Changes in healthy behaviour knowledge of rural pupils

Abstract

This study was informed by the body of literature on pupil knowledge and its potential contributions to developing healthy and active lifestyles, along with the focus of various current national physical education standards on cognitive outcomes. This study aimed to investigate the impact of conceptual-based intervention programmes in two rural schools on pupils’ healthy behaviour knowledge. Schools used a similar intervention model of healthy behaviour concepts taught as part of physical education classes, by classroom teachers, and in wellness week activities. The intervention in both schools was based on the Fitness for Life curricular model (i.e. materials for classroom teachers and for physical education teachers). Pupil healthy behaviour knowledge was tracked over time to determine the impact of the intervention programmes. Analysis of variance (ANOVA) was used to compare average percentage of correct responses within school, across time within a year, and across years. ANOVA results showed a significant main effect for grade (F(5, 1152) = 33.10, p < 0.0001, η² = 0.125), school (F(1, 1152) = 75.23, p < 0.0001, η² = 0.061), time (F(1, 1152) = 164.71, p < 0.0001, η² = 0.125), and year (F(1, 1152) = 6.27, p = 0.012, η² = 0.005), and a significant three-way interaction among grade, time, and year (F(13, 1152) = 6.76, p < 0.0001, η² = 0.07). The findings of this study suggest that pupils can acquire healthy behaviour knowledge using several different intervention points of instruction, particularly in rural areas. This study also highlights the need to help pupils in these settings improve their healthy behaviour content knowledge.

Keywords

Fitness for Life conceptual physical education longitudinal

Introduction

The cognitive domain of learning has long been discussed and emphasised in theories and frameworks of teaching and learning. For example, it has been identified as one of the three domains in Bloom’s taxonomy of educational objectives (Bloom et al., 1956), one of the most widely used taxonomies worldwide. Defined as involving knowledge and the development of intellectual skills, and initially including six major categories of cognitive processes, the cognitive domain has been revised over time (Anderson et al., 2001) but still remains a key domain of educational objectives.

The significance of the cognitive domain, or what is often called conceptual knowledge, ‘which includes an understanding of the principles and relationships that permit the learner to apply knowledge usefully and use it for analysis and evaluation’ (Ennis, 2007: 138), has long been recognised within the field of physical education (PE) pedagogy. Scholars in the field have highlighted the need for, and examined, conceptual PE programmes in tertiary settings from the mid-1900s (e.g. Adams and Brynteson, 1992; Corbin, 1969; Slava et al., 1984). Similarly, conceptual PE has been emphasised more recently relative to primary and secondary school settings (e.g. Dale et al., 1998; Dodds, 1987; Johnson and Harageones, 1994; Mohnsen, 1998).

The issue of knowledge and understanding in PE has also been addressed by various PE organisations and through relevant standards and curricula. The American national PE organisation addressed this topic as early as the 1960s (American Alliance for Health, Physical Education, and Recreation, 1969), with a focus on sub-disciplinary concepts such as body mechanics, physiological effects of physical activity, and others. The most recent version of the United States (US) national PE standards (SHAPE America, 2013) continues to place emphasis on the cognitive domain, including knowledge related to movement and performance (Standard 2), as well as physical activity and health-related fitness knowledge (Standard 3). Physical educators are often also required to take, or voluntarily take, initiative in teaching other healthy behaviour knowledge (HBK) content, including, for example, healthy eating. Similarly, in the United Kingdom (UK), the National Curriculum in England: PE Programmes of Study (Department of Education UK, 2013) also places emphasis on the cognitive domain. It includes conceptual understanding targets such as pupils ‘should develop an understanding of how to improve in different physical activities and sports and learn how to evaluate and recognise their own success’ (Key Stage 2). Related to HBK, recommendations include that pupils should ‘develop the confidence and interest to get involved in exercise, sports and activities out of school and in later life, and understand and apply the long-term health benefits of physical activity’ (Key Stage 3).

Physical literacy is now a common term used to describe ‘an individual’s capacity for a physically active lifestyle’ (Longmuir and Tremblay, 2016: 28) as well as desired outcomes from PE programmes. While there are various conceptions of physical literacy in relevant literature, knowledge and understanding, or cognitive capabilities, are often considered a key element of the concept (Whitehead, 2010). However, Hyndman and Pill (2017), who studied the term across multiple contexts, suggest that the concept of physical literacy often prioritises the physical domain, resulting in the marginalisation of the cognitive and affective domains.

The cognitive domain in PE and the teaching of HBK is important because it may facilitate the development and maintenance of healthy and physically active lifestyles. Although HBK alone may not be able to lead to behaviour change, it is a necessary step towards the development of a lifetime of physical activity and healthy living (US Department of Health and Human Services, 2005). Without adequate HBK, for example, it may be considerably more difficult for individuals to become fitness independent and continue physical activity participation as adults, or to maintain a healthy diet. There is some evidence that HBK is positively associated with 5th and 6th grade children’s exercise behaviour (DiLorenzo et al., 1998), secondary pupils’ physical activity levels (Thompson and Hannon, 2012), and secondary pupils’ aerobic fitness and body composition (Williams et al., 2013). Similarly, in another study, Merkle and Treagust (1993) found that grade eight and nine pupils with high health and fitness knowledge scores were more likely to believe in change for their personal health and physical fitness when compared to pupils with low knowledge scores. This may suggest that increased knowledge could possibly lead to increased internality of the locus of control construct. However, contrary to evidence supporting significant positive associations between HBK and various pupil outcomes, the findings of some studies indicate no significant associations between the HBK and physical activity intentions and/or behaviours of middle school (Ferguson et al., 1989) and high school (Haslem et al., 2016) pupils. The conflicting findings indicate that further research is needed in this area.

