to select a learning method to solve a motor problem: a badminton service. .... in which the teacher explains to the learner what to do; tutoring, in which the ... tion, for example, has been replaced in France by reflection on content as a result ... of relevant pieces of information) and memory (memorization of observed ...
Vincent-Morin and Lafont
JOURNAL OF TEACHING IN PHYSICAL EDUCATION, 2005, 24, 226-242 © 2005 HUMAN KINETICS, INC.
Learning-Method Choices and Personal Characteristics in Solving a Physical Education Problem Madeleine Vincent-Morin and Lucile Lafont Université Victor Segalen Bordeaux 2 The goal of this study was to identify the relationships between the learning choices made by pupils and their personal characteristics, including cognitive style (field dependence–independence), a motivational variable (feeling of selfefficacy), and a cognitive variable (task representation). The participants were 64 twelve-year-old sixth graders from a suburban middle school in France (35 boys and 29 girls). Cognitive style was measured with the Group Embedded Figures Test, a perceptual test that requires finding a simple geometrical figure embedded in a complex geometrical one. Five learning conditions (autonomy, tutoring, verbal instruction, silent demonstration, and verbal demonstration) were then proposed in random order to the pupils. They were asked to select a learning method to solve a motor problem: a badminton service. The results indicated an absence of relationships between the choice of a learning condition and cognitive style. Three variables partially predicted the learning-condition choice: feeling of self-efficacy, task representation, and motor performance. The present results can be interpreted in the light of studies on children’s help-seeking behavior in problem-solving situations. Key Words: guidance, choice, cognitive style, motor problem solving
This study examines the relationship between learning and teaching in the specific area of complex motor skills. The goal was to identify the relationships between the choices made by pupils and some of their personal characteristics, including cognitive style (field dependence–independence (or FDI, as measured by the Group Embedded Figures Test [GEFT], a perceptual test that requires finding a simple geometrical figure embedded in a complex geometrical one), a motivational variable (feeling of self-efficacy), and a cognitive variable (task representation). Pupils were asked to select a learning method to solve a motor problem: a badminton service. This experiment differs from others in that it grants pupils a role in determining their learning experience by letting them select a learning condition. Allowing pupils to choose an instructional method should promote learning. This choice can be defined as a rational, contextualized decision aimed at achieving a specific goal. According to Ricoeur (1999), choices bring the individual’s free will to bear, plans are related to goal-oriented motives, and Vincent-Morin and Lafont are with Université Victor Segalen Bordeaux 2, Laboratoire VSTII, EA 498, Faculté des Sciences du Sport et de l’EP, Ave Camille Jullian, 33607 Pessac Cedex France. 226
decisions are the consequence of goal-oriented calculations. In the present study, pupils had to make a certain number of probability calculations regarding their success or failure on the task to be carried out. Their choice was based on their own experience, abilities, representation of the task, self-efficacy beliefs (Bandura, 1977), and learning expectations. From a macro-analytic perspective, research in physical education teaching is full of controversies and debates. According to Rink (2001), there is a “resurgence of the methods wars” in which each side tries to show that one method of teaching is better than others. Classically, we can distinguish direct styles of teaching (demonstration or verbal instruction with feedback) and indirect styles of teaching (exploration and discovery teaching). Indirect teaching is associated with more student-centered methodologies (Rink, 2001). Today’s physical education teaching systems advocate socioconstructivism and can be seen as socially based applications of learning theory. The first, the sport education model (Siedentop, 1994, 2002), is a curriculum system designed to provide rich sport experiences in a physical education context: pupils belong to the same team during the entire season and teachers create a festive atmosphere throughout the season. Affiliation to the team allows students to plan, play, compete, and therefore benefit from social development in their group. Another method is Teaching Games for Understanding (TGFU; Thorpe, Bunker, & Almond, 1986), a content-based approach that links tactics and skills in a game context. In this model, students play a modified game that highlights a particular tactical problem. A more recent development in physical education teaching is Cooperative Learning (CL; Dyson, 2001, 2002). This technique helps students take responsibility through roles, enhances student communication, and improves the students’ ability to work together within a group. These three perspectives emphasize several key points: students are active learners, students work in small groups, the teacher sets goals or defines problems, and students seek solutions to the problems. In France, didactics research emphasizes the three-way relationship linking teacher, students, and knowledge taught. Two major concepts are advocated in physical education: “didactic contract” (the result of implicit negotiations among pupils, a teacher, and an instructional environment) and “devolution” of a problem to elicit adaptive learning. These concepts are derived from French research in didactics conducted first in mathematics (Brousseau, 1986, 1997). From a microanalytic point of view, Beaudichon, Verba, and Winnykamen (1988) defined a multidimensional approach to learning. These authors highlighted the multiplicity of the mechanisms and processes that come into play in knowledge acquisition and proposed the idea that not all learning conditions are equally effective. In this perspective, Winnykamen and Lafont (1990) and Lafont (1994) demonstrated the complementary nature of instructional methods in motor-skill learning, which depends on the type of task, the context, and the characteristics of the learner. Proposing different learning conditions requires an understanding of the idea that teaching effectiveness is both subject dependent (MacCullagh, Weiss, & Ross, 1989) and task dependent (Newell & Walter, 1981). The present study looks at the role of certain learner-related variables in the learner’s choice among several forms of help, ranging from working alone to substantial instructional assistance such as a demonstration. The following five learning conditions were studied here: verbal demonstration, in which the teacher is a model and demonstrates and explains the skill to be learned (Bandura, 1986,
Vincent-Morin and Lafont
