Differences in the development of creative competencies in children schooled in diverse learning environments

Available online at www.sciencedirect.com Learning and Individual Differences 18 (2008) 381 – 389 www.elsevier.com/locate/lindif Differences in the development of creative competencies in children schooled in diverse learning environments Maud Besançon ⁎, Todd Lubart Université Paris Descartes, Laboratoire de Psychologie et Neurosciences Cognitives (CNRS FRE 2987), France Received 27 March 2007; received in revised form 9 November 2007; accepted 13 November 2007 Abstract Studies on the development of creativity have highlighted the impact of learning environments. In particular, pedagogical approaches are hypothesized to differ concerning their emphasis on individual initiative, and action-based learning. A semi-longitudinal study was conducted during two consecutive years with 210 children in elementary schools with traditional and alternative pedagogical approaches. Our results highlight (1) an influence of pedagogy on children's creative performance; (2) a positive influence of alternative pedagogy on creative development from year 1 to year 2 mainly for Montessori school. Children's creative performance was influenced not only by the type of task but also by the type of school. © 2007 Elsevier Inc. All rights reserved. Keywords: Creativity; Children; Pedagogy 1. Introduction 1.1. Learning environments Pedagogical practices evolve continually in order to take into account societal needs. Since the 1950s, creativity, defined here as the ability to produce novel, original work that meets contextual constraints, has been viewed as an increasingly important characteristic for professional success and personal development (Amabile, 1996; Gardner, 1996; Lubart, Mouchiroud, Tordjman & Zenasni, 2003; Ochse, 1990; Sternberg & Lubart, 1995). In this paper, the impact of different learning environments on children's creative development is examined. Various pedagogies, which contrast on their potential support for creativity, have been implemented since the beginning of the 20th century. We examine the prototypical characteristics of traditional and alternative pedagogies. Then, task-centered factors concerning different components of creativity are described. Finally, the results of an empirical study are presented in which traditional and alternative learning environments are compared in terms of their effects on creative development. 1.1.1. Traditional pedagogy The main goals of most educational systems are to transmit knowledge, rigorous working habits and societal values. According to Danvers (2003), traditional pedagogy is characterized by (1) a central role assigned to the teacher: teaching is frontal, i.e. the teacher is in front of the class; (2) an impersonal relation with pupils because there are usually many pupils in a class; and (3) the importance of abstract knowledge which is not always linked with the everyday life. Generally, teaching is dispensed in a collective way and the teacher's authority is based on the fact that the teacher masters the subject matter (Morandi, 1997). Traditional pedagogy is supposed to offer a structure in which children can evolve. At the beginning, children acquire knowledge but they do not critique it. Then, pupils develop the capacity to criticize and propose new ideas, which occurs because they have initial cultural knowledge. Class time is devoted to learning and critical thinking and there is little attention to creativity. ⁎ Corresponding author. Laboratoire de Psychologie et Neurosciences Cognitives, Université Paris Descartes 71 avenue Edouard Vaillant, 92100 Boulogne-Billancourt Cedex, France. Fax: +33 1 55 20 59 85. E-mail address: maud.besancon@univ-paris5.fr (M. Besançon). 1.1.2. Alternative pedagogy During the 20th century, some educators and psychologists developed alternatives to the traditional approach. According to Piaget (1969/2004), it would be stimulating for children to make 1041-6080/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.lindif.2007.11.009 382 M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 discoveries by themselves and thereby construct their knowledge through actions. The teachers' role is, in this view, to provide the pupil with a rich environment including situations favoring the emergence of cognitive and socio-cognitive conflicts, viewed as the engines of development. Vygotsky (1934/ 1997) introduced the concept of “Zone of Proximal Development” (ZPD): a child could then be defined with respect to a current level of development and a second, potential level of development that could be attained with assistance from an adult (Schneuwly & Bronckart, 1985). Between the two levels there is the ZPD. Thus, this concept indicates the importance of mediation in teaching methods: the teacher must engage the pupil in the ZPD and allow him or her to progress, after which, the pupil will be able to succeed alone. Montessori (1958/2004) considered that imaginative thought must be based on contact with reality and sensory impressions. Based on these principles, she proposed targeted activities (such as learning how to see, feel, hear, and graduate colors or thermal stimuli…), beginning in preschool. For example, some pupils begin to learn the letters of the alphabet with the sense of touch: children had to follow the outline of letters drawn in various textures and their teachers provide directions on how to conduct the movements. For Montessori (1958/2004), to foster imagination, teachers should let it evolve freely (freedom to choose an activity, to handle objects placed at their disposal) but within a well established framework. In a similar vein, Freinet, a