Human Movement Science xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder C.A. Batey a, C.A. Missiuna a, B.W. Timmons a, J.A. Hay b, B.E. Faught b, J. Cairney a,⇑ a b
Faculty of Health Sciences, McMaster University, 1280 Main St. W., Hamilton, Ontario L8S 4K1, Canada Faculty of Applied Health Sciences, Brock University, 500 Glenridge Avenue, St. Catharines, Ontario L2S 3A1, Canada
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Developmental Coordination Disorder Self-efficacy Physical activity Developmental disorders and autism
a b s t r a c t Purpose: Affecting 5–6% of children, Developmental Coordination Disorder (DCD) is a prevalent chronic condition. The nature of the disorder – impaired motor coordination – makes avoidance of physical activity (PA) common. The purpose of this study was to examine the effect of barrier and task self-efficacy on PA behavior in children with DCD and a group of typically developing (TD) children. Methods: Children were compared on their perceived ability to complete different intensities and duration of PA (task efficacy) and their confidence in completing PA when faced with everyday barriers (barrier efficacy). An accelerometer was used to record their activity over the subsequent week. Results: Children with DCD were found to have significantly lower task efficacy and barrier efficacy. They also spent significantly less time in moderate to vigorous physical activity (MVPA). Multivariate analyses revealed that gender modified the relationship for both groups. Separate multivariate regressions, were therefore conducted by gender. A direct effect of DCD on PA was observed for boys, but not for girls. Further analyses showed that neither task efficacy nor barrier efficacy influenced the relationship between DCD and PA. Conclusion: Results from this study confirm that children with DCD have lower task and barrier self-efficacy than TD children and that males have lower PA levels than their TD peers; however neither
⇑ Corresponding author. Address: Department of Family Medicine, 175 Longwood Road South, Suite 201, Hamilton, ON L8P 0A1, Canada. Tel.: +1 (905) 525 9140x28506; fax: +1 (905) 905 527 4440. E-mail address: [email protected] (J. Cairney). 0167-9457/$ - see front matter Ó 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.humov.2013.10.003
Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
2
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
task or barrier self-efficacy mediated the relationship between DCD and PA. Ó 2013 Published by Elsevier B.V.
1. Introduction Developmental Coordination Disorder (DCD) is a chronic, neurodevelopmental disorder that is believed to affect approximately 5–6% of school-aged children (APA, 2000; Kadesjo & Gillberg, 1999). Children with DCD have motor coordination abilities that are substantially below age-related norms, and which significantly interfere with academic achievement and/or performance of activities of daily living (American Psychiatric Association (APA), 2000). Previous research has shown children with DCD to be less physically active than typically developing children (Bouffard, Watkinson, Thompson, Causgrove Dunn, & Romanow, 1996; Cairney, Hay, Veldhuizen, Missiuna, & Faught, 2009), and it is believed that this activity-deficit places them at greater risk for obesity (Cairney, Hay, Faught, & Hawes, 2005) and poor physical fitness (Cairney, Hay, Faught, Flouris, & Klentrou, 2007; Schott, Alof, Hultsch, & Meermann, 2007), which in turn places them at increased risk for coronary vascular disease later in life (Faught, Hay, Cairney, & Flouris, 2005). In order to prevent these negative secondary consequences, a better understanding of the reasons why children with DCD participate less in PA in needed. Several psychological factors have been identified in the literature, which may help us to understand the activity-deficit between children with and without DCD. Generalized self-efficacy toward physical activity, which refers to a child’s overall perceived competence and predilection toward physical activity (Hay, 1992), is lower in children with DCD (Cairney, Hay, Faught, Mandigo, & Flouris, 2005), and is strongly associated with reduced participation in free and organized play in these children (Cairney, Hay, Faught, Wade, et al., 2005). A closely related construct – perceived athletic competence – has also been shown to be lower in children with DCD when compared to typically developing children (e.g., Cantell, Smyth, & Ahonen, 1994; Rose, Larkin, & Berger, 1997; Shoemaker & Kalverboer, 1994; Skinner & Piek, 2001). Perceived competence in this domain refers to the child’s perception of mastery of physical skills (Harter, 