Effectiveness of a Mastery Climate Cognitive-Motor Skills School-based Intervention in children living in poverty: Motor and Academic performance, self-perceptions, and BMI

ABSTRACT

Background:

Children living in poverty often are exposed to situations that negatively affect various aspects of their development. So, it is important to propose interventional programs that minimize the negative effects of living in poverty on these children.

Objective:

To verify the impact of a mastery climate cognitive-motor skills school-based intervention on academic and motor performance, body mass index, and perceptions of competence and self-worth in children living in poverty. Method: 280 children (50% girls), 7- to10-years-old, from Ceará state – Brazil, participated in this study. Intervention (IG, N = 140) and comparison (N = 140) groups were composed. The Test of Gross Motor Development – 2, the Self-Perception Profile for Children, the School Performance Test (reading, writing, math), and the Body Mass Index (BMI) were used. Children in the IG participated in a mastery climate intervention and children in the CG in a recreational sports program for 12 weeks (36 sessions /140 min each session/ three times a week) with 60 min of academic activities, 20 min recess and snacks, and 60 min of motor practices.

Results:

The intervention group showed significant improvement from pre- to post-test, and higher motor, academic (reading, math), perceived competence, social acceptance, and self-worth scores than the comparison group at pos-test (p from 0.031 to <0.001).

Conclusion:

Mastery climate cognitive-motor intervention positively impacts the motor and academic performance and the self-perceptions of the children living in poverty, reinforcing the need for these interventions to helping children become more proficient and assured about their competences.

KEYWORDS:

• School-based program
• motivation climate
• perceived competence
• nutritional status
• fundamental motor skills
• scholar performance

Introduction

Achieving motor proficiency and successfully responding to academic demands and social interactions are critical aspects of child development. Appropriate learning opportunities implemented in mastery motivation climates – high autonomy-supportive climates with a child-centered approach to foster motivation to learn, provide a foundation for children functioning in achievement contexts (Rudisill and Johnson 2018; Valentini, Rudisill, and Goodway 1999). Mastery motivational climate is related to increases in motor performance (Barkoukis, Koidou, and Tsorbatzoudis 2010; Hastie et al. 2019; Valentini and Rudisill 2004a), physical activity (Wadsworth et al. 2017), physical play (Parish, Rudisill, and Onge 2007), perceived competence (Valentini and Rudisill 2004b), social acceptance (Píffero and Valentini 2010), and verbal recall (Valentini et al. 2017).

It is important to note that some previous studies that implemented mastery motivational climate were conducted with children living in poverty. Positive outcomes were reported for motor skill proficiency (Martin, Rudisill, and Hastie 2009; Robinson and Goodway 2009; Valentini et al. 2017; Valentini and Rudisill 2004a, 2004b), perceived physical competence (Robinson, Rudisill, and Goodway 2009; Valentini and Rudisill 2004b), and verbal recall (Valentini et al. 2017). However, most studies were conducted in WEIRD (Western, educated, industrialized, rich, and democratic) countries, with children living in rural areas (Martin, Rudisill, and Hastie 2009), at-risk enrolled in governmental programs (Robinson and Goodway 2009; Robinson, Rudisill, and Goodway 2009), and with delays (Valentini and Rudisill 2004b) with no specific information regarding the socioeconomic status of those families. Specifically, in developing countries, such as Brazil, where 45.3% (18.8 million) of the children and adolescents live in poverty, with 9.1 million (21.9.%) in extreme poverty, and 6% (2.9 million) are engaging in labor to provide for themselves or families (Abrinq Foundation 2020) inclusive in hazardous occupations (DeGraff, Ferro, and Levison 2014).

Children in vulnerability prematurely enter the work environment (Abrinq Fundation 2020; Fonseca et al. 2013) and are more likely to school failure and dropping out (Stahl, Vives, and Asibey 2015). Motor delays (Valentini, Clark, and Whitall 2015) and feelings of being less capable (Nobre et al. 2015) are often reported. The families have fewer resources to promote development; consequently, children rely on a fragile social recognition system to elaborate beliefs about their abilities (Gontijo and Medeiros 2009; Nobre et al. 2015; Pereira 2010). Furthermore, the seminal work on mastery motivational climate provided evidence for its effectiveness to promote a positive attitude toward learning and effort, cognitive engagement, use of effective learning strategies, self-instruction and self-monitoring, and academic achievement in the classroom content (Ames and Archer 1988). More recently, research suggested that students enrolled in mastery climates use metacognitive strategies more often, demonstrate less procrastinated behaviors (Wolters 2004), show high adaptative motivational engagement (Pintrich and Maehr 2004), and perceived that teacher support their autonomy (Tor Stornes and Thormod 2008). Although these originals work provided insights to implement low-cost strategies in the classroom to guide children to high achievement, to the best of author’s knowledge, no studies addressed this goal in children from developing countries.

Moreover, across different countries, the effectiveness of the mastery motivational climate on children's self-perception has been investigated mainly in the physical domain (Robinson, Rudisill, and Goodway 2009; Valentini and Rudisill 2004b); children's social acceptance and self-worth, continuously overlooked – here we addressed these effects, an original contribution of the present study. Considering the enduring relationship between poverty and obesity (Conde and Monteiro 2014; Drewnowski and Rolls 2012), even longitudinally (Lee et al. 2014), and the effectiveness of physical activity programs in reducing overweight prevalence (Bogart et al. 2017), it is also critical to address this issue to develop prevention strategies. Further, we also concern with being underweight since all the families live in poverty (Silva and Nunes 2015); Finally, to the author's knowledge, there is no report of mastery motivational climate impact for underweight children. Therefore, we also investigate this parameter in the present study.