French pedagogue, proposed a “psychological” theory of the child as primarily founded on two postulates (Freinet, 1994): (1) “Vital boost”: a child is animated by a natural dynamism, which should not be opposed; (2) “Experimental fluctuation”: error is a means to access knowledge. Moreover, creativity is allowed and fostered in various activities in which children themselves create their own productions from their knowledge or personal experience. Activities such as the “free-text”, the “free drawing” or the school newspaper allow creativity. For example, in the “free-text” activity, children write a text on a subject that is important to them. This text is then corrected and can be sent thereafter to another school or discussed in the class council (with other children and the teacher). Given the fundamental differences between traditional and alternative pedagogical methods, the question of effects on children's creative development can be raised. 1.2. Studies on the influence of school on children's creativity During the second half of the 20th century, several studies tried to examine children's performances in traditional compared to alternative pedagogies. In particular, Horwitz (1979) conducted a review of the literature from the 1930s to the late 1970s. In the 1930–1940s, studies compared performances obtained by children either attending alternative or traditional pedagogy schools; children who were exposed to alternative pedagogy showed more initiative, better capacity to face problems, knowledge acquisition and social participation. In the 1950–1960s, the studies focused on performance differences between children exposed to traditional pedagogy and alternative pedagogy. First, the results showed that there was no significant difference either on school performance, problem solving tasks or creative thinking tasks. Second, children exposed to alternative pedagogies described themselves as less rigid, more subtle and imaginative; they were more open, less conventional and had fewer stereotypes concerning social roles. The results indicate also that children attending alternative schools were more cooperative, less competitive and more accessible than children exposed to traditional pedagogies. Based on research carried out in the 1970s, Horwitz (1979) found, regarding creative thinking, that children exposed to open classes outperformed those in traditional classes. However, the results did not show a complete consensus: whereas the majority of studies found better results for children in alternative pedagogy, some highlighted benefits for children in traditional pedagogy. These divergent results could be due to the fact that creativity is a concept that encompasses many definitions and, in addition, no indication is provided concerning the family and cultural environment, which may vary across schools and studies. Two studies focused particularly on the effects of Freinet's pedagogy. First, Avanzini and Ferrero (1976–1977) found that, in general, the Freinet pedagogy was as effective as traditional pedagogy. However, with regard to creativity, children in Freinet schools (in France) showed better performances than children in traditional schools. One explanation is that teachers practicing the Freinet techniques grant greater freedom to pupils and allow them to express themselves with no threat of giving a grade or judgment; thus, these pupils take more initiatives, are more open to the world, qualities that are necessary for the development of creativity.1 In another study, Frankiewicz (1984) examined in a Polish school the effect of the free-text exercise: in a first group, teachers used the free-text task whereas in the second group, teachers used traditional lessons of Polish language. Results indicated in a post-test that pupils who had the free-text task showed better performances on different indices of creative thinking (fluency, flexibility and originality). Moreover, these same children proposed more original, richer and more coherent stories than children who did not receive the free-text method. These results suggest that practicing free-text composition supports the development of creative performance. However, it is difficult to draw any strong conclusions about the superiority of this method because there is no information on pupils' initial level of performance: indeed, it is possible that pupils of the freetext group already had higher performances than other children in the pretest. Thomas and Berk (1981) conducted a literature review concerning the effects of different school environments on children's creativity, which revealed non-conclusive results as well. They emphasized that these divergent results can be due to the various definitions allowed for the school environment. In order to clarify the findings, Thomas and Berk undertook a study analyzing the creative performance of 225 children in first or second grade, in six different schools. Their hypothesis was that the environment that best supports the development of creative performance is an intermediate one, neither too 1 Although these results seem interesting, the specific details on the measures of creative performance were not reported: thus, we do not know how these abilities were evaluated. M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 structured, nor too open or flexible. Their results highlighted a complex relation for the development of creativity, which is influenced by the type of school, the pupil's gender and the type of creativity (verbal or figural). Notably, they found that (1) an intermediate environment promoted best creative development; (2) in general, boys were more creative than girls. In conclusion, the results on pedagogical effects on creativity are preliminary. It is difficult to conclude on the influence of learning environments because the few studies that have been conducted focus only on creative performance at one moment in time and do not take into account certain variables such as the nature of the creativity tasks. 