• No results found

Achievement goal profiles : the relationship with innovative learning, effort, and achievement in elementary school students

N/A
N/A
Protected

Academic year: 2021

Share "Achievement goal profiles : the relationship with innovative learning, effort, and achievement in elementary school students"

Copied!
62
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

1 Achievement goal profiles: the relationship with innovative learning, effort, and

achievement in elementary school students

Marieke Majoor (0470848)

Supervisors: Thea Peetsma & Lisette Hornstra

In this study, the relationships between students’ achievement goal profiles and school effort, achievement, and aspects of innovative learning were examined. Seven hundred and twenty-two students reported on their effort in language and math, 68 teachers reported on the students’ general school effort and levels of innovative learning, and achievement was assessed by obtaining test-scores from the school records. Results for both subject domains indicated that theoretically more beneficial achievement goal profiles were associated with higher levels of effort and achievement, while less beneficial profiles were associated with lower levels of effort and achievement. Likewise, transitions to more favorable goal profiles during the year resulted in better educational outcomes. Regarding innovative learning, higher levels of innovative instructional approaches were associated with the adoption of more beneficial goal profiles and the transition to more beneficial goal profiles throughout the school year. The results of this study suggest that the use of aspects of innovative learning might influence the achievement goal profiles that students adopt, which in turn seems to influence their school effort and achievement.

Keywords: achievement goal profiles, effort, academic achievement, innovative learning, development, multiple goal perspective

(2)

2 Introduction

Recently, the Dutch Inspectorate of Education published a study which revealed that the motivation for learning of Dutch students is relatively low compared to students from other countries (Inspectie van het Onderwijs, 2014). Moreover, motivation of many students starts to decline in the last years of elementary school (Hornstra, van der Veen, Peetsma, & Volman, 2013). These findings raise concerns, as motivation for schools plays an important role in explaining several academic outcomes such as school effort and achievement (Eccles & Wigfield, 2002). A major theory in research on school motivation is the achievement goal theory, which focuses on the reasons that students have for engaging in achievement behavior (Dweck, 1986; Nicholls, 1984). These reasons are defined as three different achievement goals students can pursue. Like students’ motivation, elementary students’ achievement goals are found to become less favorable over time as well (Bong, 2009). Over the last decade, researchers started to focus on achievement goal profiles, in which multiple goals are adopted simultaneously (Harackiewicz, Barron, Pintrich, Elliot, & Trash, 2002; Pastor, Barron, Miller, & Davis, 2007, Pintrich, 2000b). But although the relationships between the three separate goals and academic achievement have been well studied, it is still debatable which

combination of goals leads to the most adaptive pattern of school effort and achievement. Furthermore, studies have shown that achievement goal profiles are not stable (Jansen in de Wal, Hornstra, Prins, Peetsma, & van der Veen, 2015; Schwinger & Wild, 2012), which indicates that factors of the learning environment might influence students’ achievement goals. Possible factors that are assumed to effectively foster the motivation of students are innovative instructional approaches (e.g. Nichols & Miller, 1994; Nie & Lau, 2010), but little research has been dedicated to the effects of innovative learning on students’ achievement goal profiles. Therefore, the purpose of this study is to extent the knowledge on the

(3)

3 relationship between innovative learning and achievement goal profiles on the one hand, and students’ achievement goal profiles and school effort and achievement on the other.

Achievement Goal Theory

For over 20 years, achievement goals have been an important construct in research on students’ motivation for school. In the mid-1980’s, the achievement goal theory was

developed to gain insight in the adaptive and maladaptive responses of students to

achievement challenges (Dweck, 1986; Nicholls, 1984). In this theory, goal orientations are interpreted as the reasons and intentions that students have for engaging in achievement tasks (Pintrich, 2003). Goal theorists assume that goals are conscious cognitive representations, and that behavior of students is directed toward the attainment of certain goals (Pintrich, 2000a). Therefore, in educational research, achievement goal theorists focus on the purposes and reasons of students for engaging in, choosing and persisting at various learning activities.

In early research on achievement goals, a differentiation was made between two primary goals: mastery goals and performance goals (Ames & Archer, 1988; Dweck, 1986; Nicholls, 1984). Students pursuing mastery approach goals focus on learning, developing competence and understanding a task. Students endorsing performance goals try to show competence by outperforming peers. So, students adopting the former goals use self-improvement as a reference point, while students adopting the latter goals use social or normative standards (Pintrich, Conley, & Kempler, 2003). Nowadays, however, a distinction is made between performance approach and performance avoidance goals, thereby extending the dichotomous model of achievement goals into a trichotomous achievement goal

framework (Elliot, 1999; Elliot & Harackiewicz, 1996). In this distinction, performance approach goals represent the focus of students to appear able relative to others. This way, students can be positively motivated to demonstrate competence. Performance avoidance

(4)

4 goals have a negative connotation, as students endorsing these goals try to avoid failure or appearing worse than others.

Achievement Goals and Educational Outcomes

It has long been thought that mastery goals promote greater educational benefits than performance goals (Dweck, 1986). Findings have supported this claim for various educational outcomes. Students who adopt mastery goals use learning strategies that enhance recall of information and conceptual understanding, perceive tasks as valuable, and show higher persistence when faced with difficulties (e.g. Elliot & Dweck, 1988; Elliot & McGregor, 2001; Grant & Dweck, 2003; Green & Miller, 1996; Meece & Miller, 2001; Wolters, 2004). Furthermore, studies showed that mastery goals facilitate intrinsic motivation and interest (Elliot & Church, 1997; Harackiewicz, Barron, Pintrich, Elliot, & Trash, 2002). A relationship with effort has been found as well (Gonida, Voulala, & Kiosseoglou, 2009; Miller, Greene, Montalvo, Ravindran, & Nichols, 1996). However, many studies found no relationship between mastery goals and academic achievement, indicating that students pursuing mastery goals do not perform better than students who do not pursue these goals (Elliot, McGregor, & Gable, 1999; Harackiewicz, Barron, Carter, Lehto, & Elliot, 1997; Harackiewicz, Barron, Tauer, Carter, & Elliot, 2000).

Contrary to initial views that mastery goals are superior to performance goals,

performance goals do sometimes yield more beneficial outcomes compared to mastery goals. Several studies have shown that performance approach goals are positively related to adaptive outcomes such as academic self concept, task value, and effort expenditure (e.g. Bong, 2001; Church, Elliot, & Gable, 2001; Skaalvik, 1997; Elliot et al., 1999). Furthermore, in the same study in which Harackiewicz, et al. (2002) found a main effect of mastery goals on interest, they also found significant main effects of performance approach goals on academic

(5)

5 performance. In several studies, performance approach goals were the best predictors of grades in various college courses (e.g. Church et al., 2001; Elliot & McGregor, 1999; Harackiewicz et al., 2000). So, while mastery approach goals seem to facilitate intrinsic motivation, performance approach goals seem to enhance grade performance (Elliot & Church, 1997). On the other hand, less beneficial educational outcomes of performance approach goals have been reported as well. For instance, performance approach goals have been associated with surface-level learning strategies, which means that students who adopt these goals learn by memorizing and producing the factual content of the learning material (Elliot & Harackiewicz, 1996; Graham & Golan, 1991).