This study's objective was to investigate the effectiveness of a mastery climate cognitive-motor skills school-based intervention on motor and academic performance, body mass index (BMI), self-perceptions, and global self-worth of children living in poverty. Our secondary objective was investigating the relationship of those variables at pre- and post-intervention. Significant and positives changes in motor and academic performance and self-perceptions regardless of sex and age are expected for all children; no change is expected for BMI. We also assumed that children in the mastery climate would show significantly higher scores in all variables in the post-intervention than the children in the comparison group.

Method

Participants and context

The study's sample estimation was conducted considering the Generalized Estimation Equation (GEE) model as the primary analysis. To estimate the sample size for the GEE, we used the PASS – Power analysis & Sample Size (2021) and chose the model that takes into consideration the slope of multiple groups in a repeated measures design for continuous outcome sampling (Ahn, Heo, and Zhang 2014), that represents our study design. There is a relatively high risk of type I and type II errors in interventional studies since several intervenient factors are hard to control in a quasi-experimental design. In both situations, there are negative repercussions, such as time loss, cost, and ethical problems (Serdar et al. 2021). For minimizing the probability of committing both errors, in this study, we decide, prudently, to assume a restricted cut off for alpha and power (i.e. 0.01 and 0.99, respectively) and consider a priori, a low expected effect size (r = .20) for calculating the sample size (Serdar et al. 2021). The estimation was conducted considering 2-groups (intervention and comparison) X 2-times design (pre- and post-tests), a power of .99, a significance level of .001, an effect size r = .20, and an estimation of probable missing data proportion across measurement times of 30%, a sample size indicated was 280 children.

Therefore, participants in the present study were 280 children (7- to 8-year-old: 70 boys and 70 girls; 9- to 10-year-old: 70 boys and 70 girls) from 32 public schools, in three cities, located in Ceará, Brazil. The cities were located in different regions and were chosen to represent Ceara's cultural diversity - mountain, coast, and semiarid region. The Human Development Index (HDI) was adopted in the present study to provide information regarding the quality of life in each region – the HDI includes in its measure life expectancy, education (years of schooling and school entering age), and the standard of living (gross national income per capita) in each city. The state of Ceará occupied the 15th position in the HDI national rank of 27 states (HDI = 0.726). categorized as medium, and it was similar across the three cities (mountain city: HDI = 0.620; coast city: IDH = 0.648; semiarid city: IDH = 0.659) (United Nations Development Program 2016).

We adopted the following inclusion criteria: children attending public schools from families living in poverty supported economically by the federal government – state-sponsored social programs. Children with developmental disability (e.g. Autism Spectrum Disorder, Attention-Deficit Hyperactivity Disorder), motor disability (e.g. cerebral palsy), intellectual disability, genetic disorder (e.g. muscular dystrophy), emotional disturbance (e.g. bipolar disorder), and language, hearing, and visual impairments reported by parents, teachers, or caregivers, were excluded from the study. From the total sample, two groups were composed.

The Mastery Climate Group (MCG; n =  140) was formed by children attending only public schools who were not currently or previously participating in any other program. The Comparison Group (CG; n = 140) was formed by children attending youth recreational sports program and academic reinforcement sponsored by the state. The groups were paired by sex and age. Parents’ consent was obtained for each child, and each child verbally agreed to participate.

From pre- to post-test, 69 children discontinued the study participation; 23 in the MCG – 9 boys (7-8years-old n = 4; 9–10 years-old n = 5) and 14 girls (7-8 years-old n = 8; 9–10 years-old n = 6) and, 46 in the CG – 11 boys (7-8 years-old n  = 11; 9–10 years-old n = 9) and 26 girls (7-8 years-old n = 12; 9–10 years-old n = 14). Parents’ reasons were difficulties taking the children to the program, children's chores – taking care of home and siblings, children labor in familiar agriculture and crafting and at seasonal business agriculture, and address changes; some families did not report the reasons. Figure 1 presents the study design.

Figure 1. Study design.

The families’ income was based on small trade and craft businesses, artisanal fishing, and familiar agriculture. Household monthly income was less than the minimum wage for 59.5% of the families; a high rate of householders’ unemployment was reported (72%) for at least one parent. All families lived in areas with low resources (i.e. lack of parks and health centers) and were monetarily supported by the government program. Most of the parents had less than four years of formal education (96%). None of the schools have a gymnasium; six schools had poorly maintained sports courts, the others had gravel yards. None of the schools had nearby parks, but all had wasteland or partially abandoned land where children got together to play after school; Classroom teachers conducted physical education lessons at schools, none of the schools had physical education teachers for this age group. Physical education activities were mostly recreation games, once a week, and usually, boys played soccer and girls small games; some days were play in the playground.

Instruments

Children’s motor proficiency was assessed using the Test of Gross Motor Development, second edition – TGMD-2 (Ulrich 2000), validated for the Brazilian children (Valentini 2012). The TGMD-2 subtests raw scores for Locomotor – LOC (run, gallop, hop, leap, jump, slide) and Object Control – OC (strike, dribble, catch, kick, throw, roll) skills were used. The BMI was obtained using a digital balance and a portable stadiometer; and calculated following the World Health Organization recommendations (WHO 2007). The academic performance was assessed using the School Performance Test – SPT (Stein 1994). The SPT consists of three subtests, writing, reading, and math, organized progressive order of difficulty to assess the child independently of the grade levels. The present study used raw scores and performance categorization based on the percentiles (inferior, below average, average, above average, superior) (Stein 1994). The perceptions of competence and a general sense of self-worth were assessed using the Self-Perception Profile for Children – SPPC (Harter 1982) validated for Brazilian children (Valentini et al. 2010). The subscales (school competence, athletics competence, social acceptance, behavioral conduct) and the global self-concept scores were used.