Winnykamen & Lafont, 1990); silent demonstration, in which the teacher demonstrates the skill but does not make comments (Bandura, 1986); verbal instruction, in which the teacher explains to the learner what to do; tutoring, in which the teacher gives help and support only when the learner needs it (Bruner, 1983); and autonomy, in which the learner works alone in a problem-solving situation. These same conditions were used by Winnykamen and Servant (1986), who analyzed 11- and 12-year-old children’s choices for solving the Tower of Hanoi problem. Their results showed that the type of aid chosen most frequently was tutoring. Tutoring was followed by verbal instruction and then verbal demonstration. In a pilot study conducted with 11-year-old participants, Lafont and VincentMorin (2000) analyzed the choice of an instructional method for learning a complex motor skill (the cartwheel) and the stability of that choice after task performance. The experimental procedure and learning-condition choices were identical to those used by Winnykamen and Servant. The results were partially different from those of the previous study, however, and showed that participants chose the autonomy condition most often, followed by verbal demonstration. Looking solely at the participants’ first two choices, the three methods mentioned the most often (11 times each) were autonomy, verbal demonstration, and verbal instruction. The relationship between the chosen instructional method and initial performance indicated that participants who selected autonomy or verbal instruction exhibited higher initial performance than participants who opted for demonstrative modes of learning. Taking a multiprocess approach, our study attempts to show that the suitability of a learning method depends on certain characteristics of the learners. We also wanted to show that instructional teaching methods could still be useful, even though they have been discredited for physical education in France. Demonstration, for example, has been replaced in France by reflection on content as a result of an innovative trend promoting pedagogical approaches based on problem solving (Amade-Escot, 2000) or environmental design (where the teacher sets up a material device that makes the task goal evident and he or she then lets the pupil find the solution alone; Famose, 1990). This approach is an application of the goal-setting theory. Amade-Escot (1991) and Brau-Antony (2003) investigated the behavior of teachers managing didactic situations and observed four teaching modes: (a) formal teaching of gestural models; (b) teaching by controlling the surroundings (associative learning); (c) teaching by devolution during play activity (incidental learning); and (d) teaching by devolution of a problem (adaptive learning). The procedures used by the majority of teachers employed transmission and could be placed in the first two categories. Nevertheless, teaching modes based on problem solving are highly thought of because of the devolution to pupils and indirect influence, which advocates technicisme (gestural techniques without tactics or decisional choices) for the first category (Amade-Escot, 1991, Brau-Antony). In a meta-analysis of articles published in Spirales (French journal of physical education didactics), Lafont (2002) showed that problem-solving situations were preferred to demonstration–imitation modes of teaching in gymnastics. Demonstration alone and demonstration accompanied by verbal explanation are two types of tutoring methods derived from Bandura’s (1986) sociocognitive learning theory. According to this author, knowledge and skills can be acquired through observation of others. This form of acquisition relies on attention (selection of relevant pieces of information) and memory (memorization of observed
behavior), motor reproduction, and motivational processes. Bandura’s model was validated by Caroll and Bandura (1982) in the field of complex motor skills and by Lafont (1994) in morphokinetics (technical). Bruner (1983) defined the process of tutoring as a “scaffolding” process, a form of assistance provided by a tutor to a novice. The tutor performs several functions for the novice: reducing the degrees of freedom of the task, orienting attention, controlling frustration, signaling important characteristics, and demonstrating. In the present study, the type of tutoring proposed to the participant depended on the particular difficulties he or she encountered in solving the problem. According to the taxonomies proposed by Poulton (1957) and Paillard (1974), a badminton service exhibits a mixture of morphokinetic (technical, gestural) and teleokinetic (spatial goal-oriented) properties and requires closed skills. Diverse forms of instruction can therefore be proposed to learners: a problem-solving approach, as well as demonstration, might be appropriate for teaching this type of task. The experiment presented here examines the relationships between participants’ choices among five forms of instruction and certain psychological variables (cognitive style, self-efficacy beliefs, and task representation before execution). The first point studied was the relationship between participants’ choices and cognitive style, or FDI. This concept is based on work by Witkin (1967) and Huteau (1987). Huteau postulated that cognitive style is indicative of individual differences in personality that affect the subject’s manner of perceiving, memorizing, and solving problems. Cognitive style shows up in the cognitive and socioaffective domains and corresponds to the participant’s way of functioning in these domains. According to Witkin (1965), FDI can be used to differentiate individuals based on their ability to perceive an element out of context and to adopt an analytical attitude in problem solving. In the field-dependent perception mode, perception is dominated by the general organization of the field as a whole, whereas in the field-independent mode, the parts of the field are perceived separately, irrespective of how the background is organized. Huteau, whose perspective is noticeably different, defines the concept of differentiation at the “me/not me” level, arguing that field-independent (FI) participants possess strong analytical abilities, along with poor social skills or a high level of autonomy in interpersonal relations. Conversely, field-dependent (FD) participants have lower analytical abilities and good social skills. Several studies allow us to predict which method participants will choose based on their cognitive style. According to Huteau, FI individuals are better than FD individuals at breaking down and reconstructing data. FD individuals conceptualize less easily, whereas FI participants are able to function at either the analytical or the global level because their abilities are more differentiated. A series of arguments in other areas point in the same direction: FD participants have a greater need for a secure environment; FI participants have a more confident posture and better psychomotor balance (Huteau). At the social-behavior level, FD participants are more dependent. According to Konstadt and Forman (1965), compared with FI participants, FD participants are more sensitive to negative reinforcement and look more often at the experimenter. Thus, it was hypothesized here that FD individuals would choose a demonstrative instructional method with substantial assistance, such as silent demonstration or verbal demonstration, more often than would FI participants. Inversely, FI individuals should choose independent problem-solving modes or tutoring.
