The Role of Physical Education Homework to Adolescent Girls’ Physical Activity in Finland

Introduction: There is an urgent need for active strategies to promote physical activity (PA) among female adolescents, as they are at greater risk of physical inactivity. In addition to Physical Education (PE) in school, to reach adequate levels of PA, young people need to participate in PA during the school day and/or after-school hours and/or in the before-school period. One week of objectively measured and self-reported information enlightens the status quo of adolescent girls’ PA and the role that PE homework plays. Purpose: The purpose of this study was to examine girls’ objectively measured PA in a lower secondary school as part of the Physical Education Homework Study, a project conducted in Finland. An additional aim was to provide insights into physical education (PE) homework as part of PA. Methods: Different levels of PA (i.e., light, moderate, vigorous), sedentary time, number of total steps and running steps were measured using accelerometers among girls in the 7th, 8th and 9th grades, aged between 12 and 15 years (n = 88) for a period of 1 week. In addition, self-reported structured diaries were used to gather information about after school activities, including PE homework. An analysis of variance (ANOVA) and Tamhane’s post-hoc test was used to compare the different grades on the basis of participants’ PA levels and sedentary time. Data from self-reported PE homework frequencies and time spent on PE homework 0.000) and 30 minutes more than the girls from the 9th grade (p = 0.000). Conclusion: The PE homework provided a variety of PA for the adolescent girls’ after-school hours, even if the role of homework was not time-consuming (20 - 51 minutes per week). A possible way of promoting active time among adolescents during after-school hours is to assign them PE homework, hence, active PE assignments could replace sedentary activities in adolescent girls’ lives.


Introduction
In most countries, physical activity (PA) participation radically decreases during lower secondary school, especially among girls (Currie et al., 2008: pp. 105-107;Ekelund et al., 2012;Haapala et al., 2017). As a result, many adolescents fail to meet the recommended levels of PA (Currie et al., 2008;Pate & O'Neill, 2009;Husu, Vähä-Ypyä, & Vasankari, 2016). According to the Finnish National Physical Activity Recommendations, adolescents should be physically active more than 60 minutes per day (Tammelin & Karvinen, 2008). In Finland, only 10% -17% of adolescents aged between 13 and 15 attain moderate-to-vigorous levels of PA, with the most radical decline in PA occurring among girls aged between 11 and 15, in their transition from childhood to adolescence Tammelin, Laine, & Turpeinen, 2013). According to Pate & O'Neill (2009), the objectively measured PA of adolescent girls declines at a rate of approximately 4% annually. This means that with time, the level of PA among girls recedes further from current PA guidelines.
Nearly half of young people's total daily PA accumulates after school, although the levels of after-school activity start to decrease gradually during preadolescence, while sedentary exposure increases as early as ages 9 to 11 (Wickel, Issartel, & Belton, 2013). Since female adolescents seem to be at greater risk of physical inactivity, there is an urgent need for effective strategies to promote PA among them (Aelterman et al., 2012). The after-school period represents a segment of the day in which adolescents' health-enhancing levels of PA should be promoted (Beets et al., 2009). However, adolescents, particularly girls, accumulate low levels of moderate-to-vigorous physical activity (MVPA) during after-school hours, and their MVPA levels decline at a faster rate (Okely et al., 2011;Pate & O'Neill, 2009). According to Atkin et al. (2008), technology-based sedentary behaviour is the most prevalent activity among adolescents in after-school hours; the sedentary time of girls lasted for 35 minutes per day.
Meanwhile, the times spent physically active amounted to 19 minutes for girls per day (Atkin et al., 2008). School is an important institution for PA promotion, as the vast majority of youths attend school. Thus, physical education (PE), as a structured school activity, has great potential in terms of public-health reach (Scruggs, 2007). There have been calls for schools to expand their efforts to increase PA-related opportunities to obtain higher levels of PA in adolescents, particularly adolescent girls.
Even if school PE lessons are well organized and PA levels during these lessons are adequate, PE lessons do not exert a sufficient influence over adolescents' PA and health (Fairclough & Stratton, 2005). MVPA is performed most often during leisure time, while school sports contribute only marginally to MVPA (Pfitzner et al., 2013). In Finland, a great deal of effort has been set into activating school days, for example, the Finnish Schools on the Move (Haapala et al., 2017). In addition to school PE, to reach adequate levels of PA, children and young people need to participate in PA during the school day and/or during after-school hours and/or in the before-school period. A possible way of promoting PA among youths is to assign PE homework (Gabbei & Hamrick, 2001;Williams & Hannon, 2013;Kääpä et al., 2017). Homework assignments could be designed to practise activities learned earlier at school; to be effective, the practicing should take place during after-school hours. Physical homework provides an opportunity for students to apply and practice the skills learnt in PE lessons (Gabbei & Hamrick, 2001;Williams & Hannon, 2013). According to Kääpä et al. (2017), students can find PE homework enjoyable and beneficial. For example, students enjoyed PE homework that included practicing with family member. In addition, it has been found that Finnish students enjoy participating in planning PE homework (Kääpä et al., 2017). The Finnish National Core Curriculum allows and even encourages the use of leisure time to practice skills learnt during lessons (Finnish National Core Curriculum, 2014). PE homework assignments can be implemented in nearby sporting environments or facilities. Becoming familiar with the physical environment and facilities while undertaking PE homework helps to make PA part of one's lifestyle (Williams & Hannon, 2013;Smith & Claxton, 2003).

