2020.101796
sha: 
doc_id: 321036
cord_uid: nqg9tz4s

Because achievement goals are context-specific, the study first investigated the evolution of two achievement goals of 697 regular athletes, namely self-approach goals (improving oneself) and self-avoidance goals (avoiding regression), before and during the confinement situation and the physical exercise restrictions due to the Covid-19 outbreak. Secondly, we sought to examine the potential predicting role of self-approach and self-avoidance goals on athletes’ intention to exercise during confinement, while self-avoidance goals were usually not related to this outcome in a more traditional context. Using a retrospective correlational design, the results of repeated measures ANOVA highlighted that self-approach goals scores decreased while self-avoidance goals scores increased and became the athletes’ goals with the highest score during confinement. The results of hierarchical regression analyses showed that self-approach and self-avoidance goals were both found as positive predictors of intention to exercise during confinement. This study reinforces the assumptions that (a) a specific achievement setting encourages the adoption of different achievement goals (which was identified here with regular athletes in an unusual context of confinement), (b) self-avoidance goals are not always maladaptive, and (c) shifting among multiple goals according to the requirements of the situation may be beneficial.

Pol & Kavussanu, 2011). The question arises whether the confinement situation and the 1 physical exercise restrictions due to the Covid-19 outbreak may impact the achievement goals 2 adoption of regular sportspeople, especially the adoption of avoidance-based goals which 3 focus on failure (Elliot et al., 2011) . Answering this question was the first aim of the study. 4

Self-approach (improving oneself) and self-avoidance (avoiding regression) goals were 5 investigated in the present study because the strong constraints on physical exercise due to 6 confinement could promote self-avoidance goals. In contrast, these goals are usually 7 underrepresented among all athletes in a more conventional sport context because they are 8 likely to be prevalent among older athletes on the downside of their sports career (Mascret, 9 Elliot, & Cury, 2015) . 10

The second aim of the study was to examine the potential predicting role of self-11 approach and self-avoidance goals on athletes' intention to exercise during confinement. To 12 date, no study has investigated this issue even before confinement. But self-based goals were 13 previously included in broader goals called mastery-approach goals (striving to develop 14 competence through task mastery and personal improvement) and mastery-avoidance goals 15

(not doing poorly relative to task demands or one's own performance trajectory). The recent 16 review of Lochbaum, Zanatta, and Kazak (2020) highlighted that mastery-approach goals 17 were positively related with intention to exercise (and many other positive outcomes) whereas 18 mastery-avoidance goals were unrelated with intention to exercise, and often associated with 19 negative outcomes such as worry or reduced enjoyment (Gardner, Vella, & Magee, 2017). 20

Similarly, athletes' self-avoidance goals (avoiding regression) were negatively related with 21 happiness (Briki, 2019). We wondered, however, whether self-avoidance goals may for once 22 be positive predictors of athletes' intention to exercise during confinement because they could 23 be more congruent with this particular and temporary context.

A total of 697 French participants (356 men, 341 women, M age = 31.4 years, SD = 2 10.19) voluntarily filled out the questionnaire. Sample size was based on the maximum 3 number of participants that could be recruited during the week of data collection. Participants 4 had to be of legal age (18 years old in France) and had to practice sport or physical exercise 5 on a regular basis before the confinement due to Covid-19 (M time of practice = 7.15 hours per 6

week, SD = 3.53), assessed using a single item "On average, how many hours per week did 7 you exercise before the start of confinement?". Two hundred and seventeen athletes were 8 practicing sport and physical exercise at recreational level, 353 at departmental to regional 9 level, and 127 at national to international level. 10 Procedure 11

Nonprobability snowball sampling was used to recruit participants (Kosinski, Matz, 12

Gosling, Popov, & Stillwell, 2015). A Web link to access the online questionnaire was sent to 13 a first wave of participants. They were also asked to forward this link to their own network. 14 The completion of the questionnaire was individual and anonymous. Participants signed an 15 informed consent electronically. Data were collected in mid-March 2020 for one week, one 16 month after the implementation of confinement in France. Nonprobability snowball sampling 17 in individual Web-based sessions was selected in the present study because of social 18 distancing rules that have been put in place by the French authorities to control the spread of 19