Intervention design and implementation

A 12-week (3 times per week / 36 lessons /140 min session) intervention was design following the mastery climate guidelines (Valentini, Rudisill, and Goodway 1999) and implemented for the mastery motivational climates group. With the help of two undergraduate students, a physical education teacher, and three classroom teachers delivered the mastery motivational climates cognitive-motor intervention. Two physical education teachers conducted the sport-sponsored program; three classroom teachers delivered the academic reinforcement. The researchers did not manipulate the climate in the comparison group; the climate and activities were observed and reported. Along with the intervention, the teacher's prevailing climate for the CG was teacher-centered, with children having free choice of games and recreation tasks during the lessons.

Intervention group: mastery climate cognitive-motor skills school-based intervention

Like previous studies, mastery climate strategies were implemented according to the TARGET structure – Tasks, Authority, Recognition, Group, Evaluation, and Time (Ames 1992; Valentini and Rudisill 2004a, 2004b). Regarding the task dimension, the motor practice was organized in stations and games. Five to seven stations were implemented, containing a diverse range of body and space awareness, balance, locomotor skills, and ball skills tasks. Novelty and different levels of difficulties were presented in every lesson. The tasks’ demands were distinguished by manipulating distances, heights, stability, paths, platforms, weight, size, colors, shapes, and implements. The last part of the lesson incorporates the skills learned that day in small and large games.

A diverse range of tasks was implemented for reading, writing, and math reinforcement; the content was aligned with the school's curriculum. Strategies consistent with the principles of interactive teaching (Roth and Dabrowski 2014) were used for reading (e.g. comprehension, text structure, vocabulary, punctuation awareness, independent and guided reading, reading fluency), writing (e.g. organize text, use appropriate grammar, write about personal topics, express ideas, use words beyond grade-level), and math (e.g. basic operations, fractions, decimals) skills. The strategies allow teaching across different grades using notepads, books, games, and puzzles.

The lessons incorporate tasks choices regarding the authority dimension, establish cooperative protocol and regulations, make decisions, and take responsibility. Often, children were invited to elect their favorite activities and tasks and participate in the design of new ones. For the recognition dimension, the teacher encouraged children's efforts, achievements, and progress. Parents and schoolteachers received reports about children's improvements and were invited to attend the intervention, observe the practice, help with motor and academic activities, and encourage children to persist.

Children were organized according to different motor proficiency, age, sex, goals, and interest for group dimension. Children had the autonomy to choose peers to practice motor skills, reading, writing, and sol math problems during the lesson; however, the nature of a setting led children to change groups and interact with other children regularly. Circle time activities, interpersonal conflict resolution, and cooperative peer activities were strategies to improve peer interactions. The children were also instructed to respect individual differences, feelings, and capacities; and cooperate with others with more limitations.

Evaluation strategies were implemented to help children to understand the importance of efforts to improve performance. Children self-monitored their practice using cues words for each skill. Feedback was provided by teachers using personal parameters to avoid social comparisons. Children were encouraged to assess their levels of abilities and progress and establish goals to improve performance. School teachers, administrators, and parents received information regarding children's development, learning, and new achievements.

Time strategies included providing adequate and individual time to learn the tasks. The organization of the tasks in the stations potentializes the use of the few pieces of equipment, increases participation in the motor activities, and respects each child's learning time. Motor skills activities were organized in stations, and games incorporated the skills practice at the stations. A physical educator teacher conducted the interventions with the help of two undergraduate students. The lessons began with instructions, modeling, and cue words about the skills’ motor criteria. The station’s motor practice in was conducted in small groups (4–5 children each station), and team games were played in small or large groups. Activities included conventional equipment (e.g. bows, targets, baskets, balls of different sizes and weights, rods, rackets, ropes, cones) and recyclable materials (e.g. plastic bottles and wood chips) and equipment made by the children and the teachers.

The academic reinforcement content was related to reading, writing, and math skills. The lesson begins with content instructions and how it was related to the school's content. The activities comprised group and individual tasks focusing on reading and writing, oral dictation, math games (addition, subtraction, multiplication, and division), and problem-solving. Materials such as notepads, textbooks, logic games, and puzzles were used. The sessions occurred in classrooms and patio areas. Three classroom teachers delivered all sessions.

Comparison group: State-sponsored recreational sports program & academic reinforcement

The comparison group participated in a state-sponsored program that provided recreational sports, arts, academic reinforcement (reading, writing, math), and crafts that follow the scholar curriculum. The prevailing climate in the program was described along with the TARGET structure. Researchers did not manipulate this climate, just reported.

Regarding task dimension, soccer, dance, and capoeira (Afro-Brazilian martial art) practices were organized in groups. The dance lessons were choreographies practice mainly for the girls (Funk and Axé -Brazilian dance with Samba and ludic elements). Modeling was expensively used in the dance lessons but with few verbal instructions. The soccer lessons were for boys and girls, and soccer fundamental skills (passing, receiving a ball, trapping, dribbling, shooting) and matches were practiced. Children had the free choices during the recreational activities (e.g. tag games, play with toys children brought from home); the teacher’s role was to supervise children’s behaviors and mediate conflicts. No lesson plan and a variety of tasks were observed. Few pieces of equipment were used; more resources were available for the soccer practices. The academic reinforcement's focus was the proper completion of the homework sent by the schools and provided support for children's individual needs; the content was aligned with the school's curriculum. The tasks were group activities, dictation, math resolution problems (sum, subtraction, multiplication, division), handwriting, and craft activities (e.g. drawing, painting).

Regarding the authority dimension, the lessons incorporate direct instruction – the lesson structure, sequence, and activities were led by teachers in soccer, dance, and academic activities. Children's choices and collaboration during soccer, dance, and academic activities were not taken into consideration. However, in recreational games, children had the freedom to do whatever they would like to play. Teachers also implemented a behavioral protocol without children's insights nor cooperation.