Participants
The participants for this study were girls in the 7th, 8th and 9th grades in a lower secondary school in central Finland. The girls were aged between 12 and 15 years. The half of the girls live near the school and half of them commute to school by buss. From a total of 124 girls, 105 (84.7%) agreed to participate in the study. Girls and their parents were informed about the study beforehand, and informed consent was obtained from the parents/guardians. According to the layer of the University Ethical Board, this protocol covers needed ethical permissions concerning this type of research. The study consisted of week-long objective PA measures using accelerometers and structured self-report diaries during the measurement week. There were missing data from 17 participants as a result of students (12) who were absent from school during the measurement week, and some accelerometer problems that were encountered. The actual study data consists of information from participants who had at least 4 days of accelerometer data (n = 88) and questionnaires concerning their self-report diaries (n = 81).
In self-reported data, the division was 28 girls from the 7th grade, 26 girls from the 8th grade and 27 girls from 9th grade. In the objectively measured data analysis, to get the most comprehensive overview about the adolescents' physical activity, all of the participants acceptable (at least 4 days, 10 h minimum/day) accelerometer data were used (n = 513). For this reason, the measurement days consisted of 190 days of data from the 7th grade girls, 152 days of data from the 8th grade girls and 171 days of data from the 9th grade girls. Girls' who had both objectively measured and self-reported data were included in analyses of compound measurement results (No PE homework n = 238, PE homework twice a week or more n = 248).