Vincent-Morin and Lafont
The second aim of the present experiment was to determine the relationships between certain motivational variables and the choices made by the participants. It was assumed that the choice of an instructional method would be tied to the perceived likelihood of success on the task. According to Bandura (1977), the feeling of self-efficacy corresponds to the judgment individuals make about their ability to achieve a certain performance level (self-efficacy level) on the task with a given degree of certainty (self-efficacy strength). In other words, the participant weighs the costs and benefits of executing the task. This measure is justified by the fact that studies on help seeking have found a relationship between the amount of help sought (in our study, each learning condition involves a different degree of assistance) and self-efficacy beliefs. We hypothesized that participants who had a low self-efficacy level would choose a strong guidance procedure such as silent or verbal demonstration. The final topic explored here was the participants’ representation of the task before execution. For our purposes, the notion of representation is defined as participants’ mental constructs that incorporate the goal to be attained, motor operations needed to attain it, and constraints to be taken into account. Such representations in physical education are the result of norms, social values, and common practices. In summary, the present study aimed to examine the relationships between cognitive style (FDI), self-efficacy, task representation, and the choice of a help method. One of the questions raised was: Which of the three characteristics under study would best explain the choice of an instructional method? In addition, from an exploratory point of view, the performance of boys and girls on behavioral and cognitive variables was compared and the effectiveness of the different learning conditions chosen by the participants was examined. Finally, correlations among relevant variables were computed. Specifically, it was expected that self-efficacy beliefs would be positively related to motor performance.
Method Participants The participants were 64 twelve-year-old sixth graders from a suburban middle school in a well-off neighborhood of a university town in France. The group was comprised of 35 boys and 29 girls from middle- to upper-class backgrounds. They had 4 hours of physical education classes per week. The pupils who volunteered to participate gave written consent. They were told that they would be filmed, but that the study was entirely confidential. They were novices in the skill to be learned: a badminton service.
Task, Materials, and Design The task was a badminton service. This task was chosen because it is easy to model and explain, and because it can be learned autonomously in a problemsolving situation. It has mixed properties, being both morphokinetic and teleokinetic. The participants carried out the task on a badminton court in the gymnasium of their school. All participants were videotaped during the third, fourth, and fifth stages of the experiment (see below). During these stages, the experimenter (a physical education teacher) explained the task goal and proposed the five learning
conditions in random order. Each participant chose the preferred learning condition and then carried out the task. The experimenter gave the instructions to the participants while being assisted by two physical education students from the university (tutors), one male and one female, who helped in the demonstration, verbalinstruction, and tutoring conditions. The students had been trained to teach the skill in the four guidance conditions before the experiment.
Dependent Measures The measures included the Group Embedded Figures Test (GEFT) score (Witkin, Oltmann, Raskin, & Karp 1971), badminton-service performance, selfefficacy beliefs, and task representation. GEFT Score. The GEFT is a form of the embedded figures test (EFT) for groups. Developed by Oltman, Raskin, Witkin, and Karp (1971), the GEFT is a perception test frequently used for measuring FDI. It is similar in its presentation and procedure to the EFT and consists of 18 complex figures (17 of which are also part of the EFT); participants must find and identify simple geometrical figures embedded in the complex ones. It has three parts: the first part is a training phase, which is not scored; the second and third parts consist of nine complex figures each. The task is to locate and trace with a pencil as many simple forms as possible during two 5-minute test periods. The GEFT score is equal to the total number of simple forms accurately outlined in the complex figures. Participants who scored 9 or above (on a scale of 0 to 18) were labeled field independent; those scoring 5 or below were labeled field dependent. Participants scoring between 6 and 8 were classified as intermediate. Two experimenters independently rated performance on the test. Badminton-Service Performance. Two physical education teachers who were badminton experts and who were unaware of the purpose of the study rated performance on each trial independently. Five success levels were defined on a scale from 1 to 5. A score of 1 was given when the shuttlecock was not hit or when it was hit but stayed near the subject. A score of 2 indicated that the shuttlecock was hit but did not go over the net or was hit with an incorrect movement of the racket. A score of 3 indicated that the shuttlecock went over the net, but it did not fall in the prescribed area and did not go in the right direction. A score of 4 meant that the shuttlecock went over the net along the diagonal, but the stroke was too short. A score of 5 indicated that the badminton service was correctly performed. This scale was validated by two expert badminton teachers. Each participant carried out eight trials in the chosen learning condition. The first trial was called the pretest and the last, the posttest. Self-Efficacy. The feeling of self-efficacy was assessed via questions related to the skill level each participant felt he or she possessed. The participant had to choose one of four levels, which were worded as follows: “When I serve, the shuttlecock doesn’t go over the net” (Level 1). “When I serve, the shuttlecock goes over the net but out of the prescribed area” (Level 2). “I serve correctly, but it’s easy for my opponent” (Level 3). “I serve correctly, and I make it hard for my opponent” (Level 4). The last question pertained to the strength of the subject’s perceived self-efficacy. It was assessed on a scale ranging from 0 to 100%. Each number (0, 10, 20, 30 . . . 100) represented a self-efficacy strength. The participants checked the number corresponding to the level they thought they had reached.