The relationship between performance avoidance goals and educational outcomes has been more straightforward, as these goals have generally been linked to maladaptive

outcomes such as test anxiety, self-handicapping, and low performance on tests (Elliot, 1999; Elliot & Church, 1997; Elliot & Harakiewicz, 1996). Contrary to mastery approach goals, performance avoidance goals have been found to have a negative relation with intrinsic motivation. Likewise, while performance approach goals seem to enhance performance, a negative relation exists between achievement and performance avoidance goals (Church et al, 2001; Darnon, Butera, Mugny, Quiamzade, & Hulleman, 2009; Elliot & Church, 1997; Sideridis, 2005). An explanation for these negative outcomes on academic performance might be that performance avoidance goals have been related to task distraction and low working memory (Huijun, Dejun, Hongli, Peixia, 2006; Senko & Miles, 2008), which are both affected by feelings of anxiety and worry. Nevertheless, not all studies find negative effects of

performance avoidance goals. While examining the development in students’ performance avoidance orientation, Peetsma and van der Veen (2013) did not find more negative developments in learning behavior and academic achievement for students with a less

(6)

6 favorable (high or increasing) development in performance avoidance orientation than

students with different developmental trajectories.

The Multiple Goal Perspective and Achievement Goal Profiles

Originally, it was assumed that students primarily adopt one goal. Students were considered to be either mastery oriented or performance oriented (Dweck, 1986; Nicholls, 1984). However, several studies showed that mastery goals and performance goals are either unrelated or positively correlated and that students can therefore pursue multiple goals simultaneously (e.g. Meece & Holt, 1993; Pintrich, 2000b). Yet, the question of which

combination of goals is the most beneficial is still under debate. Some researchers have found that adopting both mastery goals and performance approach goals resulted in the most

favorable outcomes (Bouffard, Vezeau, & Bordeleau, 1998; Harackiewicz, Barron, Tauer, & Elliot, 2002; Luo, Paris, Hogan, & Luo, 2011; van der Veen & Peetsma, 2009). Harackiewicz et al. (2002) explained this finding by stating that mastery goals promoted motivation by fostering interest in courses or tasks, while performance goals promoted motivation by fostering achievement. So, mastery approach goals and performance approach goals simultaneously produced independent positive main effects on different outcomes. Similar results were found by Luo et al. (2011), who found that students with goal profiles with high mastery and performance approach goals and low performance avoidance goals showed greater effort in school and achieved the highest scores in mathematics. Students who had moderate scores on all three achievement goals, or moderate scores on mastery goals and high scores on both performance goals achieved the least beneficial outcomes. So, these studies suggest that a goal profile with high scores on both mastery and performance approach goals and low scores on performance avoidance goals is associated with the most adaptive pattern of effort and achievement.

(7)

7 However, the results of Luo et al. (2011) also showed that students with a moderate score on mastery approach goals and a low score on both performance goals belonged to the group of highest achievers in math as well. This finding is more in line with studies in which it was demonstrated that higher levels of mastery goals combined with lower levels of

performance goals result in the best outcomes with regard to school effort and achievement. The study of Ng (2006) showed that a mastery-focused profile, with high levels on mastery goals and low levels on performance goals, was associated with higher levels of engagement. In addition, Meece and Holt (1993) found that students who scored high on mastery goals but low on the other goals had higher grades and test scores than students with other profiles. Yet, in another study, no differences in achievement were found between students who

predominantly endorsed mastery goals and students who endorsed both mastery and

performance goals (Valle, Cabanach, Nunez, Gonzalez-Pienda, Rodriguez & Pineiro, 2003), thereby showing that there is no consensus on which achievement goal profile leads to the most adaptive outcomes with regard to academic achievement and effort.

There are a few reasons why it is difficult to draw conclusions based on the studies described above. First, the goal profiles were created with different types of achievement goals. Some profiles only consisted of mastery (in some studies referred to as learning goals) and performance goals (Bouffard, Vezeau, & Bordeleau, 1998), while in other studies a distinction was made between performance approach and performance avoidance goals (Luo et al., 2011). In other studies (Meece & Holt, 1993; Ng, 2006), performance avoidance goals were replaced with work avoidance goals (which means that students try to complete their school work with the least amount of effort). Also, in some studies, different achievement goals such as social reinforcement goals (Valle et al., 2003) or ego-social goals (Meece & Holt, 1993) were used. Furthermore, the participants consisted of different age groups. Only one study was conducted with elementary school students (Meece & Holt, 1993). Also, few

(8)

8 studies used longitudinal designs. An even greater limitation is that most studies made use of analytical methods such as median split procedures and cluster analyses, while more

sophisticated methods to create achievement goal profiles, such as latent profile analysis, are available (see Pastor, Barron, Miller, & Davis, 2007). Advantages of latent profile analyses are that the number of profiles are determined with more stringent statistical criteria, and that information is available about the accuracy of the classifications (Muthén & Muthén, 2000; Vermunt & Magidson, 2002).

Three studies in which latent profile analyses were conducted and in which the relationships between goal profiles and achievement and/or effort were studied, were identified. The first study in which the relationship between achievement goal profiles and achievement was examined is the study of Tuominen-Soini, Salmela-Aro, and Niemivirta (2008). In this study, four goal profiles were identified among secondary school students. One profile was found to be superior to the others with regard to achievement. Students with a profile with high mastery goals, high performance approach goals, and moderate performance avoidance goals had the highest achievement scores, whereas students with other profiles did not differ in academic achievement.

Schwinger and Wild (2012) also examined the relationship between achievement goal profiles and academic achievement. Contrary to the study of Tuominen-Soiniet et al. (2008), this study distinguished between three achievement goal profiles among elementary school students, all of which were relatively high on mastery goals. In addition, Schwinger and Wild (2012) also investigated the longitudinal effects of goal profiles. Regarding these longitudinal effects, no clear patterns between goal profiles and achievement scores were found, so none of the three profiles turned out to be consistently superior to the others. However, in some school years, it was found that students who highly pursued multiple goals (high on both performance goals, and relatively high on mastery approach goals) had lower test scores

(9)

9 and/or school grades than students who moderately pursued multiple goals (moderately high on all three goals) and students who were identified as primarily mastery-oriented

(moderately high on mastery goals, low on both performance goals). Therefore, it seems that a high score on both performance goals leads to disadvantageous academic results. However, given the negative relationship between performance-avoidance goals and academic

achievement, it also seems plausible that the relatively high score on this goal mostly influenced the test results.