Setting
In Finland, the school day usually begins at 8:00 am and ends at 3:00 or 4:00 pm.
The normal school-day structure consists of 7 to 8 lessons lasting for 45-minute each, with a 15-minute break between the lessons and a 30-minute lunch break at noon. The Finnish Schools on the Move program has concentrated on activating students during the school day and using the breaks in an active way (Haapala et al., 2017). The PE Homework Study focuses on activating after-school hours (Kääpä et al., 2017). During the study period, the lower secondary school students had 2 × 45 mins PE lessons in a row (lasting 90 minutes) once a week. In Finland, this double classes one day a week is normal national standard. In Finland, organized after-school sport activities for adolescents usually take place in sport clubs and in teams. According to Blomqvist et al. (2015), over half (53%) of the Finns in their study aged between 9 and 15 took  (Blomqvist et al., 2015).
The PE teacher (first author) had occasionally given the girls in the lower secondary school PE homework in past years; thus, the concept of PE homework was familiar to them. PE homework was an aspect of PE that was undertaken by all of the girls. Homework assignments followed PE requirements, and they were given to all students in every PE lesson once a week during the study year. This meant that PE homework was given between 30 and 40 times per school year. PE homework assignments often consisted of body-conditioning exercises such as push-ups, squats or ab crunches; jogging with a family member or friend; stretching; or some skill or activity from the PE lesson. An example of PE homework is balance practices, where the student stands on their right foot in the morning while brushing their teeth and on their left foot at night while brushing their teeth. PE skill practices consisted of tasks practiced in PE lessons that they had to practice for PE homework (e.g., throwing and catching a ball).
An example of PE homework combining skill practice and parental involvement is teaching the correct performance of squats to a family member. Involving the students in planning their PE homework was sometimes part of the PE lessons' "cooling-down" session. Often, the students could choose the scale or complexity of the homework themselves and they had an influence on the activity. Improving the girls' autonomy and self-determination was the main goal in involving them in planning, choosing the scale and deciding on the amount of PE homework (Deci & Ryan, 2000). Girls participated in their PE homework assignments actively, 89.7% of the girls always completed their PE homework (Kääpä et al., 2017).

Self-Reported Data
The students' self-reported PA was assessed using a structured diary questionnaire. The aim was to identify the frequency and type of PA that girls engaged in. The diary questionnaire consisted of categorized structures where students marked their activity frequency, level of self-initiated PA, organized sport activities, occupational activities and PE homework in which they participated within the study week. In addition, they estimated the time spent on PE homework. In this study, only self-reported PE homework reports were used. Girls this age are expected to be able to fill out this kind of structured diary that does not require a lot of creativity or writing. Self-reported data supplements the objectively measured data identifying the context of physical activity.

Procedures
The accelerometer-and diary-based measurements were implemented in November 2016, 3 months following the beginning of the semester, meaning that several PE homework assignments had already been completed. The teacher delivered the devices during PE lessons and instructed the girls to wear the flexible (elasticated) belt with the accelerometer around their waist, placing the monitor on the right hip. The teacher gave oral instructions, and written instructions were attached to the diary questionnaire, which was delivered during the same lesson. The girls were instructed to wear the belt for seven consecutive days during their waking hours, apart from during water-based activities such as showering or swimming. The girls started the monitoring straight away and tested the accelerometers; if a malfunction was detected, the device was replaced immediately.
The girls returned the accelerometers and their self-report diaries a week later, during their following PE lesson, fulfilling instructed seven consecutive days requirement. To obtain a comprehensive overview, only the data from the participants who used the accelerometer for at least 4 days during the week and at least 10 hours each day were accepted for the study, and the data were analysed in 6-second epochs (Troiano et al., 2008). In studies with large sample, reliability can be achieved with 3 days of recording. Studies with smaller sample sizes like this, it is recommended to measure a greater number of days to improve measurement precision (Mattocks et al., 2008). Monitoring is also recommended to be performed over an entire day (Trost et al., 2000).