Despite the potential contributions of knowledge to developing healthy and active lifestyles, relevant research findings consistently show that youth have limited HBK and hold many misconceptions about relevant concepts (Brusseau et al., 2011; Desmond et al., 1990; Hopple and Graham, 1994; Keating et al., 2009; Kulinna, 2004; Merkle and Treagust, 1993; Placek et al., 2001; Stewart and Mitchell, 2003; Thompson and Hannon, 2012; Williams et al., 2013). At the same time, several interventions have been found to be effective in increasing youth’s knowledge levels (Hodges et al., 2014; Ignico and Corson, 1997; Leonetti et al., 2016; McCormick and Lockwood, 2006; Palmer et al., 2005; Prewitt et al., 2015; Trost et al., 2000). Further, there is some preliminary evidence that participation in conceptual rather than traditional high school/secondary PE programmes may be associated with: increased participation in vigorous physical activity for males (Dale and Corbin, 2000); decreased likelihood of engagement in sedentary behaviours one to three years post-programme, particularly in females (Dale and Corbin, 2000; Dale et al., 1998); and positive long-term effects (at 20 year follow up) on pupils’ moderate physical activity patterns and reduced sedentary behaviours (Kulinna et al., 2018).

Given the positive health outcomes associated with increased physical activity and reduced sedentary behaviours (Cooper et al., 2015), there are significant gains to be made by helping youth acquire the requisite knowledge necessary to engage in healthy behaviours. However, youth living in rural settings across the US are more likely to be classified as overweight or obese (Ismailov and Leatherdale 2010; Johnson and Johnson, 2015), potentially due to limited activity options, reduced access to healthcare, and a lower availability of healthy foods (Murimi and Harpel, 2010). The lack of resources can make it challenging for youth to make informed decisions about PA or other healthy behaviours. Since higher levels of conceptual knowledge regarding fitness, physical activity, and healthy eating can have a positive influence on the likelihood of healthy behaviours (Dale and Corbin, 2000; Dale et al., 1998; Kulinna et al., 2018), it is logical to predict that enhancing the HBK of youth living in rural areas can have a significant impact on their ability to make informed decisions regarding their health (Merkle and Treagust, 1993).

Purpose

The HBK of youth living in rural areas in the US has not received much attention by scholars in the field, and this study addresses this gap in the literature. This study was part of a broader multi-year project that aimed to create healthy and active schools in a rural area in the southwestern US. In an attempt to address the health-related challenges (e.g. low physical activity levels, increased overweight and obesity levels, increased risk of drug or alcohol use) that youth in this rural area were facing, a multi-component ecological approach was employed that included the addition of a conceptual aspect to the participating schools’ existing PE programme and across the classroom curriculum and other aspects of the school environment. Accordingly, the purpose of this study was to investigate the impact of school conceptual-based intervention programmes on rural youth’s HBK.

Methods

Participants and setting

The larger project targeted physical activity within and beyond PE, HBK, and school meals, although participating schools used slightly different approaches in addressing targeted areas that took into consideration their unique contexts. Some outcomes (i.e. social capital relationships) from this project have been already published (Jordan et al., 2016, 2018). University Institutional Review Board (IRB) approval as well as school district and principal approvals were obtained prior to starting the study. Also, pupil assent and parental consent forms were collected.

Two schools participated in this study with their grade 3–8 (approximate ages 8 to 13) pupils by implementing different HBK programmes. School 1 was a K–8 Title I school (a school designated with a high concentration of low-income students that receive additional funding from the US federal government to support educational outcomes; www2.ed.gov/programs/titleiparta/index.html) located in the rural southwestern US, with a total enrolment of 519 pupils and more than 90% of the pupil population eligible for free or reduced-price lunch. School 1 participated in the study both in years 1 and 2. School 2 was also a rural K–8 (approximate ages 5–13) Title I school located within 30 miles (50 kilometres) of School 1, with a total enrolment of 634 pupils and more than 90% of the pupil population eligible for free or reduced-price lunch. School 2 participated in the study only in year 2. Table 1 outlines the number of boys and girls in each grade across each measurement time for each school.

Table 1.

Number of boys and girls in each grade across measurement times and within each school.

	School 1				School 2^a
	Year 1		Year 2		Year 1		Year 2
	Girls	Boys	Girls	Boys	Girls	Boys	Girls	Boys
Grade 3	0	0	51	35			50	44
Grade 4	31	34	28	35			51	55
Grade 5	30	38	28	28			45	28
Grade 6	28	40	28	43			58	46
Grade 7	36	24	36	53			43	49
Grade 8	28	24	0	0			53	36
Total	153	160	171	194	0	0	300	258
School total	313		365		0		558

^a School 2 only participated in the project during year 2.