Schwinger and Wild (2012) also examined the relationship between achievement goal profiles and school effort. Both cross-sectional as longitudinal, they did not find any effect of goal profiles on effort. In contrast, Tuominen-Soini, Salmela-Aro and Niemivirta (2012) did find effects of achievement goal profiles on school engagement among upper secondary school students. High mastery/low performance (approach and avoidance) oriented students and high mastery/high performance (approach and avoidance) oriented students reported the highest levels of school effort, and low mastery/high performance avoidance oriented students and students who scored equally low on all achievement goals reported the lowest levels of effort. The authors also examined the longitudinal profile effects on engagement. Students who retained a high mastery/low performance oriented profile or a high mastery/high performance oriented profile, and students who made a transition to a more adaptive profile reported more school engagement than students who made a maladaptive transition or retained a profile with equally low scores on all achievement goals.

The studies described above show some promising results with regard to associations between goal profiles and developments in academic outcomes. However, the results vary considerably between the studies. Furthermore, only one study investigated the effects of goal profiles among elementary school students, even though research has shown that already in elementary school, students’ achievement goals become less favorable over time (Bong,

(10)

10 2009). Thus, transitions in achievement goal profile and subsequently in academic

achievement and school effort might already be visible during these years, as is partly shown by Schwinger and Wild (2012). Moreover, Schwinger and Wild (2012) and Tuominen-Soini et al. (2012) seem to be the only researchers that have focused on the longitudinal

relationships between achievement goal profiles and academic achievement and/or school engagement, so more research is necessary to gain knowledge on the extent to which growth in effort and achievement is associated with transitions in goal profiles.

Innovative Learning

Besides looking at individual differences in achievement goals and their association with achievement and effort, it is also important to look at the relationship between school context and achievement goal profiles. Pintrich (2000a) mentioned that achievement goals may not just reflect individual stability, but that these goals could also be sensitive to

contextual characteristics. This was supported by the study of Jansen in de Wal et al. (2015), who reported that transitions in students’ goal profiles occurred most frequently across school years, when environmental changes are bigger and students often get a different teacher. Similar results were obtained by Schwinger and Wild (2012), who found that across school years, two thirds of the students switched goal profiles at least one time during the five years period of study. Thus, as studies indicate that goal profiles are subject to change, research is necessary to find out which factors influence these transitions.

An approach that has been highlighted over the last decade is innovative learning. Innovative learning is based on the principles of socio-constructivism. These principles state that learning is a collaborative and active process in which students construct knowledge and meaning (Loyens & Gijbels, 2008; Wilson, 2012; Windschitl, 2002). Therefore, in innovative learning environments, students are more responsible for their own learning process and are

(11)

11 more actively involved in their learning activities (Bolhuis & Voeten, 2001). The role of the teacher is to activate students and to guide them in their learning process. This approach contrasts with a traditional approach, in which the role of the teacher is merely to transfer knowledge (Mascolo, 2009).

Three important characteristics of innovative learning are collaborative learning, authentic learning and self-regulated learning. Collaborative learning is a method of instruction in which students work together in small groups to attain a common goal. For collaborative learning to be successful, it is necessary that groups are structured in a way that students know what is expected of them and that each member has to contribute to the group task (Johnson & Johnson, 2009). Furthermore, students must have individual accountability for mastering the material. Researchers found that cooperative learning has many positive effects. When teachers adopt a collaborative instructional approach social relations are improved (Tolmie et al., 2010) and deep learning of the material is promoted (Shimazoe & Aldrich, 2010). With regard to performance, studies have shown that cooperative learning leads to better achievement and greater productivity than individual learning (Johnson & Johnson, 1999, Nattiv, 1994; Nichols, 1996; Nichols & Miller, 1994).

A second characteristic of innovative learning is authentic learning. Authentic learning refers to an instructional approach that focuses on connecting the learning content to the students’ real world outside school. The idea is that if topics are discussed that match their interests and are relevant and applicable to their daily life outside school, students are more willing to learn and better prepared to use their knowledge in other contexts (Jonassen & Rohrer-Murphy, 1999; Van Oers & Wardekker, 1999). It has been found that a focus on authentic learning enhances the self-efficacy of students (Nie & Lau, 2010).

Another element of innovative learning is the focus on self-regulated learning. Self-regulated learning can be defined as the competence of a student to plan, execute, monitor and

(12)

12 evaluate a learning process. During this process students must continuously regulate their cognitions, motivation and behavior (Wirth and Leutner, 2008). Self-regulated learners are able to take control over their own learning process by effectively using different learning strategies, being learning-oriented, and pursuing feasible goals (Zimmerman, 2002).

Innovative learning helps students to improve their self-regulatory skills by not only focusing on the learning outcomes, but also on the process of learning (Boekaerts, 1997; Bolhuis, 2003). Interventions that improve self-regulatory skills have been shown to enhance students’ academic performance (Dignath & Büttner, 2008; Dignath, Buttner, & Langfeldt, 2008; Masui & De Corte, 2005).

Because of the belief that innovative learning enhances students’ motivation, several aspects of innovative learning have been implemented in educational practice (Oostdam Peetsma, & Blok, 2007). Nevertheless, only a few studies have examined the relationships between these instructional approaches and aspects of student motivation. Lau (2012) showed that upper secondary school students had higher levels of motivation when they were in a classroom where authentic learning and self-regulated learning were emphasized. Nie and Lau (2010) found the same effects for authentic learning when they compared motivational beliefs of secondary school students who received more innovative or more traditional instructions in English class. Guthrie, Wigfield, and VonSecker (2000) also compared the effects of

traditional reading instructions and innovative instructions, and found that an emphasis on authentic and collaborative learning increased the interest of elementary students in reading. More support for the positive association between collaborative learning and motivation was obtained by Hänze and Berger (2007), who showed that students in collaborative learning environments reported higher levels of intrinsic motivation than students who received a more traditional instruction. Thus, all three aspects of instruction seem to be related to enhanced motivation in general.

(13)

13 Even less studies could be found in which the relationship between aspects of

innovative learning and students’ achievement goals was examined. Nichols and Miller (1994) compared secondary school students in more cooperative learning environments with students in more traditional learning environments, and found that the former group of students were more mastery-oriented than the latter group. In addition, when students in the cooperative environment started to receive traditional instructions, they reported lower levels of mastery orientation. Similar results were obtained by Nichols (1996), who also showed that mastery-orientations increased for students who received cooperative learning. In contrast, Thoonen, Sleegers, Peetsma, and Oort (2011) did not find an effect of collaborative learning on mastery goals in a group of primary school students. In addition, neither did they find a relationship between teacher focus on self-regulated learning and mastery goals. However, an effect was found for authentic learning, as students in classrooms with higher levels of authentic learning reported higher levels of mastery-oriented goals.