1.2.1. Effects of school environment may vary depending on children's initial level According to Sternberg and Lubart (1995), creativity is a cognitive ability which requires a confluence of six distinct and interrelated resources: intelligence, knowledge, intellectual style, personality, motivation and environmental context. Individual differences in abilities reflect different patterns of prior learning (Anastasi, 1970; Snow, 1978, 1980). According to Snow (1994), different levels of ability development and different patterns of ability differentiation result from different types of educational programs. In addition, each individual's learning history is also unique and personal because individuals perceive situations differently according to their own history or their interests. Thus, children's creative performance could be influenced by their initial creative aptitude, by their learning environment, and by the interaction between these two variables. In this vein, it is possible that the effects of pedagogy will be greater for pupils with certain levels of initial creative ability compared to others. The effect of the pedagogy is hypothesized to be greatest when children have relatively low creative potential. 1.2.2. Domain- and task-centered factors Creativity is partly domain general (Plucker, 1998) but largely domain specific and even task specific (Baer, 1999; Lubart & Guignard, 2004). Models of the creative process have distinguished various phases of processing and kinds of thinking that may come into play. One broad distinction opposes divergent thinking, in which the goal is to explore multiple cognitive paths, and convergent thinking which seeks to focus on a single, perhaps optimal path (Simon, 1960). Complex creative performance tasks involve certainly both kinds of processes, in various degrees and in specific sequences that favor the generation of new ideas in a task. In this vein, Lubart and Guignard, (2004) proposed that the moderate correlations observed between different creative performance tasks stem from the fact that there is a different mix of cognitive operations, and knowledge, involved in each creative domain and task. In related work, Maker (1996) distinguished several types of problem solving tasks in the school setting. A continuum is proposed from closed problems (Type I and Type II) to intermediate problems (Type III and Type IV) to open problems (Type V and Type VI). According to Maker and Nielson (1995), only the first four types of problems, characterized mainly by known standardized methods and solution paths, are employed in 383 the school system (Types I, II, III,IV). However, only the last three (Types IV, V and VI) of problem solving activities engage creativity. Problem types I to III emphasize convergent thinking, problem type IV focuses on divergent thinking, and problem type V and VI combine both kinds of thinking. These different considerations argue in favor of using a range of creative performance tasks in order to measure the effects of pedagogy on creative development. Both divergent thinking measures and integrative creative thinking tasks are necessary. Divergent thinking tasks require children to generate many ideas from a given starting point and have been the basis of creativity testing in schools during the past fifty years. These measures, in both verbal and figural domains, must be used together with tasks that involve integrating numerous generated ideas into a cohesive production. In this way, integrative tasks in the verbal (story-telling) and artistic (drawing composition) domains complete classic divergent thinking measures. 1.3. Overview of the empirical contribution In the current study, we examined the influence of learning environment on the development of children's creative performance. We hypothesized that children schooled in alternative pedagogies (Montessori and Freinet) will have greater creative performance than children schooled in traditional pedagogy. This study is the first, to our knowledge to compare these three types of pedagogies (Montessori, Freinet and traditional) in terms of multiple measures of creative performance. Creativity was measured with two types of tasks (divergent and integrative ones) across two content domains (verbal and graphic expressive domains), in order to examine the consistency of pedagogical effects on creative performance. In addition, we examined the extent to which the school pedagogy effect may be greater over time (a one year span) for some children than for others, as a function of children's initial scores on a creativity task, children's grade level, and gender. To address these issues, we conducted a longitudinal study with approximately one year between test and retest for children from traditional and alternative pedagogies. 2. Empirical study 2.1. Method 2.1.1. Population After obtaining school and parental consent, 211 children (96 boys and 115 girls) participated. Children were enrolled in 1st grade to the 4th grade, the first year of the study and from 2nd grade to the 5th grade, in the second year (from 7 to 12 years old). Children came from four primary schools in Paris or its suburbs.2 Five judges (M age = 27; SD = 1.73) evaluated story creativity and five other judges assessed drawing creativity (Mage = 30.8; SD = 10.8). Judges were 10 university professionnals or PhD students who work regularly in the field of creativity (Table 1). 