Vincent-Morin and Lafont
Task Representation. The experimenter asked each participant to fill out a questionnaire designed to assess task representation (i.e., the goal and the action required to reach that goal). Task representation was assessed using four multiplechoice questions about the rules for serving in badminton. The questionnaire was validated by two physical education teachers who were badminton experts. Questions 1 and 3 referred to the service area and were illustrated by six diagrams each. Questions 2 and 4 referred to actions. The four questions were: 1. Each diagram represents a badminton court and the position of the badminton server in a particular area. Choose one or more answers. 2. This question is about the movement of the racket by the badminton server. Choose one and only one answer among the three possibilities. Only one answer is right; the others are wrong. 3. This question is about the position of the server’s feet when he serves. Choose one and only one answer among the three possibilities. Only one answer is right; the others are wrong. 4. Each diagram represents a badminton court with the correct position of the badminton server and six areas for the opponent. Choose one and only one answer.
Procedure The experiment consisted of the following six stages. Stage 1. The participants were tested collectively in a classroom of their school on the French version of the GEFT (Witkin, Oltmann, Raskin, & Karp, 1971). Stage 2. Each participant answered the self-efficacy questionnaire and a task-representation questionnaire. Stage 3. The procedure was the same as in the pilot study mentioned above. The tutor explained the task goal: learn the badminton service. Then the experimenter proposed the five learning conditions in random order. A comic strip depicting each condition was shown on separate sheets of paper; in four of the conditions (silent demonstration, verbal demonstration, verbal instruction, and tutoring), the comic strip showed a badminton court with a pupil and a teacher each holding a badminton racket. The teacher was talking to the pupil and the teacher’s words were written in a balloon. In the fifth condition (autonomy), the pupil was alone and was saying what he or she was doing (the pupil’s words were also written in a balloon). The experimenter talked to each participant individually and read the words in the comic strip. She described each learning condition as follows. In silent demonstration, “the teacher shows the pupil how to do the badminton service and then the pupil tries to do it him/herself.” In verbal demonstration, “the teacher shows the pupil how to do the badminton service and explains it at the same time. Then the pupil tries to do it him/herself.” In verbal instruction, “the teacher explains how to do the badminton service to the pupil, and then the pupil tries to do it him/ herself.” In tutoring, “the pupil tries to do the badminton service alone, and if he/ she doesn’t succeed, the teacher helps him/her.” In autonomy, “the pupil tries to do the badminton service alone, and the teacher never helps him/her.” Stage 4. The tutor asked each participant to choose the learning condition he or she preferred. The choice was written down by the tutor.
Stage 5. The tutor asked each pupil to carry out the task. The participant was informed that he or she had eight trials. The number of trials was chosen based on a pre-experiment conducted with novices of the same age. In the conditions with demonstration or verbal instruction, the tutor helped pupils four times: before Trial 1, before Trial 3, before Trial 5, and before Trial 7. In the tutoring condition, the help was given at the same times but with the exception that no help was given before Trial 1; in this learning condition, then, participants had to carry out the task alone on the first two trials. In the two learning conditions with verbal instruction, the instructions were “To serve in badminton, you must be in the serving area of the half-court (right or left), behind the short service line. The shuttlecock must fall down in the diagonally opposite half-court.” In the tutoring condition, verbal instructions depended on the participant’s particular difficulties. For example, if the participant had poor coordination when hitting the shuttlecock, the verbal instructions were, “You must shorten the handle of the racquet by putting your hand nearer to the head of the racquet.” Stage 6. Each participant answered the self-efficacy questionnaire and the task-representation questionnaire a second time.
Data Analyses The statistical analyses were performed in four steps. First, the preferred learning conditions were examined, and the hypothesized relationship between cognitive style and choice of learning condition was tested using the chi-square test. Second, the effects of gender, cognitive style, motor-task performance (Trials 1 to 8), self-efficacy beliefs, and task representation on the choice of a learning condition were explored using a discriminant analysis. Third, an RMANOVA was conducted to compare the four guidance conditions, and an ANOVA was conducted to examine the gender effect and the learning effect, that is, pretest (T1) vs. posttest (T8). Fourth, correlations among the variables were computed.
Results Interrater Reliability Analyses Interrater reliability was calculated for the GEFT score (100% agreement) and for motor performance (91% on the pretest, 93% on the posttest). The results of the relationships between preferred learning condition and cognitive style are presented in Table 1.
