Which emotional profiles exhibit the best learning outcomes? A person-centered analysis of students’ academic emotions

Abstract

Previous studies on academic emotions have mostly used variable-centered approaches. Although these studies have elucidated the relationships between academic emotions and key academic outcomes, they cannot identify naturally-occurring groups of students defined by distinct academic emotion profiles. In this study, we adopted a person-centered approach to explore whether students can be grouped in terms of distinct academic emotion profiles and whether these groups differed in terms of key academic outcomes. Cluster analyses showed four distinct profiles across both domain-general (Study 1) and domain-specific (Study 2) academic emotions. Students with high levels of positive academic emotions and low levels of negative academic emotions exhibited the most adaptive educational outcomes followed by students characterized by high levels of positive emotions and moderately high levels of shame. The most maladaptive profile was exhibited by students who are low in positive academic emotions and high in negative academic emotions. Theoretical and practical implications are discussed.

Keywords

academic emotions achievement engagement motivation person-centered approach

Students experience multiple types of emotions in school (i.e., academic emotions). For example, being in a mathematics class may simultaneously make students feel joy when the teacher engages them in interesting activities and relief when they finally finish an exam. However, for many years, educational researchers have mostly neglected the role of emotions in a students’ life focusing exclusively on cognitive, motivational, and behavioral constructs. In recent years, there has been an increasing recognition that academic emotions have a crucial impact on students’ learning outcomes (Pekrun & Linnenbrink-Garcia, 2012).

The control-value theory of emotions (Pekrun, Goetz, Titz, & Perry, 2002) assumes that emotions are connected to achievement motives, activities, and outcomes. The theory argues that academic emotions shape key learning processes. Students who experience positive academic emotions (i.e., enjoyment and pride) are likely to achieve higher grades (Villavicencio & Bernardo, 2013; Pekrun, Goetz, Frenzel, Barchfeld, & Perry, 2011), pursue mastery-oriented goals (Pekrun, Elliot, & Maier, 2006), engage in effective cognitive and metacognitive strategy use (Artino & Jones, 2012; King & Areepattamannil, 2014; Muis, Psaradellis, Lajoie, Leo, & Chevlier, 2015), and participate more actively in class activities (King & Gaerlan, 2014; King, McInerney, Ganotice, & Villarosa, 2015; Pekrun et al., 2011). Conversely, the experience of negative academic emotions (e.g., boredom) leads to lower levels of achievement (Pekrun, Hall, Goetz, & Perry, 2014) and decreased effort (Dettmers et al., 2011).

Although these investigations have revealed interesting insights regarding the role of emotions in the academic context, these studies have focused on detecting the association of specific academic emotions with academic outcomes. This strategy in assessing the linkage among educational and psychological constructs is known as a variable-centered approach.

An alternative way to look at the academic emotion-school outcome relationship is to utilize a person-centered approach by examining how particular combinations of academic emotions are associated with key school outcomes. A few studies in the social psychology literature may offer preliminary evidence about the possible impact of experiencing simultaneous emotions with opposite valence (positive or negative). For example, when individuals watch tragi-comedies, encounter bittersweet situations, or watch advertisements, they tend to elicit conflicting emotions like happiness and sadness (Larsen, McGraw, & Cacioppo, 2001). Furthermore, Brown, Gonzalez, Zagefka, Manzi, and Cehajic (2008) found that high levels of guilt and low levels of shame made individuals more likely to make reparations. However, for those high in guilt and shame, attempts at making amends may actually decrease. This study shows that varying combinations of emotions may lead to distinct outcomes.

Despite the theoretical value of investigating academic emotions through person-centered approaches, we know of no existing study which assessed the link between academic emotions and key student outcomes using this methodological lens. In fact, although some studies have used a person-centered approach in the context of motivation research (Nurmi & Aunola, 2005; Roeser & Galloway, 2002), very limited investigations adopted a person-centered approach in the field of educational psychology. Using a person-centered approach offers distinct advantages compared to variable-centered approaches. Although a variable-centered approach may offer insights on how specific academic emotions are linked to distinct educational outcomes, they can only portray how one variable is related to another but cannot shed light on students’ emotional profiles and how these profiles are associated with various academic outcomes. Furthermore, Roeser, Eccles, and Sameroff (1998) proposed that utilizing a person-oriented approach is beneficial for developing interventions that are sensitive to the diverse needs of individuals with distinct psychological profiles because this approach explores the association among the variables of interests at the level of an individual.