2 Only one children is not included in the analysis because this children had repeated a grade. 384 M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 Table 1 Distribution of children, mean age (and SD)according to the school and to grade level in the second year Girls Montessori (Paris) Freinet (Suburb Paris) Traditional (Paris) Traditional (Suburb Paris) 2nd grade 3rd grade 4th grade 5th grade 2nd grade 3rd grade 4th grade 5th grade 2nd grade 3rd grade 4th grade 5th grade 2nd grade 3rd grade 4th grade 5th grade Boys N Mean age (SD) Total sample 40 5 6 11 6.32 (0.44) 4 7 11 7.32 (0.45) 5 3 8 8.06 (0.52) 6 4 10 9.60 (0.70) 7 10 17 6.57 (0.50) 7 3 10 7.48 (0.72) 3 7 10 8.75 (0.46) 7 7 14 9.56 (0.74) 5 4 9 6.49 (0.50) 5 7 12 7.72 (0.76) 9 6 15 8.65 (0.71) 22 9 31 9.48 (0.48) 6 9 15 6.40 (0.47) 14 6 20 7.61 (0.58) 6 6 12 8.25 (0.57) 3 2 5 9.27 (0.80) 51 67 52 2.1.2. Material In order to examine distinct facets of creativity, we used various measures of creativity which differed on type of task (divergent thinking versus integrative task) and domain of expression (verbal versus figural). 2.1.3. Divergent thinking tasks We used three divergent thinking tasks from the Torrance Tests of Creative Thinking (Torrance, 1976), two were verbal and one was figural: (1) Unusual uses for a cardboard box: in this task, the child must name, orally, as many different uses as he or she can make with one or several boxes (time allowed: 3 min). (2) Improvement of a toy: we present an elephant toy to the child. This toy does not have accessories, it is uniformly grey and in a sitting position. The child is asked to imagine how to make this toy funnier and more entertaining. The child has to name, orally, all his or her ideas during 3 min. (3) Parallel lines: this task uses figural material. The child is provided with three pages. On each, several pairs of parallel lines are presented. The child must make as many drawings as possible starting from 30 pairs of parallel lines. The parallel lines must play an integral part in their drawing; several pairs of parallel lines can be used for the same drawing. At the end of the 10 min allowed for this test or when the child does not have any more ideas, the child is asked to give a title to each drawing. Different indices of creativity are calculated for these divergent thinking tasks: fluency (number of ideas), flexibility (number of categories), and originality (statistical rarity). Originality and flexibility scores are based on French norms (Torrance, 1976). 2.1.4. Integrative thinking tasks Concerning this type of task, we proposed one verbal and one figural task that required participants to integrate several elements in a complete production. (1) Invent a story: in this task, the child must invent a story from a title which is provided. The story must be as original as possible. Two parallel forms were used. The title of the story, proposed to children for the first year was “the millipede's tennis shoes” and for the second year the title was “the keyhole”. These two titles were previously used in studies with children of the same age range. (2) Invent a drawing: in this task, proposed by Urban and Jellen (1996), the child is presented a sheet of paper on which six elements are displayed. Each child must finish the drawing starting from the six elements (angle, demicircle, S-form, dashed line, small square and point). The same task (form A) was used, the two consecutive years. 2.1.5. Procedure Children who took part in this research were seen, each year, with individual testing. We proposed first the divergent thinking tasks (Unusual uses of a cardboard box, improvement of toy, parallel lines) and then the integrative tasks (Invent a story and Invent a drawing). For each verbal task, responses were tape recorded and later transcribed. In order to judge the creativity of each story and each drawing, we used the consensual assessment technique (Amabile, 1996). For this study, five judges evaluated independently the creativity of each story on a 7-point Likert scale (1 = not at all creative with 7 = very creative) and five other judges evaluated the creativity of the drawings, using the same scoring system. As Amabile (1996) stipulated, no explicit definition of creativity was proposed to the judges. Moreover, we asked them to score productions against each other. 2.2. Results For the divergent thinking (DT) tasks, results concerning originality scores are reported here because of their centrality in the conception of creativity: fluency, flexibility and originality scores were calculated and showed high intercorrelations (r N.70). For the integrative thinking tasks, inter-judge reliability was satisfactory for the two tasks (story task, drawing task: respectively inter-judge agreement α = .93 and α = .92 for the first year and α = .89 and α = .91 for the second year). Inter- 385 M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 judge correlations (r) ranged between .60 and .80, and mean creativity score for each story and each drawing were used (Table 2). The results reported concern children who were present for both years of the study (a small number of children were only present one year and are excluded from analyses). First, we analyze in a MANOVA the influence of school (and pedagogy), gender, school grade level and testing year. Second, the aptitude–treatment interaction is analyzed through regression analyses. Finally, we examine the effect of learning environment on the evolution of creative performance from one year to another (Table 3). 