Analysis
The tri-axial acceleration data, collected at a 100 Hz sample rate in raw mode (in g-units), were uploaded and stored, and a statistician from the UKK -institute transferred the data into Excel spreadsheet for further analysis. The collected data were analysed in 6-second epochs and the mean amplitude deviation test. An ANOVA and Tamhane's post-hoc test were used to compare the 7th, 8th and 9th graders in terms of their self-reported data on the frequency and time spent doing their PE homework. The girls were further divided into 2 PE homework groups: Those who did not do PE homework at all or did it only once a week and those who did it at least twice a week. Independent-sample T-tests were used to compare two groups' daily running and total steps with the objectively measured PA levels. For the independent-samples t-tests, Cohen's d was determined by calculating the mean difference between the two groups and then dividing the result by the pooled standard deviation. Cohen's d is the appropriate effect -size measure when 2 groups have similar standard deviations and are of similar size, as in these data. The significance level was set at p < 0.05. The 7th graders recorded greater levels of daily light PA than those from other grades. The 7th grade girls recorded an average daily light PA time of 3 hours 28 minutes (±32 minutes); this is 34 minutes more than the girls from 8th grade (p = 0.000) and 30 minutes more than the girls from the 9th grade (p = 0.000). Regarding daily moderate PA, there was a significant difference between the girls from the 7th and 8th grades (p = 0.017), with a 6-minute time differential benefitting the 8th graders. The half of the girls from the 8th grade reached the Finnish 60-minute PA recommendation; 38% of all girls met the recommended levels of PA. For the 7th grade girls, 35% met the recommendations, while 28% of the 9th grade girls reached the 60-minute PA recommendations every day.

Physical Activity Levels
The 9th grade girls spent more time standing than others. There was a significant difference in time spent standing between the girls from the 7th and 9th grades (p = 0.038), with a 12-minute differential between these grades.
Regarding sedentary behaviour (i.e., sitting and lying), the girls from the 8th girls' (7 hours 50 minutes) was 38 minutes more than that spent by the girls in 8th grade (7 hours 12 minutes) (p = 0.002). Moreover, the girls in the 9th grade recorded a sedentary time of 33 minutes more than that of the girls in the 8th grade (p = 0.012). There was a significant difference in time spent lying down: The 8th graders spent 38 minutes less lying down than the 7th graders (p = 0.000) and 33 minutes less than the grade 9th graders (p = 0.000).

Steps per Day
The girls averaged 8535 steps per day (Table 2)

Self-Reported PE Homework
The 7th and 8th graders did their PE homework more than twice a week on average, 7th grade girls' homework frequency was 2.17 ± 1.81 and 8th grade girls' 2.12 ± 1.66 occasions per week. Meanwhile, the 9th graders reported a frequency of 1.5 ± 1.01 occasions a week (Table 3). PE homework added on average 34 minutes active time to the girls' week. The girls from 8th grade spent an average of 51 minutes per week on PE homework, while the girls from 7th grade spent 20 minutes and the girls from 9th grade spent 32 minutes. The difference between the girls from the 7th and 8th grades was significant (p = 0.027) in terms of total time spent on PE homework per week.  Those who did PE homework twice or more per week had 371 more daily running steps (p = 0.000) and 1067 more total daily steps (p = 0.107) than those who did not do homework as often ( Table 4).
The difference in objectively measured average vigorous daily PA between the groups who did PE homework more often and those who did not was statistically significant (p = 0.000) ( Table 5). The differences comprised an average of 3 to 7 minutes, but the difference favoured the group who did more PE homework.