Instruments and administration procedures

During year 1, a 73-item multiple-choice knowledge test was used to examine pupils’ knowledge. This test was based upon the middle school Fitness for Life (Corbin et al., 2010) curriculum model and has been reported to produce valid and reliable data in a similar sample of pupils (Teatro et al., 2013). The content covered in the Fitness for Life curriculum and in the test included the following 14 content areas: (a) active aerobics; (b) physical activity pyramid; (c) active sports/recreation; (d) biomechanics/physiology; (e) body composition; (f) diet/nutrition; (g) FITNESSGRAM (fitness testing programme); (h) flexibility; (i) general knowledge; (j) lifetime physical activity; (k) muscular fitness; (l) peer/family/leadership; (m) self-assessment; and (n) skill. Pupils at School 1 completed this test two times (pre/post) in year 1 of the study. The test was completed in a classroom, in paper and pencil format. It was administered by the PE teacher during two consecutive school periods (each period was approximately 40 minutes long), with an average time-to-completion of the full knowledge test of about 60 minutes.

In year 2, the same test was used but the number of items was reduced in response to teacher and pupil feedback regarding the test’s length. To achieve this, pupil responses to individual items from year 1 were examined for the average percent of correct responses, and items were split into two short versions of the test so the same relative numbers of items that addressed specific content areas were present in both versions. The relative item difficulties across versions were compared and found to be equal (29.8% and 29.7% mean correct response rate). For year 2, both schools used Version A of the test with 37 items. Pupils at both schools completed the knowledge test two times (pre/post) during year 2 (please see Table 2 for the distribution of categories across instruments). Similar to year 1, the test was administered by the PE teacher in a classroom, in paper and pencil format. However, in year 2, given the shortened version of the test, it was administered over a single school period, with an average time-to-completion of about 30 minutes.

Table 2.

Frequency and percent-of-total for each Fitness for Life knowledge test.

	73-item exam		37-item exam		36-item exam
Category	Frequency	Percent	Frequency	Percent	Frequency	Percent
1	5	6.85	3	8.11	2	5.56
2	15	20.55	9	24.32	6	16.67
3	3	4.11	3	8.11	0	0
4	12	16.44	3	8.11	9	25
5	5	6.85	3	8.11	2	5.56
6	2	2.74	1	2.7	1	2.78
7	3	4.11	3	8.11	0	0
8	3	4.11	2	5.41	1	2.78
9	5	6.85	1	2.7	4	11.11
10	4	5.48	3	8.11	1	2.78
11	2	2.74	1	2.7	1	2.78
12	6	8.22	1	2.7	5	13.89
13	4	5.48	1	2.7	3	8.33
14	4	5.48	3	8.11	1	2.78

Intervention

Table 3 provides an outline for the intervention at each school. Intervention components were the same across schools, although the frequency of the conceptual PE sessions taught in each school differed. During the study, the content taught originated from the Fitness for Life textbooks and corresponded with the test items, which were developed and validated specifically for the Fitness for Life curriculum. The Fitness for Life curriculum has materials for classroom teachers available by grade as well as for PE teachers by level (e.g. primary, middle, and secondary). Please see the programme website for more information at https://us.humankinetics.com/pages/fitness-for-life-k-12-resources.

Table 3.

Intervention outline for each school.

	School 1		School 2
	Year 1	Year 2	Year 1	Year 2
Conceptual PE sessions taught in classroom by PE teacher	1 40-minsession / 1 week	1 40-minsession / 1 week	–	1 40-minsession / 2 weeks
Integration of HBK in PE lessons in gymnasium	Encouraged but not monitored	Encouraged but not monitored	–	Encouraged but not monitored
HBK concepts taught in wellness weeks	4 wellness weeks / year (e.g. classroom physical activity breaks, physical education lessons, Eat Well Wednesday, Get Fit Friday, signage, chants, newsletters, resources)	3 wellness weeks / year(e.g. classroom physical activity breaks, physical education lessons, Eat Well Wednesday, Get Fit Friday, signage, chants, newsletters, resources)	–	3 wellness weeks / year(e.g. classroom physical activity breaks, physical education lessons, Eat Well Wednesday, Get Fit Friday, signage, chants, newsletters, resources)
HBK concepts taught by classroom teachers	Daily through FFL physical activity breaks (e.g. plug and play videos with health messages embedded)	Daily through FFL physical activity breaks (e.g. plug and play videos with health messages imbedded)	–	Daily through FFL physical activity breaks (e.g. plug and play videos with health messages imbedded)

HBK: healthy behaviour knowledge; FFL: Fitness for Life.

