Attitude and practice of physical activity and social problem-solving ability among university students

Questionnaire surveys and psychological tests were conducted between July and December 2015. The study population comprised students aged 18 years or older enrolled in Tohoku Fukushi University. We recruited 2750 first- and second-year students to participate in the study using email. A total of 186 students (6.8%) participated in the questionnaire surveys and psychological tests. One participant who did not answer questions about physical activity was excluded, leaving 185 participants (6.7%) for analysis. Tohoku Fukushi University has four faculties (including nine departments), in Welfare, Healthcare, Education, and Management. The mission of these faculties is to deeply inquire into the truth, develop widespread knowledge and specialized skills in their fields, and build meaningful relationships, with the ultimate aim of creating paths to the wellbeing for all mankind.

The questionnaire requested the following information from each participant: age, sex, living situation, body weight, height, sleeping condition, dietary habits, physical activity, social activity, psychological distress, and life stressors. The printed questionnaires was administered to participant.

Attitude and practice of physical activity was defined as “undertaking moderate physical activity such as walking, riding a bicycle, or doing household tasks while standing or walking for at least 60 min per day in daily life.” This measure has been used in the National Health and Nutrition Survey, conducted by the Ministry of Health, Labour and Welfare of Japan [21]. The possible responses were as follows “I do not practice physical activity, and I never think about doing so,” “I do not practice physical activity, but I intend to start doing so,” “I try to do physical activity, but have not done it much yet,” “Yes, I practice physical activity, but not regularly,” and “Yes, I practice physical activity and regularly.”

The Social Problem-Solving Inventory–Revised (SPSI-R) is a 52-item self-report questionnaire that assesses problem-solving skills [22]. The Japanese version of the SPSI-R, which has been validated previously, uses a 50-item self-report questionnaire [23, 24]. The questionnaire contains five subscales: positive problem orientation (PPO), negative problem orientation (NPO), rational problem solving (RPS) (which has four subcomponents: problem definition and formulation (PDF), generation of alternative solutions (GAS), decision making (DM), and solution implementation and verification (SIV)), impulsivity/carelessness style (ICS), and avoidance style (AS). The number of items on each subscale and a sample item are as follows: PPO, four items, e.g., “I believe I can solve a problem if I try hard enough”; NPO, 10 items, e.g., “I feel afraid when I have a problem to solve”; RPS, 20 items, e.g., “I approach problems from many angles”; ICS, 10 items, e.g., “I go with the first good idea that comes to mind”; AS, six items, e.g., “I avoid thinking about problems.” Higher scores on the PPO and RPS (and its subcomponents) indicate more adaptive problem-solving capacity, whereas higher scores on the NPO, ICS, and AS reflect more a more maladaptive approach to problem solving. Each item is answered on a 5-point rating scale with anchors of 0 (“not at all true of me”) and 4 (“extremely true of me”). Sato et al. indicated adequate internal consistency for the Japanese version of the SPSI-R (Cronbach’s α = 0.89) and all subscales (Cronbach’s α = 0.72 to 0.92). In addition, the test–retest reliability over 3 weeks for all subscales as indicated by Pearson’s r was indicated from 0.38 to 0.61 [23]. In this study, Cronbach’s α ranged from 0.61 to 0.90 (SPSI-R, 0.90; PPO, 0.61; NPO, 0.88; RPS, 0.86; PDF, 0.64; GAS, 0.61; DM, 0.62; SIV, 0.77; ICS, 0.77; and AS, 0.75).

First, comparisons between physical activity categories were performed via chi-squared analysis to assess differences in characteristics.

Previous studies did not report a well-accepted cut-point for the SPSI-R [22, 23]. Therefore, we conducted analysis of covariance using continuous variables and logistic regression analysis using median scores.

Second, an analysis of covariance was performed to investigate the significance of differences in the SPSI-R among respondents in the various physical activity categories.