Thus, although a lot of research has been dedicated to achievement goals, not many researchers have looked at the relationship between achievement goals and several aspects of innovative learning. Furthermore, despite the developments in achievement goal theory and the evidence that students can pursue several achievement goals at the same time, there seem to be no studies in which the association between achievement goal profiles and contextual variables such as innovative learning were examined. More empirical research is needed to examine whether innovative instructional approaches might be related to the adoption of a certain achievement goal profile, and, as several studies have showed that goal profiles can change, whether these instructional approaches are related to transitions to a more or less beneficial goal profile.

(14)

14 A major purpose of this study is to extend the work on achievement goals by relating elementary students’ achievement goal profiles to innovative instructional approaches, academic achievement and school effort. It will be examined if there is a relation between the extent to which schools adopt a more innovative approach or a more traditional approach and students’ adoption of a certain goal profile. In addition, the relation between innovative instructional approaches and transitions in students’ goal profiles throughout the year will be examined. In the analyses, a distinction will be made between three characteristics of

innovative learning: collaborative learning, authentic learning and focus on self-regulated learning. Furthermore, both the cross-sectional as the longitudinal relationship between achievement goal profiles and academic achievement will be examined, as well as the cross-sectional and longitudinal relationships between achievement goals profiles and school effort.

Specifically, the following research questions will be addressed:

1. How do students with different achievement goal profiles differ in effort and achievement in language and mathematics?

2. How are transitions in students’ achievement goal profiles related to developments in effort and achievement in language and mathematics?

3. How are the three different aspects of innovative learning related to the adoption of certain achievement goal profiles and transitions in students’ achievement goal profiles?

(15)

15 Method

Procedure

For the present study, four waves of data were collected from students and teachers in grade five to six. Data about achievements on math and reading comprehension were obtained at the middle of both the fifth and sixth grade. Measurements for achievement goals took place halfway through fifth grade, and at the beginning and halfway through sixth grade. In regular classroom conditions, students filled in self-report questionnaires under supervision of a research assistant and their teacher. Teachers filled out questionnaires on each student as well. Data on innovative learning was collected at the beginning of both the fifth and sixth grade by means of teacher report.

Participants

Students. The sample consisted of 722 primary school students from 37 classes of 25 schools across the Netherlands. Three-hunderd-and-sixty-one participants were boys (50%). At the first measurement of the students halfway through fifth grade, the participants were between eight and twelve years old (M=10.64, SD=0.46). With regard to ethnicity, 87.5% of the students were of Dutch and western origin and 12.5% were identified as non-western immigrants. In addition, information on parental education level served as a measure for economic status. Although family income and occupation are also components of socio-economic status (Duncan, Featherman, & Duncan, 1972), parental education can serve as a proxy for SES, as it gives a good indication of family income (Sirin, 2005). Furthermore, it is a relatively stable component compared to occupation and family income. Based on the information on the highest educational level obtained by one of the parents of the students, three groups were distinguished. Parents who completed lower vocational education were classified as having a low educational level (13.3%). Parents who completed intermediate

(16)

16 vocational education were classified as having an average educational level (41.7%). Parents who completed higher education were classified as having a high educational level (28.3%). From 16.7% of the students, no information was available on the parents’ educational level. Teachers. A total of 68 teachers participated in this study. Thirty-seven were grade five

teachers and 31 were grade six teachers. The higher number of grade five teachers were due to two grade six teachers not filling in the questionnaire, students in three classes continuing with their grade five teacher in grade six, and in one school, the grade six group consisted of a combination of two former grade five groups. All teachers were of Dutch ethnicity. On

average, the grade five teachers were 37 years old, and those of grade six were 41 years old. Of the grade five teachers, 68% were women, and of the grade six teachers, 63% were women.

Instruments

Achievement profiles. The Goal Orientation Questionnaire (Seegers, Van Putten, & De

Brabander, 2002) was used to measure achievement goals for both mathematics and language. This questionnaire consists of a total of 17 items with a five-point Likert scale. Five items measure mastery approach orientation (e.g., ‘I feel satisfied when I have learned something in mathematics that makes sense to me’), six items measure performance avoidance orientation (e.g., ‘During mathematics tasks I am afraid that other students will notice my mistakes’), and five items measure performance approach orientation (e.g., ‘I enjoy getting a better grade in mathematics than my classmates’). Equivalent items were used to measure goal orientations for language. Reliability was good for all scales, with values of Cronbach’s alpha ranging from .84 to .94. For both mathematics and language, a confirmatory factor analysis was used to inspect construct validity. A model in which each subscale of the Goal Orientation

(17)

17 450.94, p<.001; CFI = .95; TLI = .94; RMSEA = .07; SRMR = .05 , as well as for language, χ² (116) = 379.34, p<0.001; CFI = .93; TLI = .92; RMSEA = .06; SRMR = .06. Also, measurement invariance was assessed, and it was found that the constructs were invariant over time and across groups.

The goal profiles were created in Mplus 6.1 by performing cross sectional latent profile analyses (for full report of these analyses, see Jansen in de Wal et al., 2015). For each domain in each data wave, solutions ranging from one to six profiles were investigated. A three profile solution turned out to be the best representative for the sample for both mathematics and language. The standardized achievement goals scores for each profile on language and mathematics at the three measurement occasions are displayed in figure 1. The first profile, which is labeled ‘multiple goals’ (profile A), is characterized by a similar score on all achievement goals. Performance avoidance goals of these students are relatively high compared to other students, while performance approach goals are slightly above average and mastery goals are slightly below average. Students in the second profile, which is labeled ‘approach oriented’ (profile B), have relatively high mastery goals and low performance avoidance goals. The third profile is labeled ‘moderate/indifferent’ (Profile C). Students in this profile have average scores on each of the three goals.

Non-stationary latent transition analyses indicated that six patterns occurred most frequently. Other patterns occurred very rarely. These transitions are presented in table 1. For both mathematics and language profile B was most stable, while profile A emptied and profile C filled up over time. A majority of 78.12% of the students had a stable profile during the study for language. For mathematics, 85.22% did not transition profiles during the three measurements. For both domains, the transition pattern from profile A to profile C (A → C → C) was the most prevalent transition pattern (6.58% and 4.40% for language and mathematics respectively).

(18)

18 [Figure 1, Table 1]

Math achievement. Scores of students on national tests from the CITO were used as measures for math achievement. These tests are administered twice a year to follow the progress of the students. For this study, the scores of the tests that took place halfway through both grade five and grade six were used. These scores were obtained from the school records. As the test was updated in 2007, two different versions were used by the schools. Six schools administered the older version, while 18 schools used the updated version. To make the scores comparable, the mean and standard deviations of the scores of the old version were transformed. As a result, the mean and standard deviations of the scores were the same for the old and the new version of the test. In one school, students (N=30) did not take the CITO test. The reliability of the CITO test is found to be good (α > 0.80) (Evers, 2002; Feenstra, Kamphuis, Kleintjes, & Krom, 2010).