Professional development and support included workshops led by the research team, a mentor teacher that provided periodic on-site support, the Fitness for Life textbooks and ancillaries, and a resource website. Four workshops were held within the academic year that focused on effective strategies for incorporating HBK into classroom lessons or PE activities. The first workshop for PE teachers focused on the following: overview and use of the Fitness for Life textbooks and ancillaries (e.g. lesson plans, videos, worksheets); scheduling options, yearlong curriculum mapping, and volume control for the conceptual PE component; evaluation (e.g. quizzes, worksheets, notebooks); customisation of content to meet pupil needs; and modelling of teaching and examples of HBK. Similarly, the first workshop for classroom teachers involved: overview and use of the Fitness for Life textbooks and ancillaries; options for scheduling and integrating HBK concepts taught in the classroom; modelling of teaching strategies and examples of HBK integration in the classroom. Subsequent workshops for both PE and classroom teachers involved further modelling of teaching and sharing of examples by the research team, teachers leading short lesson segments from textbooks and/or other activities, discussion of implementation issues and problem-solving, self-reflection activities, and planning for wellness weeks. Beyond the workshops, a mentor teacher provided individualised support to teachers throughout the years, which included email communication and face-to-face meetings, as well as classroom visits and modelling of activities. A resource website was also developed for teachers where they could access additional examples of activities and links to relevant materials. These professional development and support structures were the same across the two schools during the first year they each participated in the project. For School 1 in year 2, the workshops focused more on the planning of activities and wellness weeks as well as model teaching by both members of the research team and teachers at the school.

To monitor the delivery of HBK, classroom teachers reported their participation levels using a short fidelity questionnaire three or four times per year concurrent with school-wide wellness events. The questionnaire asked classroom teachers to report on the total number of times during the week they were able to implement each of nine components of wellness weeks. These included facilitating physical activity breaks that integrated knowledge with and without the use of the instructional videos, using conceptual learning messages from the DVDs in class, posting Fitness for Life instructional signs in the classroom, using instructional chants from the lessons in the classroom, incorporating ideas on ‘Eat Well Wednesday’ or ‘Get Fit Friday’ in the classroom, incorporating Fitness for Life information on his/her personal classroom newsletter, using worksheets from Fitness for Life resources in class, and using the Fitness for Life website. The questionnaire also included an area for the classroom teacher participant to provide any relevant comments about the particular wellness week. The textbook and resources were primarily intended to empower classroom teachers to deliver HBK during specific ‘wellness weeks’ implemented three or four times throughout the academic year. These wellness weeks were implemented school-wide, increasing the emphasis on incorporating HBK into academic classrooms and temporarily shifting the school culture to focus more on healthy eating and PA with posters, events, and newsletters that promoted HBK to families.

Data management and analysis

Test items were binary coded (0 = incorrect; 1 = correct) and summed to report the total number of items each pupil answered correctly. Subsequently, the number of items answered correctly was converted into percentages to standardise responses and allow for comparisons across schools and times.

Analysis of variance (ANOVA) using the general linear model procedure in SAS 9.4 for Windows (Cary, NC) was used to compare the average percentage of correct responses between schools, across time within a year, and across years of the project. Main effects were examined for grade, gender, school, time, and year, with unique effects modelled using type three sums of squares that control for other variables in the model. Thus, differences in response means for a main effect (e.g. school) are adjusted to report influences on the dependent variable that are not found in any other model variable. Two- and three-way interactions for a combination of effects for gender, grade, school, time, and year were modelled to identify differential responses to conditions, meaning cell means and effects are adjusted to identify unique influences across the different levels of the independent variables. Repeated measures were modelled using the variable time, which was the pre- and post-test effect for the percent of correct responses, and the variable year, which reflects the academic year in which the data were collected. There were a total of 83 observations deleted (6.57% of 1264 total observations) across all times and years due to missing data for predictor or response variables. Due to the uneven design created by adding the second school during year 2, simple effects were modelled for differences between measurements at School 1 during year 1. In addition, interaction effects from year 1 are not estimable in the least-squared means analysis; therefore, only year 2 interactions are compared across both schools and time points. These interactions adjust for unequal observations and other effects in the model and identify unique treatment effects across school and time for year 2 of the project.

All hypothesis tests were conducted using a significance level of 0.05 and Scheffe adjustments for multiple comparisons for the least-squared means analysis. Effect sizes from the ANOVA are reported as partial eta-squared (η²) and represent the proportion of the total variance in the dependent variable explained by an individual factor, with other effects from the independent and dependent variables partialled out. Cohen’s d was used to report standardised mean differences between the least-squared means of the three-way interaction of school-by-time-by-year, defining small effect sizes as η² < 0.01 and 0.2 < d < .0.5; medium effect sizes as 0.06 < η² < 0.14 and 0.5 < d < 0.8; and large effect sizes as η² > 0.14 and d > 0.8 (Cohen, 1988).

Results

Teacher fidelity

Classroom teacher fidelity results for wellness week programme implementation showed that the mean number of activities implemented ranged from nine activities (wellness week 1) to 12 activities (wellness week 3) during year 1, and from eight activities (wellness week 1) to 11 activities (wellness week 3) during year 2. This demonstrates that the classroom teachers were consistently implementing multiple aspects of the Fitness for Life primary curricular model for the classroom and the use grew over time across each year. The research team was also on site periodically during the wellness weeks and able to corroborate the implementation through observations and discussions with teachers.

Knowledge scores

Table 4 presents the least-squared means (to account for the unequal number of observations within each cell) of the average number of correct responses across schools, measurement times, and years. Table 5 presents the grade-level means and standard deviations across schools, times, and years, and Figure 1 highlights grade-level means across time and year to represent the differential response indicative of a significant interaction effect.

Table 4.

Average percentage of correct responses across schools, times, and years.