Therefore, the aim of the current study was to examine students’ academic emotion profiles in relation to key educational outcomes such as motivation (Study 1), engagement, and academic achievement (Study 2) through a person-centered approach.

Academic emotions

Academic emotions refer to the specific feelings that students experience in the educational context (Schutz & Pekrun, 2007). There are two general types of academic emotions: positive emotions and negative emotions. Academic emotions have been found to be associated with key cognitive-motivational constructs, school-related behaviors, and academic performance (Pekrun 2006, 2009). Positive emotions (e.g., joy, hope, and pride) have been found to boost students’ mastery motivation and academic achievement (Pekrun et al., 2002, Pekrun, Goetz, Perry, Kramer, & Hochstadt, 2004), whereas negative emotions (e.g., anger, anxiety, and boredom) have been found to be associated with avoidance motivation and lower levels of academic achievement (Daniels et al., 2009).

Previous studies have classified academic emotions in terms of valence and activation (Pekrun, 2000; Pekrun et al., 2002). Valence refers to the extent to which emotions are considered as positive or negative. Activation pertains to the extent to which emotions are considered physiologically activating (e.g., hope) or deactivating (e.g., relief). Pekrun and his colleagues (2002) have recognized the potential value of integrating activation (activating versus deactivating emotions) to understand the role of emotions in catalyzing academic outcomes. They have proposed that academic emotions can be further classified into: a) positive activating emotions (enjoyment hope, pride); b) positive deactivating emotions (relief); c) negative activating emotions (anger, anxiety, shame); and d) negative deactivating emotions (hopelessness, boredom). These different types of emotions have been found to be associated with a diverse set of educational outcomes.

Academic emotions, motivation, engagement, and achievement

Academic emotions matter for students’ key learning outcomes. For example, whereas positive academic emotions (e.g., pride, hope) positively predicted performance (Villavicencio & Bernardo, 2013; Pekrun, Elliott, & Maier, 2009), negative academic emotions negatively predicted academic achievement (Pekrun et al., 2009).

Furthermore, academic emotions have a crucial impact on academic motivation. Negative emotions (e.g., anger, anxiety, shame, hopelessness, and boredom) are believed to be detrimental to students’ success because these emotional states can impede students’ interest towards a particular task (Pekrun et al., 2006; Turner et al., 1998). Conversely, positive emotions (e.g., enjoyment, pride, and hope) have been associated with academic motivation (Liljedahl, 2005; McLeod, 1988) and enjoyment in school (Martin, 2003; McInerney & Ali, 2006).

The present research adopted self-determination theory (Ryan & Deci, 2000) to examine students’ academic motivation. A major tenet of this motivational framework points to the advantageous role of intrinsic motivation in facilitating positive academic outcomes. Intrinsic motivation operates when students are driven to study because they find enjoyment and fulfillment in performing academic tasks. On the other hand, extrinsic motivation is regarded as a less adaptive motivational orientation and characterizes situations wherein students embody externally-oriented reasons for studying. Extrinsic motivation can be further categorized into different types depending on the extent of autonomy one experiences when engaging in academic tasks (Ryan & Deci, 2000).

External regulation is the least autonomous form of extrinsic motivation which takes place when individuals study to gain rewards or avoid punishment. Introjected regulation happens when individuals feel that they are obliged to study or to avoid feeling guilty about not studying. Identified regulation is more autonomous which occurs when individuals study because they find the activity personally significant. The most autonomous form of extrinsic motivation is integrated regulation which takes place when individuals consider academic tasks as an essential component of their personal value system.