2.2.1. Influence of gender, school, grade and testing year In this section, we analyzed the influence of independent variables on all creative tasks. Our results highlight differences according to the type of tasks (integrative vs divergent thinking tasks). Concerning the integrative tasks (story and drawing), we observed four significant effects (see Table 4): (1) Gender influenced significantly children's creative performance (R / Rao(2,177) = 3.47 ; p b .05): girls showed better performance than boys. (2) The type of school influenced significantly children's creative performance (R / Rao(6,354) = 5.97 ; p b .001): children in the Montessori school showed better performance than children in traditional and Freinet schools. (3) Children's school grade level influenced their performance (R / Rao(6,354) = 5.82 ; p b .001): performance increased progressively from grade 1 to grade 5. The testing year did not influence significantly creative performance. So, there was an effect of age or grade level (cross-sectional comparison) but there was no significant evolution from year 1 to year 2 with the repetition of the task. Concerning the divergent thinking tasks (unusual uses for a cardboard box, improvement of a toy, parallel lines), there were four significant effects (see Table 5): (1) Children's school grade level influenced creative performance (R / Rao(9,428) = 2.59 ; p b .01): performance increased on average from grade 2 to grade 4 and stabilized or decreased in grade 5. (2) The type of school influenced significantly children's creative performance (R / Rao(9,428) = 6.49 ; p b .001): children in the Montessori school showed better performance Table 2 Means and standard deviation (M and SD) of each of the five creativity scores for each school, gender and grade in the first year of testing Montessori (Paris) 1st grade 2nd grade 3rd grade 4th grade Freinet (Suburb Paris) 1st grade 2nd grade 3rd grade 4th grade Traditional (Paris) 1st grade 2nd grade 3rd grade 4th grade Traditional (Suburb Paris) 1st grade 2nd grade 3rd grade 4th grade Gender Cardboard box Improvement toy Parallel lines Story Drawing Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls 5 (4) 7.4 (2.3) 11.4 (12.9) 8.3 (9.1) 14 (20) 6.4 (3.9) 13.3 (9.3) 13.5 (7.3) 7.9 (7.8) 6.4 (4) 13 (3) 9.9 (9) 5.9 (3.9) 16.3 (12.9) 10.9 (10.2) 12.9 (11.5) 2.8 (2.2) 7.4 (5.8) 14.3 (9.5) 6.8 (9.2) 6.9 (3.8) 7.7 (7.8) 4.4 (4.4) 6 (5.1) 10.1 (5) 4 (4) 8.5 (7.8) 7.8 (5.8) 14.2 (10) 8.3 (5.9) 4.5 (2.1) 3.3 (3.5) 10 (5.9) 11.2 (6.2) 11 (7.3) 8.8 (3.6) 16.3 (7.6) 14.4 (4.3) 16.3 (5.4) 17.2 (13.4) 8.6 (8.8) 10.3 (6.7) 16.3 (12.1) 13.3 (8.5) 7.7 (4.9) 15 (10.1) 9.1 (7.9) 11.6 (5.1) 3.3 (4) 5.6 (5.9) 9.3 (5.4) 8.6 (7.4) 8.7 (5.3) 11.5 (6.7) 7.1 (7.2) 7.6 (3.8) 12.8 (7.7) 7.5 (6.5) 7.7 (5.3) 8.4 (6.7) 14.5 (9.1) 10.3 (8.9) 6 (1.4) 8 (6.9) 12 (7.6) 24.8 (14) 21.7 (13.4) 20.5 (17.2) 30.7 (15.9) 18.8 (137) 26 (17.2) 28.8 (19) 6.5 (9.0) 13.4 (11.9) 22 (13.1) 19.4 (17.6) 16.3 (12.9) 30 (11.4) 14.6 (12.5) 16 (12.2) 7 (5.4) 3.4 (3.8) 5.6 (5.7) 11 (11.8) 8.5 (5.8) 17.8 (11.3) 11.2 (10.7) 9.8 (6.3) 16.1 (10) 13.5 (12.3) 14.2 (11.5) 17.4 (10) 40.2 (16.6) 26.5 (10.1) 36 (33.9) 15.3 (7.8) 3.5 (1.1) 4.2 (0.9) 3.1 (1.4) 3.9 (1.2) 4 (0.8) 4.3 (1.5) 4.8 (1.7) 4.3 (1.4) 1.3 (1.2) 2 (1.8) 3.7 (1.4) 3.3 (1.9) 3.8 (2.5) 3.5 (1.7) 2.5 (2.2) 4 (1.1) 0.9 (1.2) 1.6 (1.1) 2.9 (1.1) 3.1 (1.2) 3.1 (1.4) 3.6 (1.5) 2.9 (2.1) 3.8 (1.5) 1.8 (1.3) 2.7 (1.4) 2.4 (1.7) 2.7 (1.4) 3.9 (1.3) 3.9 (1.5) 3.8 (2.5) 3.5 (1.3) 2.6 (1.5) 3 (1.4) 3.5 (1.7) 3.3 (1) 1.6 (0.5) 4.5 (1.1) 5.2 (1.2) 5 (0.8) 2.0 (0.9) 2.7 (1.3) 2.1 (0.8) 3 (1.9) 2.7 (1.3) 4 (1.9) 3.1 (1.9) 4.2 (1.5) 2.5 (1.4) 3 (1.4) 2.4 (1.6) 2.7 (2.4) 2.3 (1.4) 3.3 (1.7) 2.4 (1.4) 2.8 (1.4) 2.7 (1.5) 2.9 (1.9) 1.5 (0.4) 2.4 (1.2) 3.3 (1) 3.2 (2.2) 4.5 (3.3) 3.5 (3.1) 386 M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 Table 3 Means and standard deviation (M and SD) of each of the five creativity scores for each school, gender and grade in the second year of testing Montessori (Paris) 2nd grade 3rd grade 4th grade 5th grade Freinet (Suburb Paris) 2nd grade 3rd grade 4th grade 5th grade Traditional (Paris) 2nd grade 3rd grade 4th grade 5th grade Traditional (Suburb Paris) 2nd grade 3rd grade 4th grade 5th grade Gender Cardboard box Improvement toy Parallel lines Story Drawing Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls 9.2 (9) 12.8 (11.4) 13.1 (11.4) 8.5 (9.7) 9.7 (5.5) 18 (13) 12.8 (3.9) 12 (6.7) 6.8 (3.1) 6.3 (6.5) 10.7 (6.5) 9 (10.1) 5.7 (5.2) 9.3 (6.8) 3.6 (2.2) 5.3 (2.9) 5.3 (5.6) 3 (2.1) 6.3 (5.5) 13.4 (13.6) 5.3 (5.9) 7.2 (7.3) 7.7 (7.3) 5.3 (4.4) 4.7 (2.7) 6.3 (5.1) 2.5 (2.1) 7.2 (7.3) 16.5 (17.12) 4.8 (5.4) 10.5 (6.4) 4 (3.5) 8.2 (5.9) 10.2 (7.1) 11 (7.2) 9.8 (6.3) 9.7 (3.2) 19 (14.8) 17 (13.7) 14.5 (12.0) 5.5 (4.7) 7.6 (5.0) 9.3 (11.4) 8 (2.9) 6.4 (4.1) 4 (2.6) 2.4 (2.9) 5.6 (2.3) 5 5.6) 3.4 (2.2) 8.1 (4.3) 12.6 (7.6) 4.5 (4.2) 7.8 (4.4) 5.8 (4.1) 5.8 (4.9) 7 (5.3) 11.2 (7.0) 5 (4.6) 9.9 (8.6) 13.9 (15.3) 7.8 (2.6) 8 (0) 11.3 (4.2) 15 (12.8) 18 (11.1) 27.4 (9.6) 16.8 (26.5) 26.6 (12.4) 32.2 (17.5) 39.8 (16) 31.8 (27.2) 8.8 (8.3) 11.1 (8.4) 8 (7.5) 8.7 (5.5) 14.4 (8.4) 10.3 (8.7) 9.6 (3.2) 6.1 (4.5) 13.8 (11.1) 8.8 (9.0) 9.7 (5.3) 12.8 (10.5) 7.7 (5) 16 (12.8) 11.3 (9.5) 13.6 (12.5) 12.2 (10.2) 13.7 (9.2) 8 (14.3) 12.5 (8.7) 29.2 (18) 19 (6.9) 23.5 (6.4) 15.3 (5) 3.2 (1.8) 3.8 (1.6) 3.7 (1.2) 3.2 (2.7) 3.5 (1.4) 4 (1.7) 4.4 (0.5) 4.6 (1.7) 1.7 (1.5) 2.7 (0.9) 1.6 (1.4) 3.6 (1) 3.1 (1.6) 2.9 (0.8) 3 (1.7) 2.9 (1.8) 0.9 (0.9) 1.6 (0.6) 2.4 (1.6) 3.4 (0.7) 2.1 (1.2) 3.4 (1.2) 2.2 (1.8) 3.2 (1.5) 2.5 (0.9) 3 (2.1) 3 (1.8) 2.9 (1.9) 4 (1.7) 3.2 (1.4) 4.3 (0.1) 4 (1.4) 3.4 (1.3) 2.2 (1) 3.3 (1.7) 3.1 (1.4) 3.4 (1) 5.4 (1.1) 5 (0.9) 5.2 (1.3) 1.7 (1.1) 2 (0.9) 3.9 (1.4) 3.4 (1.9) 2 (1.3) 2.8 (1) 2.3 (2) 3.1 (1) 3.8 (1.8) 1.8 (0.7) 2.9 (1.3) 2.5 (1.5) 2.2 (1.1) 3.5 (1.3) 2.6 (1.4) 3.5 (1.7) 1.9 (0.9) 2.5 (1.6) 2.1 (1.2) 2.4 (1.1) 