	Year 1		Year 2
	Time 1	Time 2	Time 1	Time 2
School 1^a	34.76% ± 11.55	46.90%* ± 20.27	34.02% ± 14.1	48.66%** ± 19.23
	n = 170	n = 156	n = 182	n = 182
School 2^b			30.89% ± 11.65	37.46%** ± 16.07
			n = 264	n = 255

Data reported represented the mean number of correct responses, ± the standard deviation, and the total number of pupils at each school during the measurement periods across the years of the study. School 2 did not begin the project until year 2.

^a One observation with missing data in time 1 year 1 for School 1 and 14 pupils did not complete time 2 measurement.

^b One observation with missing data in time 1 year 1 for School 2 and nine pupils did not complete time 2 measurement.

*Denotes significant difference (t(11) = 8.54, p < 0.001) for simple contrast between measurement times within year 1.

** Denotes significant differences (p < 0.05 with Scheffe adjustment for multiple comparisons) between time 1 percentage within same year.

Table 5.

Grade-level averages of percentages of correct responses across schools, times, and year.

	School 1				School 2
	Year 1		Year 2		Year 1		Year 2
	Time 1	Time 2	Time 1	Time 2	Time 1	Time 2	Time 1	Time 2
Grade 3			23.0 ± 7.1%	44.1 ± 19.8%			22.6 ± 7.3%	28.7 ± 11.3%
			n = 43	n = 43			n = 46	n = 41
Grade 4	35.9 ± 10.2%	34.6 ± 13.5%	26.5 ± 8.2%	38.0 ± 16.8%			25.2 ± 9.3%	37.8 ± 12.1%
	n = 32	n = 30	n = 30	n = 32			n = 49	n = 49
Grade 5	40.9 ± 11.4%	48.0 ± 12.1%	28.5 ± 8.8%	46.2 ± 15.9%			34.0 ± 8.8%	44.0 ± 15.9%
	n = 37	n = 35	n = 29	n = 27			n = 40	n = 31
Grade 6	28.4 ± 8.7%	38.6 ± 22.9%	41.1 ± 11.1%	57.6 ± 16.5%			31.8 ± 10.0%	38.6 ± 14.4%
	n = 42	n = 42	n = 36	n = 35			n = 49	n = 51
Grade 7	35.3 ± 10.8%	60.8 ± 19.1%	47.8 ± 13.1%	55.1 ± 19.4%			35.1 ± 12.5%	30.3 ± 10.8%
	n = 31	n = 23	n = 44	n = 45			n = 37	n = 41
Grade 8	34.3 ± 13.7%	60.7 ± 16.9%					39.0 ± 13.2%	46.4 ± 22.5%
	n = 27	n = 26					n = 42	n = 42

Data reported represent the mean number of correct responses, ± the standard deviation, and the total number of pupils at each school during the measurement periods across the years of the study. School 2 participated in the project only in year 2.

Figure 1.

Mean percentage (with standard deviation error bars) of correct responses across grade level, time, and year. Data are combined between schools to highlight the interaction between grade, time, and year indicating that different grade levels experienced different treatment effects between time points and across the two years of the study.

Results from the ANOVA showed an overall model effect (F(28, 1152) = 19.35, p < 0.0001, η² = 0.319), indicating differences in the average percent of correct responses between grades, schools, times, and years. There was a significant main effect for grade (F(5, 1152) = 33.10, p < 0.0001, η² = 0.125), school (F(1, 1152) = 75.23, p < 0.0001, η² = 0.061), time (F(1, 1152) = 164.71, p < 0.0001, η² = 0.125), and year (F(1, 1152) = 6.27, p = 0.012, η² = 0.005). There was also a significant two-way interaction between school and time F(1, 1152) = 21.15, p < 0.0001, η² = 0.018), and a significant three-way interaction between grade, time and year (F(13, 1152) = 6.76, p < 0.0001, η² = 0.071). As mentioned above, the unbalanced design resulted in non-estimable functions for three-way interactions between school, time, and year, so the primary treatment effect is examined within year 2 across both schools and time points. Finally, there were no significant effects for gender across school, time or year.

School 1

Analysis of least-squared means for the average percentage of correct responses indicated significant differences between times 1 and 2 in year 1 (t(11) = 8.54, p < 0.0001, d = 0.808; Table 4). In addition, significant improvements were found between times 1 and 2 in year 2 (t(28) = 9.60, p < 0.0001, d = 0.868; Table 4).

School 2

Analysis of least-squared means for the average percentage of correct responses indicated a significant difference between times 1 and 2 in year 2 (t(28) = 4.82, p < 0.0001, d = 0.47; Table 4).

Interactions

There was a significant two-way interaction between school and time F(1, 1152) = 21.15, p < 0.0001, η² = 0.018), which is visible in the different changes in pre–post average scores in Table 4. In addition, there was a significant three-way interaction among grade, time, and year (F(13, 1152) = 6.76, p < 0.0001, η² = 0.071), with differential responses to pre–post testing across grades and between each year (Figure 1).

Discussion

The purpose of this study was to investigate the impact of two school conceptual-based intervention programmes on rural youth’s HBK. Rural settings provide a unique challenge due to reduced access to health-enhancing resources, and improving the HBK of youth living in rural areas can have a positive impact on their health via informed decision making (Merkle and Treagust, 1993). Acknowledging that each school represents a unique context with specific characteristics and needs, the schools participating in this study modified the approach as needed to fit their context (e.g. more conceptual lessons in PE in School 1 and more classroom lessons in School 2).