Most recently, some scholars have recognized the theoretical value of combining external and introjected regulation to form controlled motivation and aggregating identified and integrated relation to form autonomous motivation (Ratelle, Guay, Vallerand, Larose, & Senecal, 2007; Shahar, Henrich, Blatt, Ryan, & Little, 2003). Ratelle and her colleagues (2007) have found that students with higher autonomous motivation tend to adjust effectively in the academic context. Autonomous motivation was also linked to greater achievement (Black & Deci, 2000; Cerasoli, Nicklin, & Ford, 2014; Datu, King, & Valdez, 2016).

The present study

The overarching objective of the study was to assess the extent to which naturally-occurring patterns of academic emotions were associated with key educational outcomes through adopting a person-centered approach. Our study hopes to address notable gaps in the extant literature regarding the associations of various emotion profiles with academic functioning because previous studies have commonly relied on variable-centered approaches to assess the linkage between academic emotions and positive student outcomes.

In Study 1, we tested how students’ emotion profiles are related to academic motivation (autonomous vs. controlled motivation) in a domain-general context. Specifically, we examined the cluster structure of students’ academic emotions in general schooling and we explored the extent to which varying profiles of academic emotions can adequately explain differences in the school motivation of secondary students.

In Study 2, we examined students’ academic emotion profiles in relation to their mathematics class and investigated how these profiles were related to achievement in the mathematics domain. We also examined how these clusters would differ in terms of a wide range of engagement indices. Cluster comparison in terms of the students’ academic emotion profiles could offer interesting insights on what combination of emotions may be considered adaptive in the educational setting.

The following hypotheses were posited in the study:
H1: The high-positive-low negative cluster would demonstrate the highest extent of autonomous motivation compared to the other clusters. (Study 1).

H2: Compared to other empirically-derived clusters, the high-positive-low negative academic emotion profile would demonstrate better academic engagement and achievement (Study 2). That is, we anticipated that students who are high in positive academic emotions and low in negative emotions would demonstrate the highest degree of adaptive school outcomes: Engagement (operationalized in terms of university intention, school valuing, leaving school, and affect to school) and achievement (mathematics grade).

H3: There would be differences in the cluster structures between a domain-general and a domain-specific contexts (mathematics). We based this assumption on the premise of control-value theory that control- and value-related cognitions are considered to be largely domain specific, thus we expected differential cluster structures between a domain-general (Study 1) and domain-specific (Study 2) context.

Study 1: Academic emotions and motivation

The aim of Study 1 was to investigate students’ emotional profiles and to examine how students with different emotional profiles vary in terms of their academic motivation. Students’ emotions and academic motivation were examined in a domain-general manner. That is, we measured students’ emotions and motivation towards school in general and not towards a specific subject domain.

Method

Participants

Participants were 1,147 adolescent secondary students from five private schools in Metro Manila, Philippines. Of these participants, 54.32% (N = 623) were male and 45.68% (N = 524) were female. Their average age was 14.19 years (SD = 1.39). Passive consent forms were distributed and collected before conducting the survey. The questionnaires were distributed and completed during classes. Participants were assured that their responses would not affect their grades.

Measures

Academic emotions

The Short Version of the Academic Emotions Questionnaire for Filipinos (S-AEQ-F; King 2010), which consists of 14 items, was used to measure different types of academic emotions. It was based on the Achievement Emotions Questionnaire for Learning developed by Pekrun et al. (2002). The S-AEQ-F has been previously validated among Filipino students (King, 2010).The questionnaire uses a six-point Likert scale with higher values indicating greater endorsement (1 = not like me at all; 6 = very much like me). It measures three types of positive academic emotions: enjoyment (e.g., ‘I look forward to studying’), hope (e.g., ‘I feel optimistic that I will make good progress in studying’), and pride (e.g., ‘I think I can be proud of my accomplishments at studying’) and five types of negative academic emotions: anger (e.g., ‘I get angry when I have to study’), anxiety (e.g., ‘I get tense and nervous while studying’), shame (e.g., ‘I feel ashamed because I am not as good as others in school’), hopelessness (e.g., ‘I feel hopeless when I think about studying’), and boredom (e.g., ‘Studying is dull and monotonous’).