For the recognition dimension, overall, the teachers encouraged children's efforts, achievements, and progress. Social comparisons were prevalent during the academic reinforcement, dance, and soccer practices; teachers also praised the children with higher successes. Along with the academic reinforcement lessons, children were participative, and teachers were responsive to the child’s needs. During soccer and dance practices, children had no voice.

Regarding groups, children were organized in pairs and large groups for the majority of soccer practices. Dance choreographies were practiced in smalls groups. A competitive climate was predominant for the soccer matches and dance performances.

Evaluation strategies were mainly positive feedback to motivate children during the dance, soccer, and academic lessons. During recreation activities, no evaluation strategies were observed. During academic reinforcement and soccer practices, teachers provided individual evaluation and individual performance feedback. Time strategies included providing group time for the completion of the tasks.

Procedures

The university ethics committee approved the study. The present study is the second phase of a large study conducted in the northeast of Brazil. The study's first phase was to examine social-cultural factors related to motor delays in children living in poverty (Nobre, Valentini, and Rudisill 2020). The second phase was to implement interventions for children whose motor delays were detected. Initially, we contacted the board of education and school administrators from three cities of Ceará, Brazil; 54 schools responded positively to the study, 32 schools met the criteria and were included in the first and second phase of the study; the cities were chosen due to its representativeness of northeast culture. In a meeting with the school administrator and staff, the researchers explained the study's aims and procedures, and parents were informed about the research. Parents or legal guardians were contacted and received all information about the research procedures.

One of the researchers was responsible for obtaining contextual information about schools and families. He visited all children's schools and collected information about physical education lessons and curriculum, facilities, and equipment; nearby parks were also visited. He was also responsible for interviewing the parents and helping parents to complete the parents’ sociodemographic survey. The survey contained questions regarding family occupation, income, number of adults and children at home, children's chores. All information was used to describe the context. This researcher, a physical education teacher with long experience with assessment, lived in each region for six months, conducted all the assessments with the support of a physical education teaching assistant, and the intervention with the support of two undergraduate students and classroom teachers.

The TGMD-2 assessments conducted individually in schools’ yards were video recorded for further analysis; two experienced professionals independently coded the children’s performances. The professionals were physical education teachers, doctoral students trained by one of the researchers with extensive previous experience coding the TGMD-2; the inter-rater reliability was high at pre-test (ICC values for LOC skills = .89 and OC skills = .93) and post-test (ICC values for LOC skills = .97 and OC skills = .98). The School Performance Test (administrated by the classroom teachers) and the self-perception profile for children SPPC (administrated by one of the researchers) were conducted in one quiet classroom. The families were informed regarding the right to cease participation at any time. Three classroom teachers, one physical education teacher, and two undergraduate students delivered the interventions. The recreational sports program (comparison group) and the Mastery Climate intervention (mastery climate cognitive-motor school-based group) were conducted in a local district sports court near city’s central area to facilitate children’s attendance. The academic reinforcement section was conducted in available classrooms at the schools.

The teachers kept a field diary about children's attendance, behaviors, engagement in the lesson, interest in the tasks, difficulties, cooperation, and children needing further reinforcement and feedback. The teachers also completed a self-reported checklist related to implementing the six TARGET structures; they completed the manipulation checklist along the intervention period. Frequencies from the manipulation checklist were scored regarding the agreement. Both procedures were adopted to ensure children's attendance, needs and to examine the intervention protocol adherence.

Statistics analysis

Mean (M), standard deviation (SD), the ratio of variation (delta Δ), and percentual of variation (delta Δ%) were estimated. The intervention effect, groups (intervention. and comparison) by time (pre- to post-tests), considering the possible effects of sex and age, was examined using a Generalized Estimation Equations (GEE) analysis. Within the GEE, a robust estimator (covariance matrix estimator), the normal distribution with an identity binding function, the unstructured working correlation matrix, and the criterion of the lowest Quasi Likelihood Under Independence Model Criterion (QIC) were examined (Ziegler and Vens 2010). The GEE was chosen since it held efficient and unbiased estimates for longitudinal or repeated measures research designs with nonnormal response variables and since it maintains the data of a subject in the analysis even in cases of missing a measure in a specific point of time (Guimarães and Hirakata 2012). For the significant interactions in the GEE, Bonferroni post hoc tests were conducted. The significance level p < .050 was adopted. The effect sizes were reported using Cohen-d tests with recognized cut-off points (small = .20 to .39; moderate = .40 to .79; large ≥ .80). The checklists were examined with the percentage of agreement.

Results

Intervention attendance and adherence to protocol

Adequate attendance for the intervention sessions was observed for the children in the MCG and CG; ranges were similar across the groups from 70 to 80%. The majority of the families justify the children's absences in the lessons due to helping with home chores. The teachers’ field diaries suggested that they were aware of children's difficulties and the necessity to design strategies and activities to meet children's needs and promote the intensive practice. Regarding protocol adherence to the climate, the self-reported manipulation checklist suggested an adequate appropriateness of each TARGET structure (task: 86%, authority: 89%; recognition: 97%; group 95%; evaluation: 93%; time: 90%).

Groups descriptive statistics at pre and post-tests

Motor skills, BMI, academic, and self-perceptions performance’ categories at pre- and post-test are present in table 1. We found a 100% prevalence of motor delays (very poor and poor performance) and 80% of low and below-average academic achievement – decreases in the prevalence of motor and academic delays for the MCG from pre- to post-tests. We found a healthy BMI (above 90%), low prevalence of overweight and underweight (around 3% to 5%), and no cases of obesity at pre- and post-test. Small increases in overweight and decreased underweight prevalence, from pre- to post-test, were also observed (around 3–5%). Children showed moderate perceived competence, social acceptance, and self-worth.

Table 1. MCG and CG: Motor skills, BMI, academic, and self-perceptions performance’ categories at pre- and post-test.