Preferred Learning Condition The learning conditions preferred by the participants were verbal demonstration (n = 18), tutoring (n = 15), and verbal instruction (n = 15). Only 2 participants chose to work alone without help.
Relationship Between Cognitive Style and Chosen Learning Condition There was no significant effect of cognitive style on the choice of an instructional method. No significant difference appeared, either, when the learning
Vincent-Morin and Lafont
Choices Made by Participants in Each Group
Cognitive style I FD FI/FD Total
1 1 0 2
5 6 4 15
3 6 6 15
3 4 5 12
6 8 6 20
n (%) 18 25 21 64
(28.12) (39.07) (32.81) (100)
Note. Choices: A = autonomy; T = tutoring; VI = verbal instruction; SD = silent demonstration; VD = verbal demonstration. Cognitive style: FI = field independence; FD = field dependence; FI/FD = intermediate.
conditions were grouped into two categories: substantial help, which included verbal demonstration (VD), silent demonstration (SD), and verbal instruction (VI); and little help, which included tutoring (T) and autonomy (A) (χ2 = .59, p > .05).
Discriminant Analysis of All Variables A discriminant analysis was carried out to assess the best predictors of the learning-condition choice. Because only 2 participants chose the autonomy condition, it was combined with tutoring to form a single category (T&A: little help), making only four categories in all. The ranking coefficients are presented in Table 2. There were no effects of cognitive style or gender on the choice of an instructional method. Some variables, however, partially predicted the learningcondition choice. Participants with greater feelings of self-efficacy chose the verbal-instruction condition more often; those with better task representation chose silent demonstration and verbal demonstration more frequently; participants with higher motor-performance scores on Trial 1 selected silent demonstration more often; and those with lower initial self-efficacy levels opted for verbal demonstration more often.
Gender Effect Table 3 presents the means and standard deviations of all measures for boys and girls. Motor Performance. On the pretest measure of the badminton service, the ANOVA indicated that gender had a significant effect, F(1, 62) = 5.59, p < .05: boys performed better than did girls. Boys also did better than girls on the posttest, F(1, 62) = 12.66, p < .01. Task Representation. No significant difference was found between boys and girls on the pre- or posttest. Self-Efficacy Beliefs. Perceived self-efficacy level was significantly higher for boys than for girls on the pre- and posttests, F(1, 62) = 17.0, p < .001 and F(1, 62) = 10.3, p < .01, respectively. Self-efficacy strength was also significantly higher
Table 2 Fisher’s Linear and Discriminant Functions: Coefficients of Ranking Functions for Choices and Factors Factor FDI Gender TR1 SEL1 SES1 T1 Constant
.31 13.60 1.12 7.66 .60 1.94 –28.72
.29 13.26 1.36 8.48 .28 1.73 –28.53
.26 13.20 1.80 7.75 .19 2.47 –28.60
.22 12.68 1.75 7.09 .29 1.84 –24.30
Note. Choices: T&A = tutoring and autonomy; VI = verbal instruction; SD = silent demonstration; VD = verbal demonstration. Factors: FDI = field dependence–independence; TR1 = task representation on pretest; SEL1 = self-efficacy level on pretest; SES1 = selfefficacy strength on pretest; T1 = motor performance on pretest.
Table 3 Means and Standard Deviations for Boys and Girls for Motor and Cognitive Variables Pretest
Boys N = 35
Girls N = 29
Boys N = 35
Girls N = 29
Field dependence– independence
Vincent-Morin and Lafont
for boys than for girls on the pretest, F(1, 62) = 6.87, p < .02, but no significant difference was found between boys and girls on the posttest. Cognitive Style (FDI). The gender effect was nonsignificant on this variable. Learning-Condition Choice. No difference was found between boys and girls. The chi-square test was nonsignificant.
Learning-Condition Effect The results for learning-condition effect are presented in Table 4. To examine the differences between learning conditions, a 4 ⫻ 2 RMANOVA (Learning Condition ⫻ Testing Time) was conducted. Again, the four learning conditions were tutoring and autonomy (T&A), verbal instruction (VI), silent demonstration (SD), and verbal demonstration (VD). The testing times were the pre- and the posttest. Motor Performance. We observed a significant learning-condition effect on motor performance, F(3, 60) = 3.62, p < .02. Tukey’s post-hoc test showed that silent demonstration (m = 3.79) was significantly superior to verbal instruction (m = 2.73) and to verbal demonstration (m = 2.9). Task Representation. Task representation showed no learning-condition differences before or after the learning session. Self-Efficacy. On the pretest, the learning condition had a significant effect, F(3, 60) = 3.02, p < .05. Tukey’s post-hoc test showed that the participants who chose verbal instruction (m = 3.06) had a self-efficacy level significantly higher than those who chose verbal demonstration (m = 2.3). No significant difference in self-efficacy level was observed among groups after the learning session or in selfefficacy strength before or after the learning session.
Learning Effect Motor Performance. The repeated-measures ANOVA showed that motor performance increased significantly for all participants, F(1, 60) = 15.88, p < .01. Task Representation. The task representation improved significantly for all participants pooled between the beginning and end of the learning session, F(1, 60) = 59.4, p < .01, as well as for each group taken separately. Self-Efficacy Beliefs. No significant pretest or posttest difference was found for self-efficacy strength for all participants pooled or by group.