1982). Finally, when asked to reflect on why they where inactive as children, adults with DCD recall failures at physical pursuits and fear of humiliation from peers as the main reasons for avoiding PA (Fitzpatrick & Watkinson, 2003). While fear of humiliation, poor perceived athletic competence and low self-efficacy toward physical activity may be important factors accounting for overall lower levels of physical activity participation in children with DCD, several limitations in previous research need to be addressed. First, most existing studies of physical activity behavior in children with DCD rely on self-report measures of physical activity (Cantell et al., 1994; Losse et al., 1991; Piek, Baynam, & Barrett, 2006; Piek, Dworcan, Barret, & Coleman, 2000; Skinner & Piek, 2001). Self-report measures often result in an overestimation of physical activity when used with children and adolescents (Welk, Corbin, & Dale, 2000). This could be the result of social desirability bias, but also may be due to the cognitive demands associated with retrospectively estimating intensity and duration of the behavior. Motion sensor monitors provide an objective measure of physical activity and may provide a less biased estimate of physical activity as beliefs and perceptions do not directly affect the estimation of activity. Accelerometers are able to capture random bursts of physical activity and participation in structured and unstructured activities that is difficult to capture in self-reports (Armstrong & Welsman, 2006); however, very few studies have used as them to estimate physical activity in children with DCD (Baerg et al., 2011). The relationship between perceptions of competence or self-efficacy toward physical activity and physical activity behavior has not been established using objective measurements of physical activity. In addition, research on the impact of DCD on physical activity has not yet linked perceived selfefficacy toward physical activity in specific behavioral contexts. Some have argued that measures of self-efficacy should be activity specific (e.g., ‘‘I am good at basketball’’), and connected to specific behaviors, either intention or actual behavior (e.g., I can play basketball today) (Bandura, 2004; Maddux, 1995). Global measures of self-efficacy and competence are broad constructs, capturing general (not behavior specific) attitudes and perceptions toward physical activity, and therefore are likely Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
3
to be more weakly correlated with a specific behavior (Bandura, 2004). Conceived in a behaviorspecific context, self-efficacy is defined in relation to the perceptions of ones ability to complete specific tasks and overcome barriers that might prevent successful engagement in the behavior; these are differentiated as task efficacy and barrier efficacy respectively (Maddux, 1995). As noted previously, existing studies have focused exclusively on the construct of generalized selfefficacy toward physical activity (Cairney, Hay, Faught, Mandigo, et al., 2005; Cairney, Hay, Faught, Wade, et al., 2005) or perceived physical or athletic competence (Cantell et al., 1994; Losse et al., 1991; Piek et al., 2000, 2006; Rose et al., 1997; Shoemaker & Kalverboer, 1994; Skinner & Piek, 2001). We know of no study that has examined the relationship between barrier and/or task efficacy toward physical activity and actual physical activity behavior in children with DCD. Such a study would need to have children (with and without DCD) assess their perceived task and barrier efficacy toward physical activity for in relation to a specific interval of time (e.g., subsequent 7 day period from time of reporting) and to then objectively measure actual physical activity in the same time interval (e.g., subsequent 7-days; Foley et al., 2008). Such an approach would allow for a direct coupling of perception to behavior, something that has not yet been studied in relation to PA behavior in children with DCD. Finally, in addition to addressing these specific limitations, it is important to also consider whether gender may influence the association between DCD and physical activity in this context (Cairney et al., 2009). For example, the predictors of physical activity may be gender-specific in children with DCD (Cairney, Hay, Faught, Mandigo, et al., 2005; Cairney et al., 2009). Indeed, it is not at all uncommon to see gender-specific analyses conducted in relation to physical activity in the DCD literature (Pitcher, Piek, & Barrett, 2002). Any investigation of physical activity behavior in this population should consider the potential moderating effect of gender. In light of the aforementioned limitations and