Reading comprehension achievement. Contrary to the measurements of achievement goals and student reported effort, the measurements of achievement in language comprised scores on reading comprehension tests, which is a component of language. As with the math

achievement, reading comprehension achievement was measured by means of the CITO tests. The reading comprehension tests are administered once a year, halfway through the academic year. The students’ scores on the tests of grade five and six were also obtained from the school records. A new version of the reading comprehension test was issued in 2008.

Therefore, some schools (N=16) administered the older version to their students, while some schools (N=8) used the updated version. In contrast to the mathematics test, no

(19)

19 and therefore comparable. The reliability of both versions is found to be good (α > 0.80) (Evers, 2002; Feenstra et al., 2010).

Teacher-reported effort. To assess the extent to which students showed effort in school according to their teachers, a scale consisting of three items was filled in by the teachers halfway through both fifth and sixth grade (Jungbludt, Peetsma, & Roeleveld, 1996). The scale measured school effort in general (e.g. ‘This student quickly gives up when he/she does not succeed’), and was not filled in separately for mathematics and reading comprehension or language. Values of Cronbach’s alpha ranged from .82 to .85.

Self-reported effort. A questionnaire by Roede (1989) was used to measure student investment in language and mathematics. The questionnaire comprised 14 items, seven to assess effort in mathematics (e.g. ‘During class, I like working on a mathematics task’) and seven to assess effort in language (e.g. ‘During class, I like working on a language task’). Students filled in the questionnaire halfway through fifth grade and halfway through sixth grade. The reliability of the questionnaire ranged from .77 to .83.

Innovative learning. Teacher questionnaires were administered to measure the extent to which the learning environment of the students could be viewed as innovative. Teachers filled in the questionnaires at the beginning of the fifth and the sixth grade. The questionnaires consisted of three scales that measured collaborative learning, authentic learning, and focus on self-regulated learning. Collaborative learning was measured by a scale of Thoonen et al. (2011). This scale, which consisted of five items, assessed the extent to which teachers stimulated collaborative learning (e.g., ‘I invest effort in good group assignments’). A scale of Thoonen et al. (2011) was also used to measure authentic learning. This scale consisted of four items (e.g., ‘I choose samples that relate to students’). The focus of teachers on self-regulated learning was assessed by a scale consisting of nine items (e.g., ‘I give students freedom to plan their work’), of which five items were by Thoonen et al. (2011) and four items were from

(20)

20 the Questionnaire on Instructional behavior (Lamberigts & Bergen, 2000). Teachers rated the extent to which they agreed with these items on a five-point Likert scale. Internal

consistencies indicated satisfactory to good reliability for all scales. Values of Cronbach’s alpha for the collaborative learning scale were .63 for the first measurement and .77 for the second measurement. For the authentic learning scale, Cronbach’s alpha was .76 for the first and .74 for the second measurement. For the self-regulated learning scale, the values of Cronbach’s alpha were .75 on the first measurement and .71 on the second measurement. Factor loadings of all items were over .40 on the scale they represented, thereby supporting the underlying factor structure of the questionnaire (Hornstra, van der Veen, Peetsma, & Volman, 2014).

To validate this innovative learning questionnaire and check for bias due to teacher ideals or self-serving strategies, lessons were observed at three schools, and at nine schools interviews were taken with grade five teachers. The extent to which the teachers rated their learning environment as innovative differed among these schools. A relationship was found between the teachers’ responses in the interviews and the answers on the self-reports, and the teaching behavior that was observed during the lessons (Koomen, Hornstra, Peetsma, & van der Veen, 2011). Also, interviews were taken with nine grade six teachers and 45 students a year later. A significant positive relationship was found between the level of innovativeness that was reported by the teachers and students (Hornstra et al., 2014).

Cognitive/verbal ability To control for the influence of cognitive ability on achievement scores, cognitive ability was included in the analyses as a control variable. A cognitive ability test was taken in grade three, and it consisted of 85 verbal and non-verbal items. There were two verbal subtests, ‘categories’ and ‘analogies’, that were used as control variables for achievement in reading comprehension. Three non-verbal subtests were used as control variables for math achievement: ‘composition of figures’, ‘exclusion’, and ‘number series’.

(21)

21 Factor analyses showed that the five subtests form one general cognitive ability factor. The test has a reliability of 0.91 (Van Batenburg & Van der Werf, 2004).

Data-Analyses

The initial dataset was checked for outliers. Values that were more than two standard deviations from the mean were examined, and extreme values that were not consistent with other values of the same case were removed. Furthermore, many participants had missing values on one or more of the variables. It was found that in the majority of cases, these

missing values were due to students’ individual absence on measurement occasions or the fact that classes did not participate in certain measurements. It was concluded that the values were Missing At Random. Therefore, the participants were not removed from the analyses. Instead, a multiple imputation procedure was executed in SPSS to estimate the missing values. As the imputations on the categorical variables with regard to the achievement goal profiles could not be estimated, the procedure was only done for the continuous variables. In the analyses, cases with missing values on variables regarding achievement goal profiles were left out, resulting in fluctuating numbers of participants (in all analyses, a minimum of 90% of the participants was included) on the different analyses.

Furthermore, the multinomial logistic regression analyses that needed to be performed to answer the third research question could not be done with the imputed dataset. Therefore, an aggregated dataset was generated out of the datasets that were created with the multiple imputation procedure. To check whether this proceeding was appropriate, several preliminary analyses were performed on both the imputed and the aggregated dataset. Results showed that differences in parameters were negligible, which led to the conclusion that using the

aggregated dataset to answer the third research question would not significantly influence the results.

(22)

22 Before conducting the analyses, dummy variables were created for the categorical explanatory variables. For the achievement goal profiles of both language and mathematics, the moderate/indifferent profile (profile C) was chosen as the first reference category. To be able to compare the multiple goals profile (profile A) with the approach oriented profile (profile B) as well, other dummy variables were created with the multiple goals profile as reference category. This way, all three profiles could be compared with each other. For the transition patterns, it was decided to compare patterns of students who started the school year with the same achievement goal profile but who had a different profile after the year with students who retained the same goal profile throughout the school year. Therefore, three different patterns were taken as reference category (C → C → C, A → A → A, B → B → B). Also, the scores on the continuous explanatory variables were centered to the overall mean, making the interpretations of the results more straightforward.

To assess the relation between achievement goal profiles and achievement in mathematics and reading comprehension, multilevel analyses containing two levels were performed on both mathematics and reading comprehension achievement scores. With multilevel analyses, it is possible to correct for the hierarchical structure of the data. In this study, first level units were students and second level units were classes. With regard to achievement, four separate analyses were performed; a cross-sectional and longitudinal analysis on the relationship between students’ achievement goal profiles in mathematics and achievement scores in mathematics, and similar analyses on the relationship between

students’ achievement goal profiles in language and achievement scores in reading

comprehension. In each analysis, gender, ethnicity, SES and cognitive ability were controlled for. Similar multilevel analyses were performed to assess the relationship between students’ achievement goal profiles and self-reported effort in language and math and teacher-reported effort.