3.4 (1.4) 3.2 (1.6) 1.4 (0.3) 2.9 (1.5) than children in traditional and Freinet schools. However, globally, children in alternative schools (Freinet and Montessori) showed better performance than children in traditional schools (V(3,176) = 3.65; p = .01). This difference was more pronounced in the first year than in the second year (respectively V(3,176) = 4.41; p b .001 and V(3,176) = 2.51; p = .06). (3) The testing year (year 1, year 2 of the study) influenced significantly creative performance (R / Rao(3,176) = 3.97 ; p b .01): in general, performance decreased from the first year to the second year. (4) However, there was a significant interaction between school and testing year (R / Rao(9,428) = 5.68 ; p b .001): in Table 4 Multivariate tests of main effects and interactions of school, gender, grade and year on performance scores integrative tasks Table 5 Multivariate tests of main effects and interactions of school, gender, grade and year on performance scores divergent thinking tasks Gender School Grade Year Gender * School Gender * Grade School * Grade Gender * Year School * Year Grade * Year Gender * School * Grade Gender * School * Year Gender * Grade * Year School * Grade * Year Gender * School * Grade * Year R / Rao pb 3.47 5.98 5.82 .85 .57 1.62 1.60 .72 1.68 1.84 .57 .83 1.31 1.26 .92 .03 .001 .001 .43 .75 .14 .06 .49 .12 .08 .92 .55 .25 .21 .56 Gender School Grade Year Gender * School Gender * Grade School * Grade Gender * Year School * Year Grade * Year Gender * School * Grade Gender * School * Year Gender * Grade * Year School * Grade * Year Gender * School * Grade * Year R / Rao pb .94 6.49 2.59 3.97 .72 .45 1.20 .31 5.68 1.02 .78 .89 .83 1.09 1.33 .42 .001 .006 .009 .69 .91 .22 .82 .001 .42 .78 .54 .59 .35 .12 M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 387 Fig. 1. Interaction between type of school and testing year for divergent thinking tasks. the first year, creative performance in the Freinet school was similar to performance in the Montessori school. Children in these alternative schools outperformed those in traditional schools. Nevertheless, in the second year, creative performance of children in the Freinet school was similar to performance in traditional schools (see Fig. 1). In the first year, creative performance in alternative schools showed better performance than in traditional school, notably for verbal divergent thinking tasks. In the second year, creative performance in the Montessori school was higher than creative performance in Freinet and traditional schools. However, the traditional school in Table 6 Percentage of participants, by school, showing change from year 1 to year 2 (scores decrease, remain stable, increase) in each creativity task Cardboard box originality Improvement of toy originality Parallel lines originality Story Drawing Montessori Freinet Traditional 1 Traditional 2 Montessori Freinet Traditional 1 Traditional 2 Montessori Freinet Traditional 1 Traditional 2 Montessori Freinet Traditional 1 Traditional 2 Montessori Freinet Traditional 1 Traditional 2 Decrease Stable Increase 30 54.9 38.8 48.1 52.5 56.9 52.2 44.2 30 56.9 28.9 53.8 30 29.4 29.9 23.1 15 25.5 16.4 25 17.5 17.6 23.9 19.2 2.5 19.6 19.4 15.4 10 9.8 28.9 15.4 45 45.1 55.2 46.2 47.5 45.1 46.3 50 52.5 27.5 37.3 32.7 45 23.5 21.4 40.4 60 33.3 52.2 30.8 25 25.5 14.9 30.7 37.5 29.4 37.3 25 suburban Paris showed good performance for the parallel lines task. To summarize, we observed that children in alternative pedagogy, particularly in the Montessori school, showed generally higher creative performance than children in traditional pedagogy. Additionally, grade level influenced creative performance and girls' creative performances tended to be higher than boys', notably in integrative tasks. Furthermore, the testing year influenced creative performance. Finally, there was an interaction between the type of school and testing year for divergent thinking tasks. Overall, the results suggest that the evolution of creative potential (from year 1 to year 2 of the study) differs according to the learning environment. 2.2.2. Aptitude–treatment interaction and creative performance One of our hypotheses was that children's creative performance in the second year could be influenced not only by children's aptitude and by school environment, and by their interaction. In multiple regression analyses, we entered first the initial level of creativity (year 1) and type of school and second the interaction between initial creative level and type of school. The results of these analyses showed that both initial (year 1) level of creative performance and type of school accounted for variance in year 2 creative performance scores. However, with one exception for the parallel lines task (incremental R2 = .04; p b .17), the interaction term involving type of pedagogy and initial creativity score did not account for additional variance in year 2 scores, thus no strong aptitude by treatment interactions were observed. 2.2.3. Effect of learning environment on creative development We calculated a change score, corresponding to the absolute difference between the score in the second year minus the score in the first year. Then, we classified the children's evolution scores in three categories: decrease, stable (0+ / − 10%) and increase. According to our hypotheses, grade and type of school 388 M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 could influence the evolution of children's creative performance (Table 6). We examined the percentage of children whose performances increased or decreased or remained relatively stable from one year to another and according to the task. In the divergent thinking tasks, we observed that children were less original in the Freinet school (56%) and traditional pedagogies (44%) whereas children in the Montessori school were more original (53%). Finally, for the integrative tasks, children's creative performance was consistent from one year to another in each pedagogical environment. 