considerations, the following questions will be addressed in this study: 1. Are there differences in PA levels, when measured using accelerometry, between children with and without DCD? 2. Are there differences in perceived task and barrier efficacy with regard to engagement in physical activity between children with and without DCD? 3. Does gender moderate the association between DCD and PA? 4. Are differences in perceived task and barrier efficacy with regard to engagement in physical activity, mediating factors the association between DCD and PA? 2. Methods 2.1. Participants The study is a cross-sectional investigation of a subset of students who participated in a large, prospective cohort study that examined healthy growth and development of children, called PHAST (Physical Health and Activity Study Team; Cairney et al., 2009). The cohort consists of 105 children who were selected from the PHAST study based on their motor proficiency scores; the details of this are described in a previous publication (Cairney, Hay, Veldhuizen, & Faught, 2010). Briefly, all children scoring below the 10th percentile on a test of motor proficiency were invited to participate in a lab study. A sex and school matched sample of children scoring above the 10th percentile were also invited. In total, 126 children (n = 63 with probable DCD; 63 typically developing children) participated in the first wave of the lab study. The self-efficacy toward physical activity measures were not introduced until the second wave (year two) of the lab study. This study is based on the 105 children (62 boys; 43 girls) who participated in this second wave. All children were enrolled in schools in southern Ontario, Canada and were in Grade 8 during the 2008/2009 school year (ages 13–14). 2.2. Protocol Data collection took place between January and June of 2009. Trained research assistants assessed body composition, cardiovascular health, and physical fitness. An occupational therapist administered Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
4
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
the Movement ABC Second Edition (MABC-2; Henderson, Sugden, & Barnett, 2007). All of the research assistants and occupational therapist were blind to the original motor proficiency score of the child. After lab assessments and surveys were completed, children returned home with an accelerometer to wear for seven days. 2.3. Assessment of DCD All children were tested by an occupational therapist using the MABC-2 for two consecutive years within the PHAST study. The M-ABC is a widely accepted tool in DCD diagnosis that measures severity of motor impairment (Croce, Horvat, & McCarthy, 2001). In this study, the MABC-2 (Henderson et al., 2007), the most recent version of the M-ABC, was used to assess motor coordination difficulties. The MABC-2 is divided into three age bands for ages 3–6, 7–10 and 11–16. The third age band (ages 11–16) was used for this study. The MABC-2 is comprised of eight items under three motor skill categories of manual dexterity, aiming and catching, and balance. It has demonstrated good reliability and validity (Brown & Lalor, 2009; Henderson et al., 2007). The 15th percentile is commonly used as a threshold for case identification in research studies (Piek et al., 2000; Schott et al., 2007). The term ‘‘probable’’ (pDCD) was used in this study as not all criteria of the Leeds consensus (Sugden, 2006) were measured when assigning children to their respective groups. For the purpose of this study, children who consistently scored below the 15th on the MABC-2 for both years of assessment (n = 29) were compared to all other children (n = 76). This criterion was established, as it is unlikely that children with DCD would have been able to score above the 15th percentile in either of the 2 testing years. Therefore, using only those children who scored as pDCD in both years is a conservative cut-off for case identification and increases the likelihood that these children have the significant and persistent motor coordination difficulties consistent with a diagnosis of DCD (Table 1). 2.4. Measures 2.4.1. Self-efficacy toward physical activity: Task efficacy The self-efficacy scale (McAuley & Mihalko, 1998), adapted for use with children and adolescents by Foley et al. (2008), was used to measure perceptions of competence toward physical activity. Children rated their confidence to participate in regular physical activity at light, moderate and hard intensities on a scale ranging from no confidence to complete confidence (0–100% confidence) (Foley et al., 2008). As an example, children were asked ‘‘How confident are you that you can complete 10 min of physical activity at a light intensity level three times next week?’’ The physical activity intensity and duration increased in subsequent questions. The questions also specifically asked the child about their anticipated physical activity levels in the upcoming week, relating to when they would be wearing the accelerometer. An overall task value was calculated by summing the scores for each item and then dividing by the number of items, with higher scores representing greater task efficacy (participate in physical activity for longer periods of time at a greater intensity level). Internal consistency of this scale has been shown to be very high with a = 0.95 (Foley et al., 2008). In this sample, the internal consistency was also high (a = 0.97). 2.4.2. Self-efficacy toward physical activity: Barrier efficacy The barrier efficacy scale (McAuley & Mihalko, 1998), again adapted by Foley et al. (2008) for use with children and adolescents, was used to assess perceived barriers that might interfere with intenTable 1 Comparing classification of children as probable DCD in years 1 and 2. Year 1 Year 2 615th percentile on MABC-2 >15th percentile on MABC-2
615th percentile on MABC-2 29 19
>15th percentile on MABC-2 6 51
Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
5
tions to be physically active. Children rated their confidence performing physical activity in the presence of six common barriers experienced by adolescents (Foley et al., 2008;). Children were asked to rate their confidence on a visual scale (on a line from 0% to 100%) in participating in 60 min of PA most days during the next week if they were faced with the situations of bad weather, a lot of school work, good television shows, a lot of activities with family and friends, or if they were feeling tired or sore. The scores for each item are summed and then divided by the number of items: the result is a continuous measure (ranging from 0 to 6) with higher scores indicating higher barrier efficacy. Internal consistency in this scale has proven to be high (a = 0.86) (Foley et al., 2008). In this study, the a is 0.87. 2.4.3. Physical activity – Accelerometer The RT3 Triaxial Actical Accelerometer (Actical, Version 2.0, Mini Mitter, Respironics, 2006) was placed over the right hip of the child and worn for the next 7 days following the laboratory assessments. Methods of accelerometer collection can be found in Baerg et al. (2011). For the week that the child wore the accelerometer, they were given a record book to keep track of when the accelerometer was worn. The child indicated when they put the accelerometer on in the morning and when they took it off at night, as well as if they took it off at any time during the day. When the accelerometer data was collected, it was compared to the log book that each child recorded and cleaned so that only the values corresponding to the time the child indicated that they were wearing the accelerometer were used for analysis. This ensures that we have obtained an accurate record of activity counts and data is not included that does not reflect the activity of the child (for example, activity counts from movement of the accelerometer in the mother’s purse). Using minute-by-minute data, time spent in minutes in each intensity level of physical activity was determined for each child. Sedentary activity was defined as an activity count less than 100/min, light activity was between 100 and less than 1500 activity counts per minute, moderate activity was from 1500 to less than 6500 activity counts per minute, and vigorous activity was defined as 6500 activity counts or greater per minute, based on the validation work of Puyau, Adolph, Vohra, Zakeri, and Butte (2004). Total wear time each day was determined for each child, and an average daily wear time was calculated. Every child wore the accelerometer for at least 3 days, with 90% of the children achieving P6 days of wear. Average amount of time spent per day in sedentary, light, moderate, and vigorous activity was calculated. Wear time was then adjusted for each variable (intensity) by determining the proportion of time spent in each intensity level per day and averaging this for the week. 2.4.4. Body mass index (BMI) BMI was determined by measuring the subject’s weight and height during their visit to the lab at Brock University using the weight of the child in kilograms divided by the square of their height in meters (kg/m2). Using the age of the child and the standards developed for adolescents by the World Health Organization, categories of underweight (
Contents lists available at ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder C.A. Batey a, C.A. Missiuna a, B.W. Timmons a, J.A. Hay b, B.E. Faught b, J. Cairney a,⇑ a b