Covid-19. Concerning physical exercise, the French authorities only allowed short trips linked 20 to individual physical exercise, within the limit of one hour per day and within a maximum 21 radius of one kilometer around home. Collective physical exercise was strictly forbidden. The 22 study was approved by the National Ethics Committee for Research in Sports Sciences 23 (CERSTAPS IRB00012476-2020-13-04-57) and met the requirements of the Declaration of 24 confinement which had not been experimentally manipulated but imposed by the authorities 1 because of the health situation. 2

Self-approach and self-avoidance goals were assessed twice with the two 4 corresponding subscales of the 3 x 2 Achievement Goal Questionnaire for Sport (AGQ-S, 5

Mascret et al., 2015) initially validated in French language. Participants were first asked to 6 answer items on a Likert scale from 1 (strongly disagree) to 7 (strongly agree) assessing 7 retrospectively their self-approach goals (e.g., "When I was exercising before confinement, my 8 goal was to have better results than I had in the past") and self-avoidance goals (e.g., "When 9 I was exercising before confinement, my goal was to avoid doing worse than I usually do") 10 when they had physical exercise prior to the start of confinement. We used a retrospective 11 design to assess self-based goals before confinement because (a) it was not possible to initiate 12 the study before the start of confinement, (b) we were interested in the overall self-based 13 goals adopted when participants had physical exercise before confinement, and (c) with ω ranging from .73 to .94. We expected self-approach goals to decrease and self-21 avoidance goals to increase between the measures before and during confinement due to the 22

Covid-19 outbreak (Hypothesis 1). 23

Intention to exercise during confinement was assessed based on a French translation of 24 the measure adapted from Chatzisarantis, Biddle, and Meek (1997). Participants responded to 25 J o u r n a l P r e -p r o o f Confinement due to Covid-19 and achievement goals 5 the three items (e.g., "I plan to exercise at least three times a week during confinement") on a 1 7-point scale ranging from 1 (very unlikely) to 7 (very likely). Internal consistency was high 2 (ω = .97). Physical exercise is defined in the present study as a "planned, structured and 3 repetitive bodily movement, the objective of which is to improve or maintain physical fitness" 4 (Caspersen, Powell, & Christenson, 1985, p. 126). We expected both self-approach and self-5 avoidance goals to be positive predictors of intention to exercise in the particular context of 6 confinement, whereas self-avoidance goals were not usual predictors of intention to exercise 7 in a more conventional context. 8

Preliminary analyses were previously conducted. No missing data were found. Two 10 participants were excluded because they were identified as outliers using Mahalanobis 11 distance (p < .001) at the multivariate level. Skewness and Kurtosis estimates indicated that 12 the measures of self-approach and self-avoidance goals were normal in distribution. All 13 statistical analyses were conducted using JASP software (version 0.12.2). 14 A repeated measures ANOVA with the factors Goals (Self-approach, Self-avoidance) 15 and Time (Before confinement, During confinement) revealed a significant effect of Time 16 (F(1, 1392) = 201.09, p < .001, η p ² = .126) and a significant Goals X Time interaction (F(1, 17 1392) = 399.32, p < .001, η p ² = .223). Post hoc comparisons using the Tukey HSD test 18 indicated that the mean score for self-approach goals during confinement (M = 3.69, SD = 19 1.90) was significantly lower than before confinement (M = 5.47, SD = 1.51, p < .001). 20

Conversely, the mean score for self-avoidance goals during confinement (M = 4.72, SD = 21 1.69) was significantly higher than before confinement (M = 4.42, SD = 1.79, p < .001).

Hypothesis 1 was confirmed. Ancillary analyses showed that before confinement self-23 approach scores were significantly higher than self-avoidance goals (p < .001). The pattern 24 was reversed during confinement (p < .001). While improving themselves was athletes' goal with the highest score before confinement, avoiding regression became their goal with the 1 highest score during confinement. Figure 1 illustrates the previous results. After standardizing the relevant variables to reduce multicollinearity, we conducted 5 hierarchical regression analyses to examine how gender (women = 0, men = 1), age, sport 6 level (Step 1), self-based goals (Step 2), and the interactions between self-based goals and 7 gender, age, and sport level (Step 3) predicted intention to exercise during confinement. 8

Gender, age, and sport level were included in the hierarchical regression analyses to control 9 for these variables. The results showed that self-approach and self-avoidance goals were both 10 positive predictors of intention to exercise during confinement. Hypothesis 2 was confirmed. 11