Performance’ categories		MCG N(%)		CG N(%)
		Pre-test	Post-test	Pre-test	Post-test
Motor Skill	Very poor	112 (95.7)	53 (45.3)	89 (94.7)	83 (88.3)
	Poor	5 (4.3)	30 (25.6)	5 (5.3)	9 (9.6)
	Below average	0 (0)	23 (19.7)	0 (0)	2 (2.1)
	Average	0 (0)	11 (9.4)	0 (0)	0 (0)
BMI	Underweight	10 (8.5)	4 (3.4)	3 (3.2)	1 (1.1)
	Health Weight	107 (91.5)	109 (93.2)	86 (91.5)	86 (91.5)
	Overweight	0 (0)	4 (5.6)	5 (5.3)	7 (7.4)
Academic Reading Skills	Low	84 (71.8)	48 (41)	54 (57.4)	37 (39.4)
	Below average	12 (10.3)	16 (13.7)	5 (5.3)	2 (21)
	Average	11 (9.4)	26 (22.2)	24 (25.5)	17 (18.1)
	Above average	6 (5.1)	4 (3.4)	5 (5.3)	2 (21)
	High	4 (3.4)	23 (19.7)	6 (6.4)	36 (38.3)
Academic Writing Skills	Low	84 (71.8)	48 (41)	54 (57.4)	37 (39.4)
	Below average	12 (10.3)	16 (13.7)	5 (5.3)	2 (2.1)
	Average	11 (9.4)	26 (22.2)	24 (25.5)	17 (18.1)
	Above average	6 (5.1)	4 (3.4)	5 (5.3)	2 (2.1)
	High	4 (3.4)	23 (19.7)	6 (6.4)	36 (38.3)
Academic Math Skills	Low	90 (76.9)	74 (63.2)	83 (88.3)	83 (88.3)
	Below average	15 (12.8)	15 (12.8)	3 (3.2)	3 (3.2)
	Average	6 (5.1)	13 (11.1)	6 (6.4)	6 (6.4)
	Above average	5 (4.2)	3 (2.6)	0 (0)	0 (0)
	High	1 (0.9)	12 (10.3)	2 (2.1)	2 (2.1)
Perceived Scholar Competence	Low	14 (12)	18 (15.4)	16 (17)	16 (17)
	Moderate	87 (74.4)	75 (64.1)	61 (64.9)	65 (69.1)
	High	16 (13.7)	24 (20.5)	17 (18.1)	13 (13.8)
Perceived Athletic Competence	Low	12 (10.3)	6 (5.1)	16 (17)	13 (13.8)
	Moderate	93 (79.5)	94 (79.5)	66 (70.2)	64 (68.1)
	High	12 (10.3)	17 (14.5)	12 (12.8)	17 (18.1)
Perceived Behavioral Conduct Acceptance	Low	18 (15.4)	7 (6)	4 (4.3)	4 (4.3)
	Moderate	86 (73.5)	97 (82.9)	81 (86.2)	84(89.4)
	High	13 (11.1)	13 (11.1)	9 (9.6)	6 (6.3)
Perceived Social Acceptance	Low	23 (19.7)	18 (15.4)	19 (20.2)	21 (22.3)
	Moderate	69 (59)	76 (65)	63 (67)	58 (61.7)
	High	25 (21.4)	23 (19.7)	12 (12.8)	15 (16)
Global self-worth	Low	21 (17.9)	15(12.8)	9 (9.6)	9 (9.6)
	Moderate	92 (78.6)	79(67.5)	61 (64.9)	63 (67)
	High	4 (3.4)	23(19.7)	24 (25.5)	22 (23.4)

Note: MCG: Mastery Climate Group; CG: Comparison Group; BMI: Body Mass Index.

Changes over time: The intervention effect

Motor performance

Table 2 presents the means and standard deviations by groups and post-hoc tests results (within and between groups) for motor performance, BMI, and academic performance at pre- and post-tests. The GEE analysis showed a significant group-by-time interactions for LOC (χ² Wald = 100.00, df = 1, p < .001) and OC (χ² Wald = 106.90, df = 1, p < .031) skills. The Bonferroni post hoc test showed significant increases in LOC and OC scores from pre- to post-test for the MCG with a large effect size. For the CG, significant increases were found only in LOC skills, moderate effect size. For the LOC and OC skills, groups were similar at the pre-test. At the post-test, the MCG scores were higher with large effect sizes compared to CG. The ratio of change was significantly higher for the MCG for LOC and OC skills with large effect sizes.

Table 2. MCG and CG: Groups M and SD and post-hoc results (within and between groups comparisons) for motor and academic performances, and BMI at pre and post-test.