Faculty of Health Sciences, McMaster University, 1280 Main St. W., Hamilton, Ontario L8S 4K1, Canada Faculty of Applied Health Sciences, Brock University, 500 Glenridge Avenue, St. Catharines, Ontario L2S 3A1, Canada
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Developmental Coordination Disorder Self-efficacy Physical activity Developmental disorders and autism
a b s t r a c t Purpose: Affecting 5–6% of children, Developmental Coordination Disorder (DCD) is a prevalent chronic condition. The nature of the disorder – impaired motor coordination – makes avoidance of physical activity (PA) common. The purpose of this study was to examine the effect of barrier and task self-efficacy on PA behavior in children with DCD and a group of typically developing (TD) children. Methods: Children were compared on their perceived ability to complete different intensities and duration of PA (task efficacy) and their confidence in completing PA when faced with everyday barriers (barrier efficacy). An accelerometer was used to record their activity over the subsequent week. Results: Children with DCD were found to have significantly lower task efficacy and barrier efficacy. They also spent significantly less time in moderate to vigorous physical activity (MVPA). Multivariate analyses revealed that gender modified the relationship for both groups. Separate multivariate regressions, were therefore conducted by gender. A direct effect of DCD on PA was observed for boys, but not for girls. Further analyses showed that neither task efficacy nor barrier efficacy influenced the relationship between DCD and PA. Conclusion: Results from this study confirm that children with DCD have lower task and barrier self-efficacy than TD children and that males have lower PA levels than their TD peers; however neither
⇑ Corresponding author. Address: Department of Family Medicine, 175 Longwood Road South, Suite 201, Hamilton, ON L8P 0A1, Canada. Tel.: +1 (905) 525 9140x28506; fax: +1 (905) 905 527 4440. E-mail address: [email protected] (J. Cairney). 0167-9457/$ - see front matter Ó 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.humov.2013.10.003
Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
2
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
task or barrier self-efficacy mediated the relationship between DCD and PA. Ó 2013 Published by Elsevier B.V.
1. Introduction Developmental Coordination Disorder (DCD) is a chronic, neurodevelopmental disorder that is believed to affect approximately 5–6% of school-aged children (APA, 2000; Kadesjo & Gillberg, 1999). Children with DCD have motor coordination abilities that are substantially below age-related norms, and which significantly interfere with academic achievement and/or performance of activities of daily living (American Psychiatric Association (APA), 2000). Previous research has shown children with DCD to be less physically active than typically developing children (Bouffard, Watkinson, Thompson, Causgrove Dunn, & Romanow, 1996; Cairney, Hay, Veldhuizen, Missiuna, & Faught, 2009), and it is believed that this activity-deficit places them at greater risk for obesity (Cairney, Hay, Faught, & Hawes, 2005) and poor physical fitness (Cairney, Hay, Faught, Flouris, & Klentrou, 2007; Schott, Alof, Hultsch, & Meermann, 2007), which in turn places them at increased risk for coronary vascular disease later in life (Faught, Hay, Cairney, & Flouris, 2005). In order to prevent these negative secondary consequences, a better understanding of the reasons why children with DCD participate less in PA in needed. Several psychological factors have been identified in the literature, which may help us to understand the activity-deficit between children with and without DCD. Generalized self-efficacy toward physical activity, which refers to a child’s overall perceived competence and predilection toward physical activity (Hay, 1992), is lower in children with DCD (Cairney, Hay, Faught, Mandigo, & Flouris, 2005), and is strongly associated with reduced participation in free and organized play in these children (Cairney, Hay, Faught, Wade, et al., 2005). A closely related construct – perceived athletic competence – has also been shown to be lower in children with DCD when compared to typically developing children (e.g., Cantell, Smyth, & Ahonen, 1994; Rose, Larkin, & Berger, 1997; Shoemaker & Kalverboer, 1994; Skinner & Piek, 2001). Perceived competence in this domain refers to the child’s perception of mastery of physical skills (Harter, 1982). Finally, when asked to reflect on why they where inactive as children, adults with DCD recall failures at physical pursuits and fear of