Ancillary analyses finally showed that results did not change when the control variables were 12 removed. A significant interaction was also found between self-avoidance goals and sport 13 level and was then decomposed by simple slopes analyses (Aiken & West, 1991) at ±1SD. 14 This interaction effect was not represented because simple slope analysis did not reveal 15 significant findings (p = .726 for low sport level and p = .457 for high sport level). Table 1  16 shows the detailed results of the final regression analyses. 17

Insert Table 1 about here  18 19 Discussion 20

The results of the present study indicated that the confinement situation and the 21 physical exercise restrictions due to the Covid-19 outbreak influenced self-based goals of 22 regular athletes of different sport levels. Self-approach goals decreased but kept their 23 predicting role on intention to exercise already found in the literature before confinement 24 (Lochbaum et al., 2020). Self-avoidance goals increased, became the goals with the highest 25 score, and appeared to be positive predictors of intention to exercise during confinement. 26 J o u r n a l P r e -p r o o f Confinement due to Covid-19 and achievement goals 7 These results reinforced the fact that achievement goals are context-specific. Benita and 1 colleagues (2014, 2017) highlighted that adopting mastery-approach goals in a controlled 2 context (with constraints) promoted lower positive emotional experiences than in an 3 autonomous context (with choice and volition). The strong constraints associated with 4 physical exercise, implemented by the French authorities to limit the spread of Covid-19, may 5 be experienced by regular athletes as a form of controlled context which have impacted both 6 their self-based goals and their predicting role on intention to exercise. Moreover, self-7 avoidance goals combine positive (intrapersonal) and negative (avoidance) components and, 8 theoretically, positive or negative outcomes may be induced depending on the predominant 9 component (Senko & Freund, 2015) . In the context of physical exercise during confinement, 10 self-avoidance goals were more likely to be positive predictors of intention to exercise 11 because they were more functionally-congruent with the context (Conroy, Cassidy, & Elliot, 12 2008), which was not the case before confinement. The present study conducted with athletes 13 in the particular context of confinement reinforces the general assumptions that "different 14 achievement goals may be better suited for different types of situations" (Barron &  15 Harackiewicz, 2001, p. 708), that mastery-avoidance goals -especially self-avoidance goals -16 are not always maladaptive (Mascret, Nicolleau, & Ragot-Court, 2020; Senko & Freund, 17 Some limitations of the present study need to be acknowledged. First, the study used a 20 retrospective design and the particular context of confinement could have been influenced 21 athletes' answers when they were asked to remember and to rate their self-based goals before 22 confinement. Secondly, the study did not investigate participants' types of sport. It would 23 have been interesting to examine for instance whether the pattern of results remained the same 24 between individual and team sports. 25 J o u r n a l P r e -p r o o f Confinement due to Covid-19 and achievement goals 8 While improving themselves was the goal with the highest score before confinement, 1 avoiding regression became the athletes' goal with the highest score during confinement and 2 predicted their intention to exercise during this period. This is of particular broad interest 3 because these results confirmed that human beings are capable of changing their goals in 4 order to adapt as well as possible to a particular situation, here a situation as exceptional as 5 confinement linked to a pandemic. .188 (.182***) Self-approach goals 1 .51*** Self-avoidance goals 1 .38*** Step 3

.201 (.013) Self-approach goals X Gender .07 Self-avoidance goals X Gender -0.03 Self-approach goals X Age -0.01 Self-avoidance goals X Age -0.01 Self-approach goals X Sport level -0.19 Self-avoidance goals X Sport level -0.21* Note. *p < .05, **p < .01, ***p < .001

A meta-analytic review of

Hierarchical Model of Approach and Avoidance Motivation in the 13 sport, physical activity, and physical education literature

The 2× 2 Achievement Goals in Sport and 16 Physical Activity Contexts: A Meta-Analytic Test of Context, Gender

Development and validation of a scale 1 assessing achievement goals in driving

Are mastery-avoidance achievement goals always 4 detrimental? An adult development perspective

Achievement goals and motivational responses in 7 tennis: Does the context matter?

A novel approach to assessing achievement goals in the context of 10 the 2 × 2 framework: Identifying distinct profiles of individuals with different 11 dominant achievement goals

Achievement goals and self-regulation in the sport 14 context