Motor and Academic Performances & BMI	MCG and CG M(SD) & Statistics Result				Between-Groups Comparisons
	MCG		CG		p	Cohen’s d
Locomotor Skills
Pre	23.6(4.4)		24.6(4.5)		.087	.23
Post	32.5(5.1)		26.2(4.9)		< .001	1.26
Within Groups p, d	p < .001	d = 1.74	p < .001	d = .33	–	–
Δ absolute, Δ%	Δ = 8.8	Δ% = 27.4	Δ = 1.6	Δ% = 6.5	< .001	1.62
Object Control Skill
Pre	23.9(5.3)		27.1(5.7)		.551	.59
Post	32.9(5.1)		27.2(5.1)		< .001	1.28
Within Groups p, d	p < .001	d = 1.81	p = .952	d = .05	–	–
Δ absolute, Δ%	Δ = 9	Δ% = 37.7	Δ  = .30	Δ% = .40	< .001	1.58
BMI
Pre	15.6(1.3)		16.2(1.2)		< .001	.48
Post	16.3(1.3)		17.0(1.3)		< .001	.54
Within Groups p, d	p < .001	d = .06	p < .001	d = .31	–	–
Δ absolute, Δ%	Δ = .70	Δ% = 9	Δ = .80	Δ% = 4.9	.005	.40
Reading Skills
Pre	35.9(10)		50.6(12)		<.001	1.35
Post	45.8(9.1)		56.9(11.1)		.004	1.11
Within Groups p	p < .001	d = .40	p < .001	d = .30	–	–
Δ absolute, Δ%	Δ = 10.1	Δ% = 27.6	Δ = 6.1	Δ% = 12.5	.018	.33
Writing Skills
Pre	11.8(6.4)		16.5(8.5)		< .001	.64
Post	14.6(5.8)		19.0 (8.8)		.020	.61
Within Groups p, d	p < .001	d = .21	p < .001	d = .27	–	–
Δ absolute, Δ%	Δ = 2.8	Δ% = 23.7	Δ = 2.6	Δ% = 15.2	.874	.02
Math Skills
Pre	6.5(1.1)		9.0(1.2)		< .001	2.19
Post	8.7(1.5)		9.0(1.3)		.231	.21
Within Groups p, d	p < .001	d = .50	p = .437	d = .00	–	–
Δ absolute, Δ%	Δ = 2.1	Δ% = 33.8	Δ = -.10	Δ% = .00	< .001	.89

Note. MCG: Mastery Climate Group; CG: Comparison Group; BMI: Body Mass Index; M: Mean; SD: Standard Deviation; Delta: Δ; Cohen’s d: small: .20 to .39; moderate: .40 to .79; large: ≥.80.

Body Mass Index

The GEE analysis showed a significant group-by-time interaction for BMI scores (χ² Wald = 7.15, df = 1, p = .008). The Bonferroni post hoc test showed significant increases in BMI scores from pre- to post-test for the MCG and CG, with small effect sizes. Groups were significantly different at pre-and post-tests with moderate effect size; CG showed higher scores in both tests. The ratio of change was also significant higher for the CG with moderate effect size (see table 2).

Academic Performance

Yet in table 2, regarding reading performance, the GEE analysis showed a significant group-by-time interaction for reading scores (χ² Wald = 7.31, df = 1, p = .007). The Bonferroni post hoc test showed significant increases in reading scores from pre- to post-test for the MCG and CG, with moderate and small effect sizes, respectively. Groups were significantly different at pre-and post-tests with large effect sizes; the MCG showed lower scores than the CG. However, the ratio of change was also significantly higher for the MCG with a small effect size.

Regarding writing performance, the GEE analysis showed main effects for group (χ² Wald = 6.15, df = 1.0, p = .013) and time (χ² Wald = 54.32, df = 1, p < .001). The Bonferroni post hoc test showed significant increases in writing scores from pre- to post-test for both groups. Groups were significantly different at the pre- and post-test, with moderate effect sizes; the MCG showed lower scores than the CG. The ratio of variation was non-significant. Therefore, the gains were similar for both groups.

Regarding math performance, the GEE analysis showed a significant group by time interaction for math scores (χ² Wald = 13.65, df = 1.0, p < .007). The Bonferroni post hoc test showed significant increases in math scores from pre- to post-test for the MCG with moderate effect size; no significant increases were found for CG. Groups were significantly different at pre-test with a large effect size; the groups were similar at the post-test. However, the ratio of variation was significantly higher for the MCG. In addition, a large effect size was observed.

Self-Perceptions

Table 3 presents the means and standard deviations by groups and post-hoc tests results (within and between groups) for the self-perception of competence, social acceptance, and global self-worth at pre- and post-test. Regarding perceptions of academic competence, the analysis showed a significant group-by-time interaction for scholastics scores (χ² Wald = 26.50, df = 1, p < .001). The Bonferroni post hoc test showed significant increases in perceived academic competence scores from pre- to post-test for the MCG with moderate effect size; no changes were found for CG. Groups were similar at pre-test and significantly different at the post-test with small effect size. The MCG showed higher scores compared to the MCG at the post-test. The ratio of variation was significantly higher for the MCG with moderate effect size.

Table 3. MCG and CG: Groups M and SD and post-hoc results (within and between groups comparison) for self-perceptions at pre and post-test.

Self-Perceptions	MCG and CG M(SD) & Statistics Results				Between-Groups Comparisons
	MMCG		CG		P	Cohen’s d
Scholastics Competence
Pre	15.6(3.7)		16.5(3.6)		.06	.25
Post	17.2(3.4)		16.1(3.7)		.025	.31
Within Groups p, d	p < .001	d = .43	p = .197	d = .07	–	–
Δ absolute, Δ%	Δ = 1.6	Δ% = 10.3	Δ = -.17	Δ% = −2.4	< .001	.63
Athletics Competence
Pre	14.9(2.8)		14.1(2.9)		.06	.28
Post	17.9(2.3)		14.5(3.1)		.001	1.27
Within Groups p, d	p < .001	d = .94	p = .010	d = .19	–	–
Δ absolute, Δ%	Δ = 3.2	Δ% = 20.1	Δ = .54	Δ% = 2.8	< .001	1.26
Social Acceptance
Pre	15.2(3.1)		16.3(3.7)		.027	.33
Post	16.5(3.3)		16.3(3.9)		.687	.69
Within Groups p, d	p < .001	d = .45	p = .143	d = .04	–	–
Δ absolute, Δ%	Δ = 1.1	Δ% = 8.6	Δ = .20	Δ% = .00	< .001	.44
Behavior Conduct Acceptance
7–8 years-old
Pre	15.8(4.1)		15.9(4.2)		.998	.02
Post	15.9(3.6)		15.7(5.1)		.613	.05
Within Groups p, d	p = .567	d = .03	p = .513	d = .11	–	–
Δ absolute, Δ%	Δ  = .10	Δ% = .60	Δ = -.10	Δ% = −1.3	.463	.13
9–10 years-old
Pre	16.(3.8)		15.7(4.8)		.333	.07
Post	17.3(3.9)		15.6(4.6)		.001	.40
Within Groups p, d	p < .001	d = .61	p = .758	d = .05	–	–
Δ absolute, Δ%	Δ = 1.4	Δ% = 8.1	Δ = -.10	Δ% = -.60	< .001	1.02
Global Self-worth
Pre	16.7(3.1)		18.2(3.2)		<.001	.48
Post	18.2(3.5)		17.9(3.9)		.362	.08
Within Groups p, d	p < .001	d = .58	p = .362	d = .24	–	–
Δ absolute, Δ%	Δ = 1.5	Δ% = 9.0	Δ = -.30	Δ% = −1.6	< .001	.75