humiliation from peers as the main reasons for avoiding PA (Fitzpatrick & Watkinson, 2003). While fear of humiliation, poor perceived athletic competence and low self-efficacy toward physical activity may be important factors accounting for overall lower levels of physical activity participation in children with DCD, several limitations in previous research need to be addressed. First, most existing studies of physical activity behavior in children with DCD rely on self-report measures of physical activity (Cantell et al., 1994; Losse et al., 1991; Piek, Baynam, & Barrett, 2006; Piek, Dworcan, Barret, & Coleman, 2000; Skinner & Piek, 2001). Self-report measures often result in an overestimation of physical activity when used with children and adolescents (Welk, Corbin, & Dale, 2000). This could be the result of social desirability bias, but also may be due to the cognitive demands associated with retrospectively estimating intensity and duration of the behavior. Motion sensor monitors provide an objective measure of physical activity and may provide a less biased estimate of physical activity as beliefs and perceptions do not directly affect the estimation of activity. Accelerometers are able to capture random bursts of physical activity and participation in structured and unstructured activities that is difficult to capture in self-reports (Armstrong & Welsman, 2006); however, very few studies have used as them to estimate physical activity in children with DCD (Baerg et al., 2011). The relationship between perceptions of competence or self-efficacy toward physical activity and physical activity behavior has not been established using objective measurements of physical activity. In addition, research on the impact of DCD on physical activity has not yet linked perceived selfefficacy toward physical activity in specific behavioral contexts. Some have argued that measures of self-efficacy should be activity specific (e.g., ‘‘I am good at basketball’’), and connected to specific behaviors, either intention or actual behavior (e.g., I can play basketball today) (Bandura, 2004; Maddux, 1995). Global measures of self-efficacy and competence are broad constructs, capturing general (not behavior specific) attitudes and perceptions toward physical activity, and therefore are likely Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
3
to be more weakly correlated with a specific behavior (Bandura, 2004). Conceived in a behaviorspecific context, self-efficacy is defined in relation to the perceptions of ones ability to complete specific tasks and overcome barriers that might prevent successful engagement in the behavior; these are differentiated as task efficacy and barrier efficacy respectively (Maddux, 1995). As noted previously, existing studies have focused exclusively on the construct of generalized selfefficacy toward physical activity (Cairney, Hay, Faught, Mandigo, et al., 2005; Cairney, Hay, Faught, Wade, et al., 2005) or perceived physical or athletic competence (Cantell et al., 1994; Losse et al., 1991; Piek et al., 2000, 2006; Rose et al., 1997; Shoemaker & Kalverboer, 1994; Skinner & Piek, 2001). We know of no study that has examined the relationship between barrier and/or task efficacy toward physical activity and actual physical activity behavior in children with DCD. Such a study would need to have children (with and without DCD) assess their perceived task and barrier efficacy toward physical activity for in relation to a specific interval of time (e.g., subsequent 7 day period from time of reporting) and to then objectively measure actual physical activity in the same time interval (e.g., subsequent 7-days; Foley et al., 2008). Such an approach would allow for a direct coupling of perception to behavior, something that has not yet been studied in relation to PA behavior in children with DCD. Finally, in addition to addressing these specific limitations, it is important to also consider whether gender may influence the association between DCD and physical activity in this context (Cairney et al., 2009). For example, the predictors of physical activity may be gender-specific in children with DCD (Cairney, Hay, Faught, Mandigo, et al., 2005; Cairney et al., 2009). Indeed, it is not at all uncommon to see gender-specific analyses conducted in relation to physical activity in the DCD literature (Pitcher, Piek, & Barrett, 2002). Any investigation of physical activity behavior in this population should consider the potential moderating effect of gender. In light of the aforementioned limitations and considerations, the following questions will be addressed in this study: 1. Are there differences in PA levels, when measured using