Correlations Among Variables Cognitive Style. Cognitive style was positively correlated with the initial task representation (r = .40, p < .01) and with the posttest task representation (r = .37, p < .01). Motor Performance. Motor performance on the pretest (T1) was positively correlated with task representation at the end of the session (TR2): r = .30, p < .05. Motor performance on the posttest was positively correlated with self-efficacy level at the beginning (SEL1) and end (SEL2) of the session (r = .32, p < .05 and r = .52, p < .01, respectively). Self-Efficacy. The self-efficacy level before the learning session was positively correlated with the self-efficacy level at the end of the session (r = .52, p < .01) and with self-efficacy strength at the beginning (r = .45, p < .01). At the end of the
17 15 12 20 64
.90 .88 1.21 1.02 .99
SD 3 2.93 3.33 3.40 3.17
Mean 17 15 12 20 64
.79 1.16 .88 .94 .95
SD 2.82 3.06 2.66 2.30 2.68
6.59 5.33 4.91 4.65 5.37
T&A VI SD VD Total
17 15 12 20 64
1.97 3.01 2.64 2.49 2.59
SD 6.35 5.06 6.08 5.05 5.59
Mean 17 15 12 20 64
2.37 3.59 1.97 3.05 2.85
SD 2.59 2.26 3.33 2.35 2.57
17 15 12 20 64
17 15 12 20 64
1.06 1.16 1.37 1.42 1.29
.64 .45 .77 1.03 .81
3.65 3.20 4.25 3.45 3.59
2.52 2.53 2.75 2.00 2.40
17 15 12 20 64
17 15 12 20 64
1.50 1.08 1.28 1.53 1.39
.87 .91 .62 .97 .90
Note. Measures: TR1 = task representation on pretest; TR2 = task representation on posttest; SEL1 = self-efficacy level on pretest; SEL2 = self-efficacy level on posttest; SES1 = self-efficacy strength on pretest; SES2 = self-efficacy strength on posttest; T1 = motor performance on pretest; T8 = motor performance on posttest. Choices: T&A = tutoring and autonomy; VI = verbal instruction; SD = silent demonstration; VD = verbal demonstration.
Cognitive and Motor Variables in the Four Learning Conditions (Part 2)
1.76 1.73 2.25 2.10 1.95
T&A VI SD VD Total
Cognitive and Motor Variables in the Four Learning Conditions (Part 1)
Learning-Method Choices 237
Vincent-Morin and Lafont
session, the self-efficacy level was positively correlated with the initial self-efficacy strength (r = .30 p < .05). Self-efficacy strength at the beginning and end of the session were correlated (r = .52, p < .01).
Discussion The main purpose of this study was to analyze the role of cognitive style (FDI) on the choice of an instruction method. Contrary to predictions, no difference between FDI groups was found. The results of this study are not fully consistent with the hypothesis set forth, in that regardless of whether the pupils were found to be field independent or field dependent, they primarily chose strong guidance. The preferred choice of a strong instructional method can be explained by the nature of the task. Indeed, a badminton service is defined by the rules of the game and by environmental factors that impose heavy constraints at the motororganization level and thereby increase the difficulty of the task. Moreover, the newness of the task might also account for these results. We can assume that the pupils wanted to have precise visual and verbal information before undertaking the task. In line with Huteau’s (1987) work, the experimenter represents a source of information, which is manifested in FD pupils by the need for a social relation and in FI participants by the need for information about the task. This phenomenon could explain the fact that all participants preferred one of the stronger forms of help. Furthermore, the tendency to choose a highly assisted method can be interpreted in terms of the pupils’ habits and expectations in matters of learning. These participants might have been more inclined to expect the experimenter to offer help, just as pupils expect help from a teacher. The choices preferred by the pupils were identical to those observed by Winnykamen and Servant (1986). Again, in addition to cognitive style, the other variables studied pertained to self-efficacy beliefs, task representation, and gender. Silent demonstration was the choice most often made by the pupils who had the most elaborate task representation, as well as by those who obtained the highest scores on the first trial. This choice can be explained by the need to confirm knowledge validated in the task-representation questionnaire and by learning habits (i.e., the tendency to seek substantial help from the experimenter). Pupils who had the highest self-efficacy level chose verbal instruction. This form of learning involves less help than does silent demonstration and verbal demonstration. By contrast, pupils with the lowest feelings of self-efficacy opted for the most assistance: verbal demonstration. This result is consistent with Puustinen and Winnykamen’s (1998) observations. Their study showed that children who had the lowest feelings of self-efficacy were the ones who asked for the most help. It would seem, then, that it is the initial feeling of self-efficacy that accounts for the instructional method chosen more so than the subject’s field-dependent/independent cognitive style. As far as gender, the results obtained showed that it did not affect the choice of a learning method. Gender, however, did have an impact on motor performance: the boys outperformed the girls. Boys also had higher self-efficacy scores than did girls. One possible reason is that boys might be more concerned with social comparison, which denotes a greater ego: studies have shown that males are generally more ego-oriented and more motivated to win than females (White & Duda, 1994).