Note. MCG: Mastery Climate Group; CG: Comparison Group. M: Mean; SD: Standard Deviation; Delta: Δ; Cohen’s d: small: .20 to .39; moderate: .40 to .79; large: ≥.80.

Concerning athletic competence, the analysis showed a significant group-by-time interaction for athletic scores (χ² Wald = 86.40, df = 1, p < .001). The Bonferroni post hoc test showed significant increases from pre to post-test in perceived athletic competence scores for the MCG with strong effect size and for the CG with a small effect size. Groups were similar at pre-and post-tests with large effect size. The MCG showed higher scores compared to the CG at the post-test. The ratio of variation was also larger for the MCG and showed a large effect size.

Regarding social acceptance, the GEE analysis showed a significant group-by-time interaction in social scores (χ² Wald = 21.45, df = 1.0, p < .001). The Bonferroni post hoc test showed significant increases in social acceptance scores from pre- to post-test for the MCG with moderate effect size. The ratio of variation was also significantly higher for the MCG. Also, a moderate effect size was observed.

Concerning behavioral conduct, the GEE analysis showed a significant group-by-time-by-age interaction for behavioral conduct (χ² Wald = 6.60, df = 1, p = .010). The Bonferroni post hoc test showed significant increases in the behavioral conduct only for 9-10-years-old children in the MCG with moderate effect size. The ratio of variation was significantly larger for the MCG with a large effect size.

In relation to self-worth, the GEE analysis showed a significant group-by-time interaction for global self-worth scores (χ² Wald = 33.48, df = 1, p = .007). The Bonferroni post hoc test showed significant increases in self-worth from pre- to post-test for the MCG and CG, with moderate and small effect sizes, respectively. The ratio of variation was significantly higher for the MCG. Also, a large effect size was observed.

Discussion

We conducted a mastery climate cognitive-motor intervention for children living in poverty. The intervention was successful in helping children in the intervention group improving motor, academic (reading, math), perceived competence, social acceptance, and self-worth scores from pre- to post-test. Furthermore, the intervention group showed higher scores than the comparison group at the post-test.

Motor performance

The mastery climate intervention effectively changed LOC and OC skills performance of the MCG, regardless of sex and age, supporting our hypothesis. The results were like previous mastery climate interventions for vulnerable children (Píffero and Valentini 2010; Sampaio and Valentini 2015; Valentini 2002; Valentini et al. 2017); the ratios of changes were higher for the MCG than the CG.

For the CG, improvements were found for LOC skills, but not for OC skills, like the previous studies that also implemented a direct instruction approach (Martin, Rudisill, and Hastie 2009; Píffero and Valentini 2010; Valentini and Rudisill 2004b). Although the frequency of practice was similar for both groups in the present study, the participation in recreational sports was not strong enough to generate effective changes for ball skills. The lack of instruction and specific strategies for children with motor deficits and the recurrent practice of games already well known by children are plausible explanations for these results.

Body Mass Index

We expected no significant BMI changes. However, increases from pre- to post-test were observed for both groups. The changes in BMI were not related to increases in overweight and obesity; in fact, four children in the MCG and one in the CG moved from underweight categorization to healthy weight; the other children maintained their healthy weight, and few were overweight. Although there is evidence for reducing BMI in children enrolled in school-based program (Bogart et al. 2017), this program, contrary to ours, enrolled children with obesity. No intervention studies with underweight children reported gains in BMI to the authors’ knowledge, limiting our capacity to compare the results.

Two plausible explanations for these results may be outlining. First, all children came from families living in poverty; with a family that received governmental support, we speculate that considering their socioeconomic conditions, a restricted food intake and less access to healthy food was the reality. As the intervention started and we provided healthy snacks, children may have more intake than usual. Second, the study started after summer vacation, and schools also provided daily meals (i.e. rice, beans, vegetables, meat, pasta, soups) for all children to complement the food intake at home; children had more food available to them at school also. These factors lead to increases in BMI for underweight children. Our suppositions were drawn from the knowledge about the communities and schools and information about the families’ socio-economic status.

Academic Performance

The results for academic performance showed the intervention's effectiveness on all variables (reading, writing, math), regardless of sex and age, with moderate and small effect sizes, confirming our hypothesis. Positive changes were observed for the CG for reading and writing skills, with small effect sizes. Up to date, no study has reported the impact of mastery climate cognitive-motor skills on school performance for children, limiting our ability to compare results. However, to some extent, our results are comparable with previous classroom studies implementing mastery climate. positive changes in effort-related cognition, self-instruction, self-monitoring strategies (Ames 1992; Ames and Archer 1988), metacognitive strategies, and proactive behaviors (Wolters 2004) have been reported more often for children in mastery climate than traditional climates.

Self-perception of competence

Unlike previous mastery climate interventions, we assessed several dimensions of self-perceptions and global self-worth to understand its repercussions more deeply. We found increases in self-perceptions of school and athletic competence, social acceptance, and self-worth for the MCG, regardless of sex and age. Regarding behavioral conduct, positive changes were observed only for older children (9 and 10 years-old) in the MCG. For the CG, we only found increases in self-perceptions for the perceived athletic competence.