Motivation

Students’ reasons for studying were assessed with an adapted version of the Academic Self-Regulation Scale (Ryan & Connell, 1989) with the following subscales: Intrinsic motivation (four items, ‘I’m studying because I enjoy doing it’), identified regulation (four items ‘I’m studying because this represents a meaningful choice to me’), introjected regulation (four items ‘I’m studying because I want others to think I’m smart’), and external regulation (four items, ‘I’m studying because others oblige me to do so’). Another way of looking at motivation is to divide it into autonomous and controlled motivation. We followed the procedure performed by Pelletier, Fortier, Vallerand, and Brière (2001) and Vansteenkiste, Lens, Dewitte, Dewitte, and Deci (2004) by averaging intrinsic motivation and identified motivation to create a composite score for autonomous motivation, and introjected and external regulation for controlled motivation.

The English versions of psychological questionnaires were administered in Study 1 and Study 2 because English serves as the official medium of instruction among secondary school students in the Philippine context.

Data analysis

Descriptive statistics (e.g., mean and standard deviation), reliability coefficients, and correlational coefficients were calculated. We first used hierarchical cluster analysis using Ward’s method (Ward, 1963) to classify students in terms of their academic emotions. Next, we performed k-means cluster analysis to validate whether the empirically-derived clusters using the hierarchical cluster analysis could be replicated using a different clustering procedure. The 21st version of the Statistical Package for Social Sciences was used to perform all the relevant analyses.

Results and brief discussion

The values of skewness and kurtosis did not exceed 2 and 7 which indicated that the data is normally distributed based on the criteria of Finney and DiStefano (2006). A review of Mahalanobis distance values also revealed no multivariate outlier in the research.

The descriptive statistics and correlational analyses are presented in Table 1. As expected, positive emotions were linked to higher autonomous and controlled motivational orientations. Mixed findings were observed in terms of negative emotions, as shame and hopelessness were associated with higher controlled motivation while the remaining negative emotions were associated with lower autonomous and controlled motivation.
Table 1.
Descriptive statistics and intercorrelations amongst measured variables (Study 1, n = 1,147).

M SD α Skewness Kurtosis 1 2 3 4 5 6 7 8 9 10

Academic emotion

1. Enjoyment 4.73 .99 .75 −.54 −.08 – .48* .65* −.01 −.29* .−30* .02 −.04 .73* .39*

2. Pride 4.60 1.05 .61 −.63 .21 – .540 .03 −.14* −.13* .03 −.03 .52* .31*

3. Hope 4.49 1.06 .83 −.53 .38 – .01 −.14* −.20* .02 −.03 .57* .34*

4. Anxiety 3.17 1.30 .62 .10 −.56 – .41* .47* .58* .44* −.01 .12*

5. Boredom 2.42 1.38 .83 .78 −.27 – .69* .38* .51* −.29* −.10

6. Anger 2.50 1.33 .88 .63 −.42 − .37* .42* −.31* −.10

7. Shame 3.29 1.35 .68 .06 −.63 − .52* .01 .11*

8. Hopelessness 3.10 1.39 .74 .18 −.65 – −.01 .08

Motivation

9. Autonomous 4.65 .94 .72 −.43 .05 – .48***

10. Controlled 4.25 1.06 .84 −.51 .15 –

Note: * p < .05, p < .01, * p < .001.

Table 2.
Mean scores on study variables across motivational profiles (n = 1,147).

Cluster1– (n = 419, 36.53 %) High shame group
Cluster 2 – (n = 290, 25.28%) Moderate
Cluster 3 – (n = 228, 19.88%) Maladaptive
Cluster 4– (n = 210, 18.30%) Adaptive
F p η²