3. Discussion and conclusion Concerning the first hypothesis of an influence of pedagogy on children's creative performance, our results indicated that, in general, children schooled in alternative pedagogies (Montessori and Freinet) obtained higher performances than children schooled in traditional pedagogy. These findings replicate results obtained by Avanzini and Ferrero (1976–1977) and Frankiewicz (1984). However, our results are not completely congruent with those of Thomas and Berk (1981), which revealed, like Horwitz (1979), mixed findings. In fact, creative performance, according to various studies, could depend on gender, creative task and pedagogy. Thomas and Berk emphasized that these divergences could be due to the variations of school environment in “traditional” and “alternative” type pedagogies. This issue will be discussed below because we observed variations between schools in the current results. Moreover, Thomas and Berk (1981) found in their study a gender effect according to which boys showed better creative performances than girls. They suggested that teachers may tolerate creativity among boys more easily than among girls. Our results do not indicate a major gender effect but, when we observed a gender effect girls showed better creative performance than boys. These differences between girls and boys in creative tasks were found, notably by Petrulytè (2007) who compared the verbal and figural creative performances of children in primary school. She observed that boys showed better performances than girls on flexibility whereas girls outperformed boys on elaboration. In our case, the differences were observed in the integrative tasks which requires elaboration. Concerning the second hypothesis on the positive influence of alternative pedagogy on creative development from year 1 to year 2, we observed that the Montessori program was associated with an overall increase in creativity, for children of all initial creative ability levels but this was not observed for children in the Freinet school. This difference could be explained in terms of the effectiveness of Montessori schools increasing creativity. However the teaching staff varied in the Freinet school, with fewer teachers engaged in the Freinet philosophy of teaching during the second year of study. Thus, some children in year 2 had a teacher who proposed a more traditional pedagogy, in the Freinet school. These variations across the two years of study support the influence of pedagogy and suggest the importance of teachers that implement well the pedagogies. Concerning other developmental trends, ours results highlight that some children increased their creative performance from year 1 to year 2 whereas others showed a decline or stability. It seems that children's performance who advanced from third grade to fourth grade (the second year of the study) tended to increase less than children who moved to second or third grade, except for integrative tasks on which performance was roughly stable for all children. Thus, these results corroborate the literature concerning the presence of a fourth grade slump for divergent thinking tasks (Charles & Runco, 2001; Torrance, 1968). Nevertheless, “grade” and “year of testing” measure partially the same construct: in fact, children are more mature from one grade to another and also from one year of testing and the next. Thus, the results need to be interpreted with caution. The differences observed between divergent thinking tasks and integrative tasks could be due to the fact that integrative tasks involve both divergent thinking as well convergent processes such as defining and redefining the problem, perception and reorganisation of information (Lubart et al., 2003). Moreover, because children do not all belong each year to the same class, peer groups could influence creative performance: children know each other and they establish relationships between one another throughout the year and the peer group plays an important role (Runco, 1999). Thus, differences introduced within the “class composition” could also influence creative performance. Consequently, if a child is in the “group 1–class A” during the first year and in the “group 2–class B” during the second year, there may be social, peer effects of the new set of children interacting together. So it would be necessary in future studies to focus more on social environmental characteristics. In summary, the results presented here indicate a relationship between learning environment and creativity. There is evidence for: (1) The influence of pedagogy. It is possible that alternative pedagogies (such as Montessori and Freinet) use different types of exercises which develop divergent thinking, an important component of creative thinking. Of course, one methodological limit of the current research is that there was only one Montessori school and one Freinet school. Thus the results need to be replicated in future work. (2) The influence of the teacher. Teachers were not the same from one year to another, notably in the Freinet school (sick leave, birth of a child…). Implementation of a pedagogy differs from one teacher to another (Brighelli, 2005; Cropley, 1999; Maker & Nielson, 1995; Smith & Renzulli, 1984). In addition, teachers who participated in the first year were volunteers, perhaps motivated intrinsically. In the second year, some teachers were more extrinsically motivated (hierarchical pressure). (3) The