Third, we conducted logistic regression analysis to evaluate the association between attitude and practice of physical activity and social problem-solving. Social problem-solving (median score 10.85) was the dependent variable, whereas the independent variable was physical activity, which was divided into five categories: “do not exercise, and never think about doing so (“I do not practice physical activity, and I never think about doing so”),” “do not exercise but intend to start (“I do not practice physical activity, but I intend to start doing so”),” “try to exercise but do not (“I try to do physical activity, but have not done it much yet”),” “exercise but not regularly (“Yes, I practice physical activity, but not regularly,”),” and “exercise regularly (“Yes, I practice physical activity and regularly”)”. Multiple logistic regression analysis was performed to calculate the odds ratios (ORs) and 95% confidence intervals (CIs) for higher social problem-solving according to these five physical activity categories. Furthermore, we repeated the logistic regression analysis after modifying the dependent variable to each subscale, including PPO, NPO, RPS (PDF, GAS, DM, and SIV taken individually), ICS, and AS. The median scores for each of these variables were as follows: PPO, 12; NPO, 23; RPS, 45; PDF, 13; GAS, 12; DM, 11; SIV, 10; ICS, 19; AS, 12.

Fifth, we also conducted stratified analyses according to differences in sex, i.e., we queried whether sex significantly interfered with the association between physical activity and social problem-solving ability. Sex, as the interaction between physical activity and social problem-solving ability, was tested through the addition of cross-product terms to the multivariate-adjusted model.

Sixth, in a supplementary analysis, we conducted multiple Poisson regression analysis to evaluate the association between attitude and practice of physical activity and SPSI-R and subscales and to calculate the rate ratios (RRs) and 95% CIs for higher social problem-solving according to these five physical activity categories.

As shown in Table 1, participants belonging to the categories of “try to exercise but do not,” “exercise but not regularly,” and “exercise regularly” were less likely to be women or living alone than those who said they never considered exercising. In contrast, the proportion of men and those living alone was similar with regards to those who responded that they never considered exercising, and those who did not exercise but intended to begin exercising.

The mean score ± standard deviation (SD) on the SPSI-R was 10.4 ± 2.7 for participants who said they never considered exercising, 10.7 ± 2.0 for participants who did not exercise but intended to start, 10.6 ± 2.1 for participants who tried to exercise but did not, 10.8 ± 2.1 for participants who exercised but not regularly, and 12.3 ± 2.8 for participants who exercised regularly. We found significant differences between the five categories of physical activity with respect to the SPSI-R, at the p < 0.01 level. Subscale analyses showed the presence of significant differences in NPO means between the five categories of physical activity (NPO; p = 0.03) (Table 2).

As shown in Table 3, the unadjusted analysis indicated that the ORs (95% CI) in reference to participants who answered “I do not practice physical activity, and I never think about doing so” were 2.03 (0.68–6.65), 1.68 (0.58–5.34), 3.11 (1.00–10.60), and 7.00 (2.11–26.13) for participants who did not exercise but intended to start, tried to exercise but did not, exercised but not regularly, and exercised regularly, respectively (p value for linear trend < 0.01). This finding remained basically unchanged, even after adjusting for age, sex, and living alone (p value for linear trend < 0.01). Thus, a greater frequency of higher SPSI-R scores was found among respondents stating that they tried or intended to exercise but did not, as well as those who exercised regularly or irregularly, when compared to those who had not even considered exercising. With regard to subscales and the subcomponents of RPS, linear associations were observed between physical activity and NPO, DM, and ICS. However, there were no significant linear associations between physical activity and PPO, RPS, PDF, GAS, SIV, or AS.

We further evaluated sex difference in the association between physical activity and social problem-solving ability (Table 4). There was no significant interaction between physical activity and social problem-solving ability (p value for interaction = 0.44).

The multiple adjusted analysis indicated that the RRs (95% CI) in reference to participants who answered “I do not practice physical activity, and I never think about doing so” were 1.57 (0.63–3.91), 1.36 (0.55–3.36), 1.78 (0.71–4.49), and 2.31 (0.93–5.74) for participants who did not exercise but intended to start, tried to exercise but did not, exercised but not regularly, and exercised regularly, respectively. Higher SPSI-R was more frequently observed among participants who exercised regularly than those who did not exercise. This linear association was marginally significant (p value for linear trend = 0.05). With regard to the subcomponents of RPS, linear associations were observed between physical activity and DM (p value for linear trend = 0.04) (Additional file 1: Table S1).