(23)

23 To answer the research question regarding the relation between aspects of innovative learning and achievement goal profiles, four multinomial logistic regressions were performed with the Generalized Linear Mixed Model in SPSS. In these analyses, it is possible to deal with categorical variables with multiple levels, as well as with the hierarchical structure of the data. First, the cross-sectional relationship between the three aspects of innovative learning (collaborative learning, focus on self-regulated learning and authentic learning) and student’s achievement goal profiles in language were examined. The same analysis was performed for students’ achievement goal profiles in mathematics. Since the dependent variables were categorical variables, reference categories were specified during the development of the model. As the group of students with achievement goal profile C was the largest, this profile was chosen as the reference category in both analyses. To be able to also examine the differences between profile A and profile B, a second analysis was performed with the latest as reference category.

Second, the longitudinal relationships between aspects of innovative learning and achievement goal profiles in language and mathematics were examined. The largest group of students was chosen as the first reference category, which was the group of students with pattern C → C → C. In addition, to check for differences between students who started the year with a similar profile, but ended the school year with a different profile, other analyses were performed with different reference categories (A → A → A, B → B → B).

Model Selection

For each of the analyses, a series of models were estimated. To check for dependency between the student level and the classroom level, empty models with only the dependent variable included were created and compared to models without a random slope. Intraclass correlations of each variable are reported in table 2. All analyses had a significant deviance

(24)

24 test, indicating significant variance of the dependent variables at the classroom level, and that multilevel analyses were required.

For the cross-sectional analyses, basic models with the control variables as fixed effects were estimated (model 1). For the longitudinal analyses, growth models with only the achievement scores of grade five as predictor were estimated (model 1). Next, additional models were created with all control variables (model 2). The last step for all analyses was to add the achievement goal profiles as predictors (final model). During each step, all models showed a significantly better fit than the previous models. The significance of the coefficients for the relation between the independent and dependent variables was tested using Wald tests (z tests). The set level of significance was 5%.

Since the outcome variables in the second research question were categorical, test-statistics describing the fit of the model were not accurate (Hox, 2010). Multilevel models with categorical outcomes are estimated using quasimaximum likelihood techniques, therefore, no reliable information is available on the fit of final models. The significance of the coefficients for the relation between the independent and dependent variables was tested using Wald tests (z tests). The set level of significance was 5%.

(25)

25 Results

Descriptive Statistics

Table 2 shows the descriptive statistics of achievement in math and reading comprehension, students’ self-reported effort in language and math, teacher-reported school effort, and the scores on innovative learning for each measurement.

[Table 2]

Relationships between students’ achievement goal profiles in language and effort and

achievement in language/reading comprehension in grade five

The mean scores for each achievement goal profile in language on teacher-reported school effort, self-reported effort in language and achievement in reading comprehension in fifth grade are displayed in figures 2 and 3. In table 3, the effects of students’ achievement goal profiles in language on teacher-reported school effort, students’ self-reported effort in language and achievement in reading comprehension are reported. With regard to teacher-reported school effort, students with a multiple goals profile in language scored significantly lower on teacher-reported school effort than students a moderate/indifferent profile (b = -.440, p = .000). Also, teacher reports of effort for students with an approach oriented profile were significantly higher than for students with a multiple goals profile (b = .459, p = .000). No differences in teacher-reported school effort were found between students with a

moderate/indifferent profile in language and students with an approach oriented profile. With regard to self-reported effort in language, students with a multiple goals profile scored significantly lower than students with a moderate/indifferent profile (b = -.155, p = .007). Students with an approach oriented profile reported significantly more effort in

(26)

26 language than students with either a moderate/indifferent goal profile (b = .148, p = .001) or a multiple goals profile (b = .303, p = .000).

The effects of students’ language achievement goal profiles on achievement in reading comprehension are in line with the effects found on teacher-reported school effort. Students with an approach oriented profile scored significantly higher on achievement in reading comprehension than students with a multiple goals profile (b = 5.842, p = .000), while the latter group also scored significantly lower than students with a moderate/indifferent goal profile (b = -4.259, p = .001). No differences in reading comprehension achievement were found between students with an approach oriented goal profile and a moderate/indifferent goal profile.

[Figure 2, Figure 3, Table 3]

Cross-sectional relationships between students’ mathematics achievement goal profiles

and effort and achievement in math in grade five

The mean scores for each math achievement goal profile on teacher-reported school effort, self-reported effort and achievement in mathematics in fifth grade are displayed in figures 4 and 5. The estimates of the cross-sectional relationship between students’

achievement goal profiles in math and teacher-reported school effort, students’ self-reported effort and achievement in math are displayed in table 4. Teacher reports of effort for students with a multiple goals profile in math were significantly lower than for students with a

moderate/indifferent profile (b = -.287, p = .007). Students with an approach oriented profile also scored higher on teacher-reported school effort than students with a multiple goals profile (b = .338, p = .002). Students with a moderate/indifferent profile in math did not significantly differ in teacher-reported school effort from students with an approach oriented profile.

(27)

27 Compared to students with a moderate/indifferent profile, students with a multiple goals profile reported significantly more effort in math (b = .260, p = .001). Accordingly, students with an approach oriented profile also reported significantly more effort than students with a moderate/indifferent profile (b = .400, p = .000). No differences in

self-reported effort in math were found between students with a multiple goals profile and students with an approach oriented profile.

Regarding achievement, students with a moderate/indifferent profile performed significantly better on mathematics than students with an approach oriented profile (b = 2.058, p = .021). No differences were found between students with a multiple goals goal profile and students with either an approach oriented or moderate/indifferent profile. So, students with these latter two profiles do not perform better on math tests than students with a multiple goals profile.

[Figure 4, Figure 5, Table 4]

Relationships between transitions in students’ language achievement goal profiles and

developments in effort and achievement in language/reading comprehension

The mean scores for each achievement goal profile in language on teacher-reported school effort, self-reported effort in language and achievement in reading comprehension in sixth grade are displayed in figures 2 and 3. In table 5, the effects of transitions in students’ achievement goal profiles in language on developments in teacher-reported school effort, self-reported effort in language and achievement in reading comprehension are displayed. After controlling for teacher-reported school effort in grade five, no differences in growth of teacher-reported school effort were found between students who retained a

(28)

28 the year with these profiles, but made a transition to another profile during the year. Similar results were found for both self-reported effort and reading comprehension achievement, indicating that students who made a transition did not show different levels of growth in effort and achievement in language or reading comprehension than students who retained the same profile.