A second possibility is that the nature of the task might have affected the outcome: badminton is labeled as primarily a male sport. Finally, the difference between boys and girls might have been reinforced by social and cultural customs: boys are more often exposed to anticipation-coincidence tasks and ball sports (Anderson, Gebhart, Pease, & Rupnow, 1983). This greater potential for boys to possess the motor resources solicited by this task might affect boys’ feelings concerning their ability to succeed. These results are in line with the study conducted by Frömel, Formankova, and Sallis (2002) on physical activity and sport preferences in 10- to 14-year-old children. These authors showed that girls preferred activities with aesthetic and recreational characteristics, whereas boys preferred sports-performance activities. The present task would be considered a sports-performance activity. Furthermore, our results fit with those obtained by Fehlker (2003), who found significant gender differences in knowledge of baseball and softball: males answered more questions regarding the rules and game knowledge than did females; females underestimated the number of questions they believed they would get right, whereas males overestimated this number. The present results can be interpreted in the light of the studies conducted by Ames (1983) and Nelson-Le Gall (1985) on children’s help-seeking behavior in problem-solving situations. These authors showed that seeking help promotes the construction of knowledge and the acquisition of skills. Participants ask for help in order to solve their problem efficiently. Ames (1983) argued that help-seeking behavior is an autonomous problem-solving strategy that fits with a motivational attitude centered on mastering the task and is dependent on self-assessments of one’s ability to perform it. In other words, we can assume that the participants with the highest task-representation scores chose silent demonstration based on an evaluation of their shortcomings in the task to be performed. Another goal of the experiment was to compare the effects of the different learning conditions on several variables. Silent demonstration was observed to be superior to verbal instruction. This result is in keeping with the work by Bandura (1986) and Carroll and Bandura (1982), who pointed out the effectiveness of external models in motor-skill acquisition. The fact that verbal demonstration did not prove more effective than other methods might be a result of the fact that, for this type of task, verbal information and visual information were redundant. It can also be explained by the lower self-efficacy level of these participants on the pretest. In spite of the progress made between the pretest and the posttest in the area of motor performance, perceived self-efficacy levels decreased for all participants. One can assume that the task characteristics led the participants to an awareness of the difficulty of the task, and this caused them to be cautious in their self-judgments. The observed relationships between posttest self-efficacy level and final performance scores are in line with the study by Shunk (1989), who found a reciprocal relationship between performance and self-efficacy beliefs. The results of this study bring out the complexity of the motor-skill learning process. The findings are partially inconsistent with the hypothesis set forth. One can assume that the participants associated the steps of the experiment with school activities and thereby granted the task a social significance that had an impact at the motivational level. This led them to choose highly assisted forms of learning in accordance with their emotional states and their estimates of their level of competence. Our results point out some new directions for constructivist pedagogies in
Vincent-Morin and Lafont
physical education: Those who argue for less-directed instruction must take into account students’ preferences and attitudes toward learning. Even though autonomybased teaching methods are currently the vogue in France, more instructive and demonstrative techniques should not be neglected, especially when students request this form of assistance. Our study shows, in addition, that the latter can be quite effective. This study emphasizes the complexity of interactions between pupils, task, and teacher or tutor. Finally, it demonstrates the need for physical education teachers to consider their pupils in an ecological context in order to reach a global understanding of the processes used in the learning activity.
References Amade-Escot, C. (1991). Caractérisation de la formation didactique initiale des enseignants d’éducation physique et évaluation de ses retombées sur les compétences professionnelles. (Characteristics of physical education teacher training in didactics and assessment of its effects on the professional abilities acquired by novice teachers). Unpublished doctoral dissertation, Université de Toulouse III, Toulouse, France. Amade-Escot, C. (2000). The contribution of two programs on teaching contents: “Pedagogical content knowledge” and “didactics of physical education.” Journal of Teaching in Physical Education, 20, 78-101. Ames, R. (1983). Help-seeking and achievement orientation: Perspectives from attribution theory. In B. De Paulo, A. Nadler, & J.D. Fischer (Eds.), New directions in helping: Vol. 2, (pp. 165-186). NY: Academic. Anderson, D.F., Gebhart, J.A., Pease D.G., & Rupnow, A.A. (1983). Effects of age, sex, and placement of a model on children’s performance on a ball-striking task. Perceptual and Motor Skills, 57, 1187-1190. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice Hall. Beaudichon, J., Verba, M., & Winnykamen, F. (1988). Interactions sociales et acquisition de connaissances chez l’enfant: une approche pluridimensionnelle (Social interactions and knowledge acquisition in children). Revue Internationale de Psychologie Sociale, 1, 129-141. Brau-Antony, S. (2003). Caractéristiques de situations didactiques proposées par des enseignants d’EPS en volleyball (Characteristics of didactic situations proposed by physical education teachers in volley-ball). In C. Amade-Escot (Ed.), Didactics in Physical Education (pp. 169-192). Paris: Revue EPS. Brousseau, G. (1986). Fondements et méthodes de la didactique des mathématiques (Foundations and methods in mathematics didactics). Recherches en Didactique des Mathématiques, 7(2), 33-115. Brousseau, G. (1997). Theory of didactical situations in mathematics. Dordrecht, Netherlands: Kluwer. Bruner, J.S. (1983). Le développement de l’enfant: savoir faire, savoir dire. (Child development: learning to do, learning to say). Paris, Presses Universitaires de France. Caroll, W.R., & Bandura, A. (1982). The role of visual monitoring in observational learning of action patterns: Making the unobservable observable. Journal of Motor Behavior, 14(2), 153-167.