Discussion
As a part of the Physical Education Homework Study, adolescent girls' objectively measured PA and self-reported PE homework were examined. The girls were more active compared to those in other PA studies (Ruiz et al., 2011;Cohen, Ghosh-Dastidar, & Lytle, 2014;Marques et al., 2015); 38% of the girls reached the 60-minute MVPA recommendations (Currie et al., 2008). The PE homework comprised an average of 34 minutes of active time per week. As a whole, those who did PE homework at least twice per week were more physically active than those who did less PE homework. There was a statistically significant difference in vigorous PA and running steps between the PE homework groups.
On average, girls spent 1 hour 2 minutes daily on moderate PA (3 -6 MET) and 8 minutes on vigorous PA (over 6 MET), with MVPA averaging 1 hour and 11 minutes. Comparing these results to those of a 9-country international study (Ruiz et al., 2011), the amount of MVPA was higher in this study than in others.
According to Ruiz et al. (2011), European adolescent girls' (aged 12.5 -17.5 years) MVPA was 49 minutes per day. In the study of the 9 European countries, 27.5% of girls met the current MVPA recommendations (over 60 minutes per day). In the present study, 35% of girls from 7th grade, 50% of girls from 8th grade and 28% of girls from 9th grade reached the 60-minute recommendation, even if MVPA was on average 1 hour and 11 minutes. The differences between the grades in MVPA comprised a few minutes, but the 7th-grade girls were the most active when light, moderate and vigorous activities were compounded.
Particularly with light PA, there were significant differences between the 7th graders and the other grades. The girls in this study were more active than in the study of Cohen, Ghosh-Dastidar, & Lytle (2014) and in that of Marques et al. (2015). In the present study, the average MVPA of the 8th-grade girls was 75   Okely et al. (2011), which investigated 1518 adolescent girls, only 1.5% met the current PA recommendations; the girls spent less than 5% of their waking hours in MVPA and 60% spent time being sedentary. In this study, PA levels during school hours might be partly due to the policy of the school. It is mandatory to go outdoors during every recess, which means that even the most sedentary students engage in light activity during breaks when moving outside and then back to their classes for the next lesson.
According to methodology and research, there is a great range of variation in step-count guidelines for adolescents: Numbers vary from 9000 to 16,000 steps per day (Silva et al., 2015).  (Gabbei & Hamrick, 2001;Williams & Hannon, 2013). PE homework is also valued in terms of involving students in planning, increasing their autonomy and increasing positive PA experiences (Deci & Ryan, 2000). The study of Behrens et al. (2015), found that the teacher's supportive interaction with participants appear to be important aspect in generating an increase in PA. In their study, when teachers promoted PA, a positive relationship with MVPA was noted (Behrens et al., 2015). If the teacher's interactions and promotions can reach students in daily life (e.g., in the form of PE homework), it could increase their PA levels.

Strengths and Limitations
The strength of this study is the combined use of an objective assessment tool and a self-reported diary questionnaire. The use of accelerometers increases the verifiability of high-quality objective data concerning the quantity and type of PA being performed. Previous studies have confirmed that the accelerometer records all ambulatory activity and that accelerometers are reliable in assessing gait parameters in children and adolescents (Ekelund, Tomkinson, & Armstrong, 2011 Comparing those who did more PE homework with those who did less requires addressing. Those who participated more in PE homework might be more active and more physically competent in the first place. However, PE homework added PA time, at least a small amount, to those who did more homework and to those who did them less. The small sample could be a limitation and might preclude the detection of significance; however, this selected sample only contained girls from one lower secondary school in Finland. The question of transferability regarding all aspects is relevant. In Finland, schools and PE are quite similar nationwide. In addition, in the Western world, PE and teaching styles are comparable. As such, the replication of the study is possible in other schools. There were no withdrawals or dropouts during the study. This could be due to the school structure and the influence schools and teachers have on girls of this age (Kohl II & Cook, 2013).
Studies, in which the students' own teacher conducts the research, imply both advantages and disadvantages. It is possible that granting the consent to participate in a study leads to heightened awareness of targeted behaviours (Beets et al., 2009). Students might also exaggerate their PA behaviour in diaries, with the desire to impress their PE teacher. However, a familiar and confidential relationship between the teacher/researcher and the participants might encourage students to take the study seriously.

Conclusion
Promoting a physically active lifestyle requires reaching beyond the PE lessons.
In this study, PE homework provided an opportunity to embed active assignments into the adolescent girls' daily lives replacing sedentary activities. In addi- This study aimed to determine the influence of organizational and pedagogical changes in school by using PE homework as a tool to influence the PA of adolescent girls. Increasing PA outside of PE has implications also for public health policies, the makers of which could participate in the discussion regarding the use of this practice on a wider scale. Educational policy-makers and school administrators can use these findings to guide curriculum planning and development, as schools are the focal point for interventions designed to incorporate health-enhancing PA in after-school hours.