M SD M SD M SD M SD

Constituting dimension

1. Enjoyment 5.21^a .67 3.94^b .93 4.36^c .92 5.27^a .75 185.257 .001 .327

2. Pride 5.02^a .70 3.66^b .89 4.39^c 1.08 5.29^d .66 209.218 .001 .354

3. Hope 4.91^a .80 3.58^b .83 4.31^c 1.03 5.12^d .86 176.003 .001 .316

4. Anxiety 3.28^a 1.11 2.78^b 1.09 4.37^c .97 2.19^d 1.10 164.134 .001 .301

5. Boredom 1.97^a .97 2.23^b .99 4.37^c .95 1.47^d .68 445.789 .001 .539

6. Anger 2.07^a 1.01 2.39^b 1.03 4.21^c .97 1.66^d .86 306.909 .001 .446

7. Shame 3.78^a 1.00 2.80^b 1.00 4.46^c 1.05 1.74^d .71 352.633 .001 .481

8. Hopelessness 3.30^a 1.19 2.60^b .96 4.37^c 1.00 2.00^d 1.33 184.893 .001 .327

Motivation

9. Autonomous 4.50^a 1.00 3.79^b 1.00 4.21^ab .99 4.43^ab 1.15 29.61 .001 .072

10. Controlled 5.04^a .73 4.03^b .84 4.27^c .92 5.12^a .81 128.30 .001 .252

Note: For each dependent variable, bolded means are the highest mean scores, means with different subscripts indicate a significant difference at p < .05 using Scheffe. Cluster means in the same row are significantly different if they have different subscripts. Constituting variable: F(24,3414) = 118.226, pillais V = 1.362; partial η²= .452; motivation: F(6,2286) = 57.88, pillais V = .264; partial η²= .132

In performing cluster analysis, hierarchical clustering approach using Ward’s method was adopted because this strategy is optimal when no apriori structure has been posited about a given data structure (Hastie, Tibshirani, & Friedman, 2009). In order to determine how many clusters should be used, we looked at the dendogram and the agglomeration schedule which provide a solution for every number of cluster from 2 to 1,147 (the total sample size). More specifically we focused on the changes in the agglomeration coefficient and found that after generating four clusters, succeeding clusters added little to distinguishing between cases.

We further verified the validity of this four-cluster solution by conducting a k-means cluster analysis wherein we specified that we wanted the sample to be divided into four clusters. The k-means cluster analysis generated profiles that were very similar to what we obtained for the hierarchical cluster analysis. This made us more confident in the validity of the four-cluster solution obtained from the hierarchical cluster analysis.

We describe each of the four clusters next: Cluster 1 (Adaptive shame) was the biggest group which constituted 36.53% of the sample (n = 419). We called this the adaptive shame group because students in this cluster had high levels of positive emotions and low levels of negative emotions except for shame which was slightly higher than the other groups. The second cluster (moderate group) comprised 25.28% of the sample (n = 290). It was characterized by moderate levels of positive and negative emotions. The third cluster (maladaptive group) was characterized by low positive and high levels of negative academic emotions. This group comprised the 19.88% of the sample (n = 228). The fourth cluster was called the adaptive group and was characterized by high levels of positive emotions and low levels of negative emotions. This group comprised 18.30% of the total sample (n = 210). Figure 1 illustrates the academic emotions profile of four clusters in Study 1.
Figure 1.
Academic emotions profile of the four clusters in Study 1.

To further examine whether the four clusters actually differed in terms of academic emotions, we performed a MANOVA with the four clusters as the independent variable and the seven academic emotions as the dependent variables. Result showed significant differences on academic emotions among the four clusters, F(24, 3414) = 118.23, Pillai’s V = 1.36; partial η²= 0.454 .

Next, we examined whether the four clusters differed in terms of academic motivation using MANOVA. Our results indicate significant differences amongst the four clusters on motivation, F(6, 2286) = 57.888; Pillai’s V = 0.264; partial η²= 0.132. Confirming H1, the results indicated that Cluster 1 (adaptive shame) and Cluster 4 (adaptive group) had the highest levels of autonomous and controlled motivation. Cluster 2 (moderate group) had relatively lower levels of autonomous and controlled motivation, while Cluster 3 (maladaptive group) had the lowest levels of controlled motivation.