[Table 5]

Relationships between transitions in students’ math achievement goal profiles and

developments in effort and achievement in math

The mean scores for each math achievement goal profile on teacher-reported school effort, self-reported effort and achievement in mathematics in fifth grade are displayed in figures 4 and 5. In table 6, the effects of transitions in students’ achievement goal profiles in mathematics on developments in teacher-reported school effort, self-reported effort and achievement in math are displayed. After controlling for teacher-reported school effort in grade five, a significant difference in growth of teacher-reported school effort was found between students who retained a multiple goals profile and students who made a transition from a multiple goals profile to a moderate/indifferent profile (b = .337, p = .001), with the latter group having higher levels of growth than the former group. No differences were found between students who originally had an approach oriented or moderate/indifferent profile, but made a transition throughout the year, and students who retained an approach oriented or moderate/indifferent profile.

Regarding self-reported effort in math, students who made a transition from an approach oriented profile to a moderate/indifferent profile during the school year reported lower levels of growth than students who retained an approach oriented profile (b = -.232, p =

(29)

29 .030). Furthermore, when comparing students who retained a moderate/indifferent profile with students who made a transition from a moderate/indifferent profile to an approach oriented profile, results showed that the latter group of students reported significantly higher levels of growth (b = .581, p = .000). No differences in growth of self-reported effort were found for students who started the year with a multiple goals profile and did not transition, and students who had a multiple goals profile at the beginning of the school year, but made a transition to a moderate/indifferent profile.

For achievement in mathematics, no differences were found between students who retained a certain goal profile and students who made a transition from that goal profile to another profile. So, a transition in achievement goal profile did not result in more

achievement growth in mathematics from grade five to grade six.

[Table 6]

Relationships between innovative learning and students’ achievement goal profiles in

grade five

To investigate the cross-sectional relationships between collaborative learning, authentic learning, and the focus on self-regulated learning and students’ achievement goal profiles in language and math, multinomial logistic regressions were performed (table 7). With regard to language, results show that when the score on collaborative learning was higher, students were less likely to have a multiple goals profile than a moderate/indifferent profile (b = -.474, p = .019). Authentic learning and focus on self-regulated learning were unrelated to the odds of having either a multiple goals profile or a moderate/indifferent profile. Furthermore, no relationships were found between the three aspects of innovative learning and the odds of having either a moderate/indifferent profile or an approach oriented

(30)

30 profile, or having either a multiple goals profile or an approach oriented profile. The same applies to mathematics, no significant effects were found for collaborative learning, authentic learning and the focus on self-regulated learning on the odds of having either one of the three profiles.

[Table 7]

Relationships between innovative learning and transitions in students’ achievement goal

profiles

In table 8, the effects of innovative learning on students’ achievement goal profiles in language are reported. A significant effect was found for authentic learning when comparing students who retained a moderate/indifferent profile with students who made a transition from this profile to an approach oriented profile. When the score on authentic learning increased with one unit, the expected odds of a student who transitioned to an approach oriented profile versus a student who retained a moderate/indifferent profile were multiplied by 3.77 (b = 1.327, p = .036). No other significant effects were found for the three aspects of innovative learning while comparing students who retained a multiple goals profile and students who made a transition to a moderate/indifferent profile.

The results of the multinomial logistic regression analyses on math are displayed in table 9. For students who started the year with a moderate/indifferent profile, results show that when the score on collaborative learning increased with one unit, students were 4.564 times more likely to make a transition to an approach oriented profile than to retain a

moderate/indifferent profile (b = 1.518, p = .041). When the score on authentic learning increased with one unit, the expected odds of a student who retained a moderate/indifferent

(31)

31 profile versus a student who made a transition to an approach oriented profile were multiplied by .062 (b = -2.786, p = .015).

(32)

32 Discussion

The aim of the present study was to examine the relationships between elementary students’ achievement goal profiles and school effort and achievement, and the relationship between aspects of innovative learning and the adoption of certain achievement goal profiles. Furthermore, the changes in these relationships over time were examined. The results showed that more beneficial profiles were associated with higher scores on effort and achievement. Moreover, transitions from less beneficial goal profiles to more beneficial profiles were associated with a larger increase in effort and achievement during the year. With regard to innovative learning, higher levels of innovative instructional approaches seemed to be related to the adoption of more beneficial achievement goal profiles. Below, the outcomes will be discussed in more detail.

The relationship between students’ achievement goal profiles and effort and

achievement

In general, the results of this study offer support for the theory of multiple goal perspective, which states that students can adopt multiple achievement goals at the same time. Several patterns were found with regard to the relationship between students’ achievement goal profiles and effort and achievement in language or reading comprehension and mathematics. A more beneficial profile with respect to educational outcomes could be identified. Looking at the cross-sectional results, students with an approach oriented profile (high on mastery goals, low on performance avoidance goals, and average on performance approach goals) seemed to put slightly more effort in their school work than students with a

moderate/indifferent profile (moderate on all three goals). The advantageous effect of an approach oriented profile on school effort became even clearer when the longitudinal

(33)

33 who made a transition from a moderate/indifferent profile to an approach oriented profile reported higher levels of growth in effort than students who retained a moderate/indifferent profile. Thus, an increase in the level of mastery goals combined with a decrease in the level of performance avoidance goals seems to result in increasing effort. Likewise, students who made a transition from an approach oriented profile to a moderate/indifferent profile reported lower levels of growth in effort than students who retained an approach oriented profile. This means that a decrease in the level of mastery goals and an increase in the level of performance avoidance goals resulted in lower levels of growth in effort. Thus, in line with previous

research on goal profiles (Luo et al., 2011; Meece & Holt, 1993; Ng, 2006; Tuominen-Soini et al., 2012), that suggests that students with an approach oriented profile appear to display the most adaptive pattern with regard to effort in their school work.

Regarding the least adaptive achievement goal profile, in most instances students with a multiple goals profile (high on performance avoidance goals, below average on mastery goals, and slightly above average on performance approach goals) had the lowest scores on both school effort and achievement. The findings were comparable across subject domains. These results support the earlier findings by Luo et al. (2011) and Tuominen-Soini et al. (2012), who both found that low mastery/high performance avoidance-oriented students reported the lowest levels of effort in school. Moreover, Luo et al. (2011) not only found the same effect on effort, but also on achievement. Even though in earlier studies these

maladaptive educational outcomes were attributed solely to the adoption of performance avoidance goals (Elliot, 1999; Elliot & Church, 1997; Elliot & Harakiewizc, 1996), these results indicate that it might actually be the combination of a high score on performance-avoidance goals with a lower score on mastery goals that contributed to the lower levels of effort and achievement. The adoption of mastery goals has been associated with more beneficial learning strategies and higher levels of intrinsic motivation and interest. Thus, a

(34)

34 lower score on mastery goals could imply that students have less interest in their tasks and use learning strategies that do not enhance conceptual understanding. In addition, students with higher levels of performance avoidance goals suffer relatively often from test anxiety and self-handicapping, resulting in feelings of anxiety and worry with regard to school tasks. To protect their self-worth, these students might not give as much effort when confronted with challenging tasks. Moreover, feelings of anxiety and worry have been found to cause task distraction and low working memory (Huijun et al., 2006; Senko & Miles, 2008), which might explain the negatively influence on test performance found in this study. Thus, these combined negative manifestations of low mastery goals and high performance avoidance goals seem to have adverse effects on the effort that students with a multiple goals profile put in their school work and on their performance.