influence of the type of task. There was evidence for a differential effect of school pedagogy as a function of the type of creative task. The pattern of results differed for divergent thinking tasks and integrative thinking tasks, from year 1 to year 2 of the study. Thus, the finding that pedagogy influences creativity must be modulated according to the component of creativity that is examined. References Amabile, T. M. (1996). Creativity in context. Boulder, CO, US: Westview Press. Anastasi, A. (1970). On the formation of psychological traits. American Psychologist, 25, 899−910. M. Besançon, T. Lubart / Learning and Individual Differences 18 (2008) 381–389 Avanzini, G., & Ferrero, M. (1976–1977). Contribution à une comparaison entre les techniques Freinet et les méthodes traditionnelles d'enseignement [A comparison between Freinet and traditional teaching methods]. Bulletin De Psychologie, 30(10–13), 455−467. Baer, J. (1999). Domains of creativity. In M. A. Runco & S.R. Pritzker (Eds.), Encyclopedia of creativity (pp. 591−596). San Diego, California: Academic Press. Brighelli, J. -P. (2005). La fabrique du crétin. [Creating Idiots] Paris: JeanClaude Gawsewitch. Charles, R. E., & Runco, M. A. (2001). Developmental trends in the evaluative and divergent thinking of children. Creativity Research Journal, 13(3–4), 417−437. Cropley, A. J. (1999). Education. In M. A. Runco & S.R. Pritzker (Eds.), Encyclopedia of creativity (pp. 629−642). San Diego, California: Academic Press. Danvers, F. (2003). 500 mots clefs pour l'éducation et la formation. [500 keywords for education and training] Villeneuve D'Ascq: Presses Universitaires du Septentrion. Frankiewicz, W. (1984). Etude de l'influence de la technique du texte libre sur le développement de la pensée créative des enfants [Study of the effect of the free-text technique on development of creative thought in children]. Scientia Paedagogica Experimentalis, 21(2), 200−210. Freinet, C. (1994). Oeuvres pédagogiques. [Educational works]. Tome 1 Lonrai (France): Seuil. Gardner, H. (1996). L'intelligence et l'école: La pensée de l'enfant et les visées de l'enseignement. [Intelligence and school: children's thinking and educational aims] Paris: Retz. Horwitz, R. A. (1979). Psychological effects of the open classroom. Review of Educational Research, 49(1), 71−85. Lubart, T., & Guignard, J. -H. (2004). The generality–specificity of creativity: A multivariate approach. In E. L. Grigorenko & R.J. Sternberg (Eds.), Creativity: From potential to realization (pp. 43−56). Washington, DC, US: American Psychological Association. Lubart, T., Mouchiroud, C., Tordjman, S., & Zenasni, F. (2003). Psychologie de la créativité. [Psychology of creativity] Paris, France: Armand Colin. Maker, J. C. (1996). Identification of gifted minority students: A national problem, needed changes and a promising solution. Gifted Child Quarterly, 40, 41−50. Maker, J. C., & Nielson, A. B. (1995). Teaching/learning models in education of the gifted (2nd ed.). Austin (TX): Pro-Ed. Montessori, M. (1958/2004). Pédagogie scientifique: tome 2, éducation élémentaire. [Scientific pedagogy: volume 2, elementary education] Genève: Desclée de Brouwer. Morandi, F. (1997). Modèles et méthodes en pédagogie. [Models and methods in pedagogy] Paris: Nathan Université. 389 Ochse, R. (1990). Before the gates of excellence: The determinants of creative genius. New-York: Cambridge University Press. Petrulytè, A. (2007). Peculiarities of verbal and non-verbal creativity, learning achievements among gifted children. Discussion in Gifted Children: challenges and Possibilities, Kaunas. Piaget, J. (1969/2004). Psychologie et pédagogie. [Psychology and pedagogy] Paris: Folio Essai. Plucker, J. A. (1998). Beware of simple conclusions: The case for content generality of creativity. Creativity Research Journal, 11(2), 179−182. Runco, M. A. (1999). Developmental trends in creative abilities and potential. In M. A. Runco & S.R. Pritzker (Eds.), Encyclopedia of creativity (pp. 537−540). San Diego, Californie: Academic Press. Schneuwly, B., & Bronckart, J-. P. (1985). Vygostki aujourd'hui. [Vygostky today] Neuchâtel: Delachaux et Niestlé. Simon, J. R. (1960). Change with age in the speed of performance on a dial setting task. Ergonomics, 3, 169−174. Smith, L. H., & Renzulli, J. S. (1984). Learning style preferences: A practical approach for classroom teachers. Theory into Practice, 23(1), 44−50. Snow, R. (1978). Theory and method for research on aptitude processes. Intelligence, 2, 225−278. Snow, R. (1980). Aptitude processes. In R. E. Snow, P. -A. Federico, & W. E. Montague (Eds.), Aptitude, Learning and Instruction, Vol. 1. (pp. 25−63) Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers. Snow, R. (1994). A person–situation interaction theory of intelligence in outline. In A. Demetriou & A. Efklides (Eds.), Intelligence, Mind, and Reasoning: Structure and Development (pp. 11−28). Amsterdam: Elveiser Science. Sternberg, R. J., & Lubart, T. I. (1995). Defying the crowd: Cultivating creativity in a culture of conformity. New York: Free Press. Thomas, N. G., & Berk, L. E. (1981). Effects of school environments on the development of young children's creativity. Child Development, 52(4), 1153−1162. Torrance, E. P. (1976). Les tests de pensée créative. [Tests of Creative Thinking] Paris: Les Editions du Centre de Psychologie Appliquée. Torrance, P. (1968). A longitudinal examination of the fourth grade slump in creativity. Gifted Child Quarterly, 12(4), 195−199. Urban, K. K., & Jellen, H. G. (1996). Test for Creative Thinking–Drawing Production (TCT-DP) Francfort: Swets Test Services. Vygotsky, L. S. (1934/1997). Pensée et langage. [Thinking and language] Paris: La Dispute.