Study 2: Academic emotions, engagement, and achievement

In Study 1, we showed that there were four distinct emotional cluster profiles and that students with different profiles exhibited distinct motivational outcomes. In particular, we found that the adaptive group and the adaptive shame group had the best motivational outcomes. However, Study 1 only examined academic emotion profiles in the general school context. We wanted to examine whether the emotional clusters we found in Study 1 would generalize to a specific school subject (i.e., mathematics). If we can replicate the emotional clusters in Study 1 to Study 2 then it would provide stronger evidence of the validity of our cluster structure. We also compared the different clusters in terms of their mathematics achievement and other engagement outcomes.

To do this, we involved a sample of secondary students and assessed their academic emotions in mathematics. The assessment of students’ academic emotion in a more domain specific academic subject is important because this enabled us to examine profile similarities and differences of Filipino students across domain-general and domain-specific contexts. We then compared the different clusters in terms of their Mathematics achievement and other related engagement-related indices.

Methods

Participants

A total of 341 secondary students from schools in Palawan, Philippines participated in the study. The average age was 13.53 years old (SD = 1.72). There were 106 males and 235 females. Of these participants, 86 (25.2%) students were in their first year, 84 (24.6%) students in their second year, 88 (25.8%) in their third year, and 83 (24.3%) students in their fourth year. Before requesting the participants to answer the questionnaires, the signed passive consent forms were collected.

Measures

Academic Emotions Questionnaire––Math(AEQ-Math)

The learning-related Achievement Emotion Questionnaire––Math (AEQ-M; Pekrun, Goetz, & Frenzel, 2005) was used to assess the emotions participants experience when studying mathematics. The instruction for the measure asked the participants to describe how they felt, typically, in a mathematics class. AEQ is composed of eight subscales: Enjoyment (‘I enjoy my Math class’), hope (‘I have optimistic view towards studying’), pride (‘I am proud of my contributions to the Math class’), anger (‘I am annoyed during my Math class’), anxiety (‘When thinking about my Mathematics class, I get nervous’), shame (‘When I say something in Math class, I can tell that my face gets red’), hopelessness (‘I feel down’), and boredom (‘I think Mathematics class is boring’). The ratings were made on a five-point Likert scale (1 = strongly disagree and 5 = strongly agree) with higher scores indicating a greater endorsement of the construct.

Facilitating Conditions Questionnaire (FCQ)

This study considered four subscales of the original FCQ (McInerney, 1992): University intention (e.g., ‘Most people who are important to me think that I should go to college or university’), school valuing (e.g., ‘People who have a good schooling get more out of life than those who don’t’), and affect toward school (e.g., ‘I like working at school’). A five-point Likert-type scale (ranging from 1 = strongly disagree to 5 = strongly agree) was used.

Achievement

Math grades were taken from the records of the teachers after the data collection for the surveys. The grades ranged from 0% to 100% with higher marks indicating higher achievement in Mathematics.

Results and brief discussion

A review of the skewness and kurtosis values indicate normality because the skewness values ranged from 0.002 to 2.0 while the kurtosis values ranged from −0.052 to 0.810. Similar to Study 1, we used the criteria of Finney and DiStefano (2006) in judging the normality of the dataset. Examination of the Mahalanobis distance values revealed no potential multivariate outlier in the study.

The descriptive statistics and correlational coefficients are presented in Table 3. Consistent with the theoretical conjectures, positive academic emotions were associated with higher engagement and achievement while negative emotions were linked to lower engagement and achievement. The results of reliability analyses, descriptive statistics, and correlational analyses are shown in Table 3.
Figure 2.
Academic emotions profile of four clusters in Study 2.

Table 3.
Descriptive statistics and intercorrelations among measured variables (Study 2, N = 341).