accelerometry, between children with and without DCD? 2. Are there differences in perceived task and barrier efficacy with regard to engagement in physical activity between children with and without DCD? 3. Does gender moderate the association between DCD and PA? 4. Are differences in perceived task and barrier efficacy with regard to engagement in physical activity, mediating factors the association between DCD and PA? 2. Methods 2.1. Participants The study is a cross-sectional investigation of a subset of students who participated in a large, prospective cohort study that examined healthy growth and development of children, called PHAST (Physical Health and Activity Study Team; Cairney et al., 2009). The cohort consists of 105 children who were selected from the PHAST study based on their motor proficiency scores; the details of this are described in a previous publication (Cairney, Hay, Veldhuizen, & Faught, 2010). Briefly, all children scoring below the 10th percentile on a test of motor proficiency were invited to participate in a lab study. A sex and school matched sample of children scoring above the 10th percentile were also invited. In total, 126 children (n = 63 with probable DCD; 63 typically developing children) participated in the first wave of the lab study. The self-efficacy toward physical activity measures were not introduced until the second wave (year two) of the lab study. This study is based on the 105 children (62 boys; 43 girls) who participated in this second wave. All children were enrolled in schools in southern Ontario, Canada and were in Grade 8 during the 2008/2009 school year (ages 13–14). 2.2. Protocol Data collection took place between January and June of 2009. Trained research assistants assessed body composition, cardiovascular health, and physical fitness. An occupational therapist administered Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
4
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
the Movement ABC Second Edition (MABC-2; Henderson, Sugden, & Barnett, 2007). All of the research assistants and occupational therapist were blind to the original motor proficiency score of the child. After lab assessments and surveys were completed, children returned home with an accelerometer to wear for seven days. 2.3. Assessment of DCD All children were tested by an occupational therapist using the MABC-2 for two consecutive years within the PHAST study. The M-ABC is a widely accepted tool in DCD diagnosis that measures severity of motor impairment (Croce, Horvat, & McCarthy, 2001). In this study, the MABC-2 (Henderson et al., 2007), the most recent version of the M-ABC, was used to assess motor coordination difficulties. The MABC-2 is divided into three age bands for ages 3–6, 7–10 and 11–16. The third age band (ages 11–16) was used for this study. The MABC-2 is comprised of eight items under three motor skill categories of manual dexterity, aiming and catching, and balance. It has demonstrated good reliability and validity (Brown & Lalor, 2009; Henderson et al., 2007). The 15th percentile is commonly used as a threshold for case identification in research studies (Piek et al., 2000; Schott et al., 2007). The term ‘‘probable’’ (pDCD) was used in this study as not all criteria of the Leeds consensus (Sugden, 2006) were measured when assigning children to their respective groups. For the purpose of this study, children who consistently scored below the 15th on the MABC-2 for both years of assessment (n = 29) were compared to all other children (n = 76). This criterion was established, as it is unlikely that children with DCD would have been able to score above the 15th percentile in either of the 2 testing years. Therefore, using only those children who scored as pDCD in both years is a conservative cut-off for case identification and increases the likelihood that these children have the significant and persistent motor coordination difficulties consistent with a diagnosis of DCD (Table 1). 2.4. Measures 2.4.1. Self-efficacy toward physical activity: Task efficacy The self-efficacy scale (McAuley & Mihalko, 1998), adapted for use with children and adolescents by Foley et al. (2008), was used to measure perceptions of competence toward physical activity. Children rated their confidence to participate in regular physical activity at light, moderate and hard intensities on a scale ranging from no confidence to complete confidence (0–100% confidence) (Foley et al., 2008). As an example, children were asked ‘‘How confident are you that you can complete 10 min of physical activity at a light intensity level three times next week?’’ The physical activity intensity and duration increased in subsequent questions. The questions also specifically asked the child about their anticipated physical activity levels in the upcoming week, relating to when they would be wearing the accelerometer. An overall task value was calculated by summing the scores for each item and then dividing by the number of items, with higher scores representing greater task efficacy (participate in physical activity for longer periods of time at a greater intensity level). Internal consistency of this scale has been shown to be very high with a = 0.95 (Foley et al., 2008). In this sample, the internal consistency was also high (a = 0.97). 2.4.2. Self-efficacy toward physical activity: Barrier efficacy The barrier efficacy scale (McAuley & Mihalko, 1998), again adapted by Foley et al. (2008) for use with children and adolescents, was used to assess perceived barriers that might interfere with intenTable 1 Comparing classification of children as probable DCD in years 1 and 2. Year 1 Year 2 615th percentile on MABC-2 >15th percentile on MABC-2
615th percentile on MABC-2 29 19
>15th percentile on MABC-2 6 51
Please cite this article in press as: Batey, C. A., et al. Self-efficacy toward physical activity and the physical activity behavior of children with and without Developmental Coordination Disorder. Human Movement Science (2013), http://dx.doi.org/10.1016/j.humov.2013.10.003
C.A. Batey et al. / Human Movement Science xxx (2013) xxx–xxx
5
tions to be physically active. Children rated their confidence performing physical activity in the presence of six common barriers experienced by adolescents (Foley et al., 2008;). Children were asked to rate their confidence on a visual scale (on a line from 0% to 100%) in participating in 60 min of PA most days during the next week if they were faced with the situations of bad weather, a lot of school work, good television shows, a lot of activities with family and friends, or if they were feeling tired or sore. The scores for each item are summed and then divided by the number of items: the result is a continuous measure (ranging from 0 to 6) with higher scores indicating higher barrier efficacy. Internal consistency in this scale has proven to be high (a = 0.86) (Foley et al., 2008). In this study, the a is 0.87. 2.4.3. Physical activity – Accelerometer The RT3 Triaxial Actical Accelerometer (Actical, Version 2.0, Mini Mitter, Respironics, 2006) was placed over the right hip of the child and worn for the next 7 days following the laboratory assessments. Methods of accelerometer collection can be found in Baerg et al. (2011). For the week that the child wore the accelerometer, they were given a record book to keep track of when the accelerometer was worn. The child indicated when they put the accelerometer on in the morning and when they took it off at night, as well as if they took it off at any time during the day. When the accelerometer data was collected, it was compared to the log book that each child recorded and cleaned so that only the values corresponding to the time the child indicated that they were wearing the accelerometer were used for analysis. This ensures that we have obtained an accurate record of activity counts and data is not included that does not reflect the activity of the child (for example, activity counts from movement of the accelerometer in the mother’s purse). Using minute-by-minute data, time spent in minutes in each intensity level of physical activity was determined for each child. Sedentary activity was defined as an activity count less than 100/min, light activity was between 100 and less than 1500 activity counts per minute, moderate activity was from 1500 to less than 6500 activity counts per minute, and vigorous activity was defined as 6500 activity counts or greater per minute, based on the validation work of Puyau, Adolph, Vohra, Zakeri, and Butte (2004). Total wear time each day was determined for each child, and an average daily wear time was calculated. Every child wore the accelerometer for at least 3 days, with 90% of the children achieving P6 days of wear. Average amount of time spent per day in sedentary, light, moderate, and vigorous activity was calculated. Wear time was then adjusted for each variable (intensity) by determining the proportion of time spent in each intensity level per day and averaging this for the week. 2.4.4. Body mass index (BMI) BMI was determined by measuring the subject’s weight and height during their visit to the lab at Brock University using the weight of the child in kilograms divided by the square of their height in meters (kg/m2). Using the age of the child and the standards developed for adolescents by the World Health Organization, categories of underweight (