Our analysis focused on school-level variables pertaining to the adoption of the Fitness for Life model and the instruction of HBK in the PE curriculum as well as in the classroom. Results from the analysis indicated significant main effects in the average number of correct responses between the two schools. The schools modified the teaching of HBK to their unique context, and each school used the developmentally (grade) appropriate Fitness for Life curricular model as their source material. A plausible explanation for this school-level difference is that the data for School 1 are made up of four observations across the two years of the project, whereas School 2 only has two observations as they only participated in year 2 of the project. Comparing the means from School 1 during year 1 to the means from School 2 during their first year of participation in the project (listed as year 2), the average changes reflect the unique situations within each school (Table 4). For example, year 1 at School 1 does not include grade 3 pupils, and this might inflate the school average relative to School 2 (Table 5). It is noteworthy that School 1 would have pupils that saw knowledge questions across all four measurement points (pre–post for both years). However, as shown in Table 4, in School 1, the time 1 means for years 1 and 2 are virtually identical, indicating there was likely little contamination or carry-over of the results from the previous year on the initial measurement of year 2.

Results also revealed a significant main effect for time, suggesting that pupils generally changed their performance on the knowledge tests between the first and second assessments within a year. Follow-up analyses indicated that for both schools and both years for School 1, the second measurement was higher than the first, with significant increases present in School 1 for year 1 and year 2 and in School 2 for year 1. These increases, demonstrated by large effect sizes (η² > 0.14 and d > 0.8) for School 1 and medium effect sizes (0.06 < η² < 0.14 and 0.5 < d < 0.8) (Cohen, 1988) for School 2, are suggestive that regular instruction and application of HBK concepts can lead to meaningful improvements of knowledge. The standardised mean difference (d) between time 1 and time 2 indicates increases of about 0.80 standard deviations for School 1 and 0.50 standard deviations for School 2. As reported in Table 4, this translates into about 7–14 more questions answered correctly on the post-test, compared to the pre-test.

Beyond the main effects, results indicated a significant two-way interaction between school and time. Table 4 shows how each school had a different profile of average number of correct responses between times 1 and 2 between the years of the study. Results also indicated a three-way interaction among grade, time, and year. Table 5 and Figure 1 display grade-level means across time and year, showing that different grades experienced different effects of the conceptual PE curriculum and HBK classroom materials, as evidenced by different rates of change in pre–post averages, differences in the same grade within the same school between year 1 and 2, and when comparing the same grade in the two schools (e.g. the rate of change for grade 4 pupils is different between the two schools, and across time and years). There are many factors that influence test scores, including teacher experience, training, availability of resources, the socioeconomic status of the pupils, the age of the pupils, and the general school climate (Rockoff, 2004; Stewart, 2008). Each of these factors would have different effects when expressed across different schools, measurement times, and years. However, the general trend highlighted from our analysis suggests that independent of other factors, the addition of HBK taught by PE teachers as part of their regular curriculum and by classroom teachers can help increase pupils’ average score on knowledge tests. This finding is consistent with the findings of other studies that used interventions to enhance youth’s HBK levels (Hodges et al., 2014; Ignico and Corson, 1997; Leonetti et al., 2016; McCormick and Lockwood, 2006; Palmer et al., 2005; Prewitt et al., 2015; Trost et al., 2000).

Strengths and limitations

The strengths of this project arise from the long-term, repeated measures design with dedicated PE teachers and classroom teachers providing instruction based on a specific curriculum and assessments aligned with the materials and activities used. We generated data with over 1100 pupils over a two-year period across two schools, which gives us the statistical power to identify differences over time and the influences of our design effects. However, while we were able to collect data on a large sample, there were logistic difficulties in distributing and recording exams to individual pupils across the two years, and there was also a high pupil mobility rate in the area. This made individual-level analyses difficult, and we were unable to model subject-specific variances across the measurement times and study years.

Another limitation of this study includes a lack of a control group that would allow direct comparisons of treatment effects. Unfortunately, we were unable to recruit control schools due to the remote location of participating schools and the scarcity of schools in the same general area.

At the same time, we are confident in the repeated measures design across two schools to provide evidence that conceptual-based PE content can increase pupil performance on knowledge tests within an academic year. However, there are potential confounders within the data, as we adjusted the length of the knowledge test from year 1 to year 2 to accommodate teacher requests for reducing the overall time-to-completion, and some pupils may have been exposed to the same questions in different versions of the test within School 1. While this may raise some questions regarding the overall effectiveness of the intervention, we believe the medium-to-large effect sizes (large effect sizes; η² > 0.14 and d > 0.8) for School 1 and medium effect sizes for School 2 (0.06 < η² < 0.14 and 0.5 < d < 0.8; Cohen, 1988) demonstrate significant average increases in the percentage of correct responses over time, and support the effectiveness of the intervention. The challenges and limitations discussed in this section are not atypical when conducting research in schools, and particularly in rural contexts.