Dyson, B. (2001). Cooperative learning in an elementary physical education school. Journal of Teaching in Physical Education, 20, 264-281. Dyson, B. (2002). The implementation of cooperative learning in an elementary physical education program. Journal of Teaching in Physical Education, 22, 69-85. Famose, J.P. (1990). Apprentissage moteur et difficulté de la tâche. (Motor learning and task difficulty). Paris: INSEP. Fehlker, J.H. (2003). Gender differences in athletics: baseball and softball knowledge. Retrieved September 10, 2004, from http://clearinghouse.mwsc.edu/manuscripts/437.asp Frömel, K., Formankova, S., & Sallis, F. (2002) Physical activity and sport preferences of 10- to 14-year-old children: A 5-year prospective study. [Acta University Palacki, Olomuc, Czech Republic]. Gymnica, 32(1), 11-16. Huteau, M. (1987). Style cognitif et personnalité. (Cognitive style and personality). Lille: Presses Universitaires de Lille. Konstadt, N., & Forman, E. (1965). Field dependence and external directedness. Journal of Personality and Social Psychology, 1, 490-493. Lafont, L. (1994). Modalités sociales d’acquisition d’habiletés motrices complexes: Rôles de la démonstration explicitée et d’autres procédures de guidage. (Social modes of motor skill acquisition: Roles of verbal demonstration and other learning conditions). Unpublished doctoral dissertation, Université de Paris V, Paris, France. Lafont, L. (2002). Techniques, modèles et didactique de l’EPS. (Techniques, models, and physical education didactics). STAPS, 59, 57-70. Lafont, L., & Vincent-Morin, M. (2000). Etude des formes d’aides choisies par les élèves pour la résolution d’un problème moteur (Study of types of help selected by pupils in a problem-solving situation). Unpublished manuscript, Université Victor Segalen Bordeaux 2, Bordeaux, France. MacCullagh, P., Weiss, M., & Ross, D. (1989). Modeling considerations in motor skill acquisition and performance. Exercise and Sport Sciences Review, 17, 475-513. Nelson-Le Gall, S. (1985). Help-seeking behavior in learning. Review of Research in Education, 12, 55-90. Newell, K.M., & Walter, C.B. (1981). Kinematic and kinetic parameters as information feedback in motor skill acquisition. Journal of Human Movement, 7, 235-254. Paillard, J. (1974). Le traitement des informations spatiales. De l’espace corporel à l’espace écologique (Processing of spatial information: From corporeal space to ecological space). Paris Symposium APLSLF. Paris: Presses Universitaires de France. Poulton, E.C. (1957). On prediction in skilled movements. Psychological Bulletin, 54(6), 467-478. Puustinen, M., & Winnykamen, F. (1998). Influence du sentiment d’auto efficacité dans la demande d’aide chez des enfants de 8 à 9 ans (Influence of self-efficacy feeling on requests for help by children 8 to 9 years old). Enfance, 2, 173-188. Ricoeur, P. (1999). Liberté (Freedom). (Electronic version 5, CD ROM). The Encyclopedia Universalis Multimedia, France. Rink, J. (2001). Investigating the assumptions of pedagogy. Journal of Teaching in Physical Education, 20, 112-128. Shunk, D.H. (1989). Self efficacy and cognitive achievement: Implications for students with learning problems. Journal of Learning Disabilities, 22(1), 14-22. Siedentop, D. (1994). Sport education: Quality P.E. through positive sport experiences. Champaign, IL: Human Kinetics. Siedentop, D. (2002). Sport education: a retrospective. Journal of Teaching in Physical Education, 13(4), 375-394.
Vincent-Morin and Lafont
Thorpe, R., Bunker, D., & Almond, L. (1986). Rethinking games teaching. Loughborough, UK: University of Technology. White, S.A., & Duda, J.L. (1994). The relationship of gender, level of sport involvement, and participation motivation to task and ego orientation. International Journal of Sport Psychology, 25, 4-18. Winnykamen, F., & Lafont, L. (1990). Place de l’imitation-modélisation parmi les modalités relationnelles d’acquisition: Le cas des habiletés motrices (Role of imitation-modeling in relational processes of acquisition: The case of motor skills). Revue Française de Pédagogie, 92, 23-30. Winnykamen, F., & Servant, G. (1986, March). Contribution à l’étude des relations d’aide pour la résolution d’un problème (Contribution to the study of help relationships in problem-solving situations). Poster session presented at the C.R.E.S.A.S. meeting, Paris, 1986. Witkin, H.A. (1965). Psychological differentiation and form of pathology. Journal of Abnormal and Social Psychology, 70, 317-336. Witkin, H.A. (1967). A cognitive style approach to cross-cultural research. International Journal of Psychology, 2, 233-250. Witkin, H.A., Oltman, P.K., Raskin, E., & Karp, S.A. (1971). A manual for the embedded figures test. Palo Alto, CA: Consulting Psychologists Press.