Like our results, for both groups, previous studies also reported increases in children's perceptions of physical competence (Brauner, Valentini, and Santayana de Souza 2017; Robinson, Rudisill, and Goodway 2009; Theeboom, De Knop, and Weiss 1995; Valentini and Rudisill 2004b). Regarding global self-worth, the MCG results are like those reported for children living in vulnerability attending a mastery climate sports program (Brauner, Valentini, and Santayana de Souza 2017). MCG participated in a climate that supported their individual needs and provided opportunities to learn new skills, strengthened their perception of motor competence (Theeboom, De Knop, and Weiss 1995).

Regarding perceptions of academic competence, social acceptance, and behavioral conduct – the present study's original contribution, positive changes were observed for the MCG. These positive changes are excited; the intervention helped children with motor and academic delays improve performance and consequently express more positive feelings about their competencies. The strategies adopted in the mastery climate effectively reinforced children’s autonomy, problem-solving, and monitoring progress; helping children nurture their self-perceptions.

Regarding social acceptance, participation in the intervention required children to cooperate to establish and follow a protocol for rules and acceptable behavior toward teachers and peers, enrolling children in a co-responsibility role. Furthermore, the strategies for group dimension (i.e. respect differences, help a peer with difficulty, use the circle time to express feelings and accomplishments, solve the interpersonal conflict by talking or ask the teacher for mediation) were strategies implemented to improve peer interactions, resulting in perceptions of conduct and social acceptance improvements. Social interactions were also strengthened by the setting's physical organization (i.e. stations) and team games. These strategies reinforce children's social acceptance by acquiring social parameters and internalizing the skills necessary to make friends and get along with them (Harter 2012a).

Positive effects were observed for older children in MCG (9–10 years-old) regarding behavioral conduct. It is essential to acknowledge that children, at this age, compare their behavior with peers; this confrontation from what is or not expected and accepted by others helps them to regulate their behavior (Harter 1988). As children age, peers and teachers’ opinions and values are strongly internalized and play a relevant role in helping children acquire self-judgment parameters (Harter 2012a). Younger children have more difficulty internalizing protocols and rules and are aware of behavior comparisons, a plausible explanation for the lack of changes in perceptions of behavior conduct.

Last, changes in global self-worth were also observed for MCG. Global self-worth refers to how much the child is happy and satisfy with who she/he is (Harter 2012b). Children living in social-economic vulnerability tend to introject negative personal attributes, perceive themselves as inferior and unskilled, and often lack social recognition that could help them to elaborate firmer beliefs about their potential (Pereira 2010). The intervention’s participation helped children recognize their ability to perform motor and school tasks, interact in different groups, and making new friends, strengthening their sense of belonging and global self-worth.

Strength and limitations

One of the strengths of this study was to provide, for the first time, evidence for the effectiveness of mastery climate in supporting reading, writing, and math improvements in children living in poverty; also, for the first time, evidence is provided for mastery climate for fostering social acceptance and self-worth in children. Another strength was enrolling a large sample from three different contexts representing different cultures within a state; 32 schools were also enrolled in the present study – our sample was not diverse. Another strength is that we also enrolled a matched comparison group of children attending youth recreational sports programs and academic reinforcement sponsored by the state – a complex design approach than a comparison group that received no intervention. Lastly, we used the GEE analysis that provided unbiased estimates for repeated measures research designs with nonnormal response variables and maintained all subjects’ data in the analysis even in cases of missing a measure at a specific point of time.

The limitation of the present study was not to examine the long-term impact of the mastery climate cognitive-motor interventions. The increases in self-perceptions of academic, social acceptance, and global self-worth may long-term affect motor and academic achievement; an investigation of those factors would provide relevant information to research and practice. Therefore, future research may address follow-up intervention effects. Another limitation was not controlling families’ eating habits and children's daily intake, since children with underweight gained weight along with the intervention and we provided plenty of healthy snacks for all children. It would also be interesting to understand further how important it was in complementing the daily food intake of children living in poverty, another recommendation to future research.

Besides, the positive changes in social acceptance and self-worth found in the present study led us to question whether these specific changes could impact child relationships in other development contexts. Therefore, our recommendation for future studies is to investigate the impact of mastery climate interventions on the relationships between the child and their peers and the child and their families – a study addressing this issue would provide interesting practical repercussions. Similarly, the positive changes in perceptions of motor and academic competence in the present study also led us to question how the teachers’ instruction could be related to such a relevant change. Therefore, another recommendation for future studies is to investigate how teachers’ instructions and feedback influence children’s motor and academic self-perceptions and the sources of information children use to judge their performance (e.g. adults, past experiences, peers) – both issues still lack evidence in the intervention literature.

Conclusion

The mastery climate promoted equity opportunities for boys and girls to learn, no sex interaction was found. These results have a direct implication to practice – the mastery climate showed the potential to help girls who often demonstrated lower motor scores, to improve their skills like the boys. Implementing climates that provide equal opportunities is crucial to promoting motor learning, and it was found in the present.

The most vulnerable populations, specifically those living in poverty, are chronically exposed to risk factors and need interventions to counterbalance those risks. Our results emphasized the relevant role of compensatory programs in helping children learn new skills and to nourish positive self-perception of competencies and social acceptance within impoverished settings with minimal resources.

Children living in vulnerability needed robust support; incorporating effective motivational strategies in compensatory programs allows for critical academic, motor, and social achievements for those children engage in further tasks. Providing realistic and positive opportunities for foster motor skills, self-concept, and socialization and inhibit unwanted behaviors of self-exclusion and isolation are essential strategies for promoting children's integral development.

Disclosure statement

No potential conflict of interest was reported by the author(s).