Conclusion

Overall, there were differing effects across the schools for measurement time and study year. Generally, post-test scores were higher than pre-test scores across a year, suggesting that adding or integrating HBK to the PE curriculum and the classroom can be an effective strategy for enhancing knowledge outcomes. Although individual-level responses to the lessons across times or years are unknown, population-averaged effects are indicative that a strategy that best fits the needs of the teachers, pupils, and administrators at a given school is capable of producing significant increases in HBK test scores. Future research studies should adopt strategies to track individual pupils over time and model participant-specific effects of HBK curricular interventions on healthy behaviours. Finally, since teachers are primary stakeholders in the delivery of conceptual PE lessons within PE settings and academic classrooms, research into effective training protocols can be used to improve the quality of instruction and improve pupil HBK outcomes.

Practical implications

This study supports the value of complex and tailored approaches for improving pupils’ HBK. The schools participating in this study had slightly different approaches to incorporating the Fitness for Life curricular model into their school (e.g. classroom lessons, content taught in PE, content taught by classroom teachers, in wellness weeks, etc.) that met their specific needs; however, both schools were successful in improving pupils’ HBK over time. The findings of this study suggest that adequate time and support (e.g. teacher training) needs to be allowed for substantial improvements in pupil HBK to manifest, which may be a function of teachers’ pedagogical content knowledge in HBK, pupils’ limited baseline HBK, and/or other factors. Given the potential of HBK to support the adoption of healthy behaviours, additional information about successful programmes and approaches targeting the development of pupils’ HBK may be useful for teachers, school personnel, and researchers.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the HealthWorks Foundation of Arizona, USA.

References

Adams

Brynteson

(1992) A comparison of attitudes and exercise behaviors of alumni from universities with varying degrees of physical education activity programs. Research Quarterly for Exercise and Sport 63(2): 148–152.

American Alliance for Health, Physical Education, and Recreation (1969) Knowledge and Understanding in Physical Education. Washington, DC: American Alliance for Health, Physical Education, and Recreation.

Anderson

Krathwohl

Airasian

, et al. (2001) A Taxonomy for Learning, Teaching, and Assessing: A revision of Bloom’s Taxonomy of Educational Objectives. New York: Longman.

Bloom

Engelhart

Furst

, et al. (1956) Taxonomy of Educational Objectives: The Classification of Educational Goals. Handbook I: Cognitive Domain. New York: David McKay.

Brusseau

Kulinna

Cothran

(2011) Health and physical activity content knowledge of Pima children. The Physical Educator 68(2): 66–77.

Cohen

(1988) Statistical Power Analysis for the Behavioral Sciences, 2nd edn. Hillsdale, NJ: Lawrence Erlbaum Associates.

Cooper

Goodman

Page

, et al. (2015) Objectively measured physical activity and sedentary time in youth: The international children’s accelerometry database (ICAD). International Journal of Behavioral Nutrition and Physical Activity 12(1): 113–123.

Corbin

(1969) Lecture-laboratory physical education program organization and evaluation. National College Physical Education Association for Men 73: 139–142. Accessed 23 May 2019.

Corbin

Le Masurier

Lambdin

, et al. (2010) Fitness for Life: Elementary School Guide for Wellness Coordinators. Champaign: Human Kinetics.

10.

Dale

Corbin

(2000) Physical activity participation of high school graduates following exposure to conceptual or traditional physical education. Research Quarterly for Exercise and Sport 71(1): 61–68.

11.

Dale

Corbin

Cuddihy

(1998) Can conceptual physical education promote physically active lifestyles? Pediatric Exercise Science 10(2): 97–109.

12.

Department of Education, UK (2013) National curriculum in England: PE programmes of study. Available at: www.gov.uk/education/primary-curriculum-key-stage-2-programmes-of-study (accessed 23 May 2019).

13.

Desmond

Price

Lock

, et al. (1990) Urban black and white adolescents’ physical fitness status and perceptions of exercise. Journal of School Health 60(5): 220–226.

14.

DiLorenzo

Stucky-Ropp

Vander Wal

, et al. (1998) Determinants of exercise among children. II. A longitudinal analysis. Preventive Medicine 27(3): 470–477.

15.

Dodds

(ed.) (1987) Basic Stuff. Series I (revised edn, Vols 1–6). Reston: American Alliance for Health, Physical Education, Recreation and Dance.

16.

Ennis

(2007) 2006 C.H. McCloy research lecture: Defining learning as conceptual change in physical education and physical activity settings. Research Quarterly for Exercise and Sport 78(3): 138–150.

17.

Ferguson

Yesalis

Pomrehn

, et al. (1989) Attitudes, knowledge, and beliefs as predictors of exercise intent and behavior in schoolchildren. Journal of School Health 59(3): 112–115.

18.

Haslem

Wilkinson

Prusak

, et al. (2016) Relationships between health-related fitness knowledge, perceived competence, self-determination, and physical activity behaviors of high school students. Journal of Teaching in Physical Education 35(1): 27–37.

19.

Hodges

Kulinna

Lee

(2014) Teaching healthy behaviour knowledge in primary school physical education. Biomedical Human Kinetics 6(1): 33–29.

20.

Hopple

Graham

(1994) What children think, feel, and know about physical fitness testing. Journal of Teaching in Physical Education 14(4): 408–417.

21.

Hyndman

Pill

(2017) What’s in a concept? A Leximancer text mining analysis of physical literacy across the international literature. European Physical Education Review 24(3): 292–313.