Although the least adaptive profile could be easily identified for the cross-sectional results, the longitudinal results did not show such a clear pattern. Only with regard to teacher-reported effort did one group show less growth in effort than the others. Students who made a transition from a multiple goals profile to a moderate/indifferent profile showed greater levels of growth in effort than students who retained a multiple goals profile. Although this result is in line with the cross-sectional results, no differences were found for self-reported effort and achievement across subject domains. Nevertheless, this study does contribute to the existing evidence (Tuominen-Soini et al., 2012) that students who transition from a seemingly less adaptive goal profile to a more adaptive goal profile actually do show more growth in school effort than students who retain the same goal profile or students who transition from a more adaptive goal profile to a less adaptive goal profile.

In general, students with a moderate/indifferent goal profile had higher scores on effort and achievement than students with a multiple goals profile, and lower or similar scores than students with an approach oriented goal profile. However, in two occasions, deviating

(35)

35 results were obtained. Moderate/indifferent students reported lower levels of effort in

mathematics than multiple goals or approach oriented students. In earlier studies, students with a moderate/indifferent profile were described as students who do know what is expected of them in terms of learning and doing well in school, but also try to minimize the amount of effort they put into school tasks (Tuominen-Soini, Salmela-Aro, & Niemivirta, 2010).

Therefore, finding these students to report the least amount of effort with regard to math does not come as a surprise. However, students with a moderate/indifferent goal profile did

outperform students with an approach oriented profile on math achievement. As positive relationships between the extent to which students and teachers report students’ effort and performance on school tests have often been found (for a review see Fredricks et al., 2004), which is also the case in this study for language, it is hard to interpret these seemingly contradictory results. It might be that for mathematics, approach oriented students were too focused on deeply comprehending the subject-matter, thereby getting too much into detail, while moderate/indifferent students were less distracted by details and obtained the best math scores with the least amount of effort. It might also be that the more superficial learning strategies that may be associated with a moderate/indifferent profile have a better match with the nature of mathematics tests, resulting in better performance on those tests. Nevertheless, as this finding also contradicts the study of Tuominen-Soini et al. (2008) who found

indifferent students to be among the lowest achievers, further research can help to broaden the knowledge on the effects of a moderate/indifferent goal profile on academic achievement.

The relationship between aspects of innovative learning and the adoption of

achievement goal profiles

The findings of this study indicate that the level of innovative instructional approaches might be related to students’ achievement goal profiles. With regard to collaborative learning,

(36)

36 students were more likely to adopt a moderate/indifferent goal profile than a multiple goals profile when they were allowed to collaborate,. As, in this study, a moderate/indifferent goal profile was found to be a more adaptive goal profile with regard to effort and achievement than a multiple goals profile, this finding supports the view that this innovative instructional approach may lead to better educational outcomes. In addition, higher levels of collaborative learning were also found to be related to transitions in students’ achievement goal profiles. Students with a teacher who adopts a collaborative instructional approach were more likely make a transition from a moderate/indifferent profile to an approach oriented profile than to retain a moderate/indifferent goal profile. Thus, the use of collaborative instructional

approaches seemed to lower the level of performance avoidance goals of students and to enhance the level of mastery approach goals. This is in line with previous studies in which mastery-orientations of students increased as a result of cooperative learning (Nichols, 1996; Nichols & Miller, 1994).

A similar result was found for authentic learning. Students were more likely to make a transition from a moderate/indifferent goal profile to an approach oriented goal profile than to retain a moderate/indifferent goal profile in a classroom with higher levels of authentic

learning. When teachers cover topics that match the interest of students and are relevant to their daily life outside school, students seem more likely to adopt higher levels of mastery approach goals, while at the same time their level of performance avoidance goals decreases. This result is in line with the finding of Thoonen et al. (2011), who found that higher levels of authentic learning related to higher levels of mastery-oriented goals. Thus again, support was found for the hypothesis that the use of innovative instructional approaches enhances the motivation of students through the adoption of a more beneficial achievement goal profile.

For mathematics, however, contrasting results were found. With higher levels of authentic learning, students were less likely to make a transition to an approach oriented

(37)

37 profile than to retain a moderate/indifferent profile. So in this case, they did not adopt a more beneficial achievement goal profile. It might be that for mathematics, this approach is less effective than a more traditional approach, and students benefit from more structured math instructions. Another explanation might be that teachers were less able to find a way to make their math lessons meaningful to their students, while they were able to do so with language. As self-reports were used to assess the extent to which teachers used authentic instructional approaches, no information is available on how teachers implemented these approaches in their lessons. In future research, classroom observations could be used to give more insight in how teachers attempt to create a more innovative learning context and the way this affects students’ achievement goal orientations.

Last, no effects were found for teacher focus on self-regulated learning, indicating that this innovative approach did not seem to influence the type of achievement goal profile students adopted. However, out of all three innovative approaches, not only did teachers report the highest scores on focus on self-regulated learning, but differences between teachers were the smallest on this instructional approach. Therefore, finding results with regard to focus on self-regulated learning might have been more difficult. Furthermore, the teachers’ self-reports did not give insights in the quality of the innovative instructions. As Dignath et al. (2008) showed, in effective interventions students are taught cognitive, metacognitive and motivational strategies, and are provided knowledge on strategy use and its benefits. As no information is available on the quality of the instructions, it is unclear whether the teachers were able to optimally convey these strategies to their students. Besides classroom

observations, intervention studies might give a clearer understanding whether students’ achievement goal profiles are enhanced when teachers promote self-regulated learning.

Referenties

GERELATEERDE DOCUMENTEN

Table 1 Mean values and standard deviations of the correlations between the fundamental fre- quency and the intensity contour and MBPS value for each

A selected number of these global e-governance practices and models in chapter 3 will be incorporated (in Chapter 4) into a workable, drafted e-governance training model

The present study compared the psychometric properties of two instruments designed to assess trauma exposure and PTSD symptomatology and asked the question: &#34; Do the K-SADS

The main focus of this research is to derive a stability model which can encounter the enhanced formability obtained when simultaneous bending and stretching is applied to

Een geografische uitbreiding van het onderzoeksgebied én een specialisatie op grafiek naast een uitbreiding van de doelgroep voor de bibliotheek, waar niet alleen studenten

Therefore, it is time to no longer consider Russian constituent ordering in the context of competition between information structure and definiteness, but in terms of

LPF, low-pass filter; BS, beam splitter; NDF, neutral density filter; R, piezo- driven reference mirror; ⌬R, scanning range of R; MM, multimode graded index detection fiber; FT,

soonlike voornaamwoord, bv. In hierdie geval word die betrokke verskynsel volgens die no- tasie wat die oorspronklike outeur gebruik h~t, beskryf. Tog word dit gekategoriseer