22.

Ignico

Corson

(1997) Effects of heart rate monitor instruction on first-grade children’s HRF knowledge and performance. Journal of the International Council for Health, Physical Education, Recreation, Sport & Dance 33(4): 11–14.

23.

Ismailov

Leatherdale

(2010) Rural-urban differences in overweight and obesity among a large sample of adolescents in Ontario. International Journal of Pediatric Obesity 5(4): 351–360.

24.

Johnson

Harageones

(1994) A health fitness course in secondary physical education: The Florida experience. In Pate

Hohn

(eds) Health and Fitness through Physical Education. Champaign: Human Kinetics, pp.165–172.

25.

Johnson

JA III

Johnson

(2015) Urban-rural differences in childhood and adolescent obesity in the United States: A systematic review and meta-analysis. Childhood Obesity 11(3): 233–241.

26.

Jordan

Lorenz

Stylianou

, et al. (2016) Examining student social capital in a comprehensive school-based health intervention. Journal of Classroom Interaction 51(2): 36–49.

27.

Jordan

Lorenz

Stylianou

, et al. (2018) The role of classroom teacher social capital in a comprehensive school physical activity program. Journal of Teaching in Physical Education 37(2): 218–224.

28.

Keating

Harrison

Jr Dauenhauer

, et al. (2009) Urban minority ninth-grade pupils’ health-related fitness knowledge. Research Quarterly for Exercise and Sport 80(4): 747–755.

29.

Kulinna

(2004) Physical activity and fitness knowledge: How much 1-6 grade pupils know. International Journal of Physical Education 41(3): 111–121.

30.

Kulinna

Corbin

(2018) Effectiveness of secondary school conceptual physical education: A 20-year longitudinal study. Journal of Physical Activity and Health 15(12): 927–933.

31.

Leonetti

Zhu

Chen

(2016) Improving students’ knowledge and values in physical education through Physical Best lessons. European Physical Education Review 23(2): 223–236.

32.

Longmuir

Tremblay

(2016) Top 10 research questions related to physical literacy. Research Quarterly for Exercise and Sport 87(1): 28–35.

33.

McCormick

Lockwood

(2006) College pupil perception of wellness concepts. The Physical Educator 63(2): 78–103.

34.

Merkle

Treagust

(1993) Student knowledge of health and fitness concepts and its relation to locus of control. School Science and Mathematics 93(7): 355–359.

35.

Mohnsen

(1998) Concepts of Physical Education: What Every Student Needs to Know. Reston: National Association for Sport and Physical Education.

36.

Murimi

Harpel

(2010) Practicing preventive health: The underlying culture among low-income rural populations. Journal of Rural Health 26(3): 273–282.

37.

Palmer

Graham

Elliott

(2005) Effects of a web-based health program on fifth grade children’s physical activity, knowledge, attitudes and behaviour. American Journal of Health Education 36(2): 86–93.

38.

Placek

Griffin

Dodds

, et al. (2001) Middle school pupils’ conceptions of fitness: The long road to a healthy lifestyle. Journal of Teaching in Physical Education 20(4): 314–323.

39.

Prewitt

Hannon

Colquitt

, et al. (2015) Effect of personalized system of instruction on health related fitness knowledge and class time physical activity. Physical Educator 72(5): 23–29.

40.

Rockoff

(2004) The impact of individual teachers on pupil achievement: Evidence from panel data. American Economic Review 94(2): 247–252.

41.

SHAPE America (2013) National Standards & Grade-Level Outcomes for K-12 Physical Education. Champaign: Human Kinetics.

42.

Slava

Laurie

Corbin

(1984) Long-term effects of a conceptual physical education program. Research Quarterly for Exercise and Sport 55(2): 161–168.

43.

Stewart

(2008) School structural characteristics, pupil effort, peer associations, and parental involvement the influence of school- and individual-level factors on academic achievement. Education and Urban Society 40(2): 179–204.

44.

Stewart

Mitchell

(2003) Instructional variables and pupil knowledge and conceptions of fitness. Journal of Teaching in Physical Education 22(5): 533–551.

45.

Teatro

Kulinna

Zhu

, et al. (2013) Validating middle school fitness knowledge assessments: The Fitness for Life Test Bank Validation Project. Paper presented at the National American Alliance for Health, Physical Education, Recreation, and Dance Association meeting, Charlotte, NC, 24–26 April 2013.

46.

Thompson

Hannon

(2012) Health-related fitness knowledge and physical activity of high school pupils. Physical Educator 69(1): 71–88.

47.

Trost

Morgan

Saunders

, et al. (2000) Children’s understanding of the concept of physical activity. Pediatric Exercise Science 12(3): 293–299.

48.

US Department of Health and Human Services (2005) Theory at a Glance: A Guide for Health Promotion Practice, 2nd edn. Bethesda, MD, USA: National Cancer Institute, National Institutes of Health.

49.

Whitehead

(2010) Physical Literacy throughout the Lifecourse. London, UK: Routledge.

50.

Williams

Phelps

Laurson

, et al. (2013) Fitness knowledge, cardiorespiratory endurance and body composition of high school pupils. Biomedical Human Kinetics 5(1): 17–21.