M SD α Skewness Kurtosis 1 2 3 4 5 6 7 8 9 10 11 12 13

Academic emotion

1. Enjoyment 3.71 0.66 0.87 −0.15 0.06 – 0.69* 0.71* −0.57* −0.68* −0.65* −0.43* −0.58*** 0.13* 0.18 −0.15 0.30* 0.30*

2. Pride 3.93 0.69 0.85 −0.45 0.21 – 0.65* −0.47* −0.45* −0.46* −0.37* −0.53* 0.20* 0.25* −0.13* 0.26* 0.26*

3. Hope 3.70 0.65 0.81 −0.37 0.15 – −0.47* −0.54* −0.53* −0.39* −0.51* 0.19* 0.23* −0.20* 0.27* 0.32*

4. Anxiety 2.61 0.71 0.89 0.01 −0.39 – 0.59* 0.69* 0.74* 0.77* −0.06 −0.05 0.11* −0.14 −0.35*

5. Boredom 2.08 0.79 0.86 0.81 0.49 – 0.78* 0.48* 0.56* −0.15* −0.20* 0.24* −0.19* −0.25*

6. Anger 2.04 0.73 0.89 0.55 −0.23 – 0.65* 0.69* −0.15** −0.13* 0.18 −0.15 −0.27*

7. Shame 2.42 0.70 0.78 0.13 −0.33 – 0.72* −0.18* −0.10 0.16 −0.06 −0.29*

8. Hopelessness 2.32 0.83 0.84 0.39 0.05 – −0.18 −0.12* 0.17 −0.16 −0.34*

Engagement

9. University intention 4.77 0.43 0.81 −0.79 0.64 – 0.77* −0.54* 0.32* 0.18

10. School valuing 4.65 0.47 0.86 −0.49 0.62 – −0.50* 0.40* 0.08

11. Leave school 1.28 0.52 0.72 −0.44 0.09 – −0.18 −0.12*

12. Affect to school 3.79 0.65 0.73 −0.06 −0.06 – 0.17**

Achievement

13. Mathematics grade 86.70 7.34 – −0.22 −0.90 –

Note: p < 0.05; p < 0.01; p < 0.001.

Similar to Study 1, we utilized hierarchical clustering approach through Ward’s method. We referred to the dendogram and the agglomeration coefficients to assess the data structure in Study 2. We also performed k-means cluster to further evaluate the validity of the four-cluster structure. As the generated profiles from the k-means cluster analysis resembled the clusters that emerged in the hierarchical clustering strategy, we adopted the four-cluster solution (derived from the hierarchical cluster analysis) in the subsequent analyses.

These four clusters broadly resembled the four clusters from Study 1. The first cluster (adaptive shame) had the highest number of participants and constituted 34.90% of the total sample (n = 119). The second cluster (maladaptive group) was characterized by low levels of positive academic emotions and high levels of negative academic emotions. This group comprised 26.40% of the total sample (n = 90). The third cluster (moderate group) constituted 26.7% of the sample (n = 91) and was characterized by moderate levels of positive and negative academic emotions. Lastly, the fourth cluster (adaptive group) was characterized by high levels of positive and low levels of negative academic emotions. This group comprised 12% of the participants (41). Figure 2 illustrates the academic emotions profile of the four clusters in Study 2

To assess possible differences on the academic emotions among the identified clusters, MANOVA was conducted. Results showed a significant difference on the academic emotions, F(24, 996) = 46.57; Pillai’s V = 1.09; partial η²= 0.362. Cluster 4 (adaptive) had the highest levels of positive emotions followed by Cluster 1 (adaptive shame), Cluster 3 (moderate), and Cluster 2 (maladaptive) respectively. In terms of negative emotions, the highest levels of negative emotions were exhibited by Cluster 2 (maladaptive) followed by Cluster 3 (moderate) and Cluster 1 (adaptive shame) respectively. Cluster 4 (adaptive) had the lowest levels of negative emotions

Next, we performed a MANOVA to examine whether differences on school engagement (university intention, school valuing, leave school, and affect to school) and achievement (mathematics grade) can be adequately explained by students’ academic emotion profiles (See Table 4). Results suggest a significant multivariate effect on all the outcome variables (engagement and achievement), F(15, 1005) = 5.78, Pillai’s V = 0.24; partial η²= 0.079 which confirmed H2.
Table 4.
Mean scores on study variables across academic emotion profiles (N = 341).

Cluster1– (n = 119, 34.9%) High shame
Cluster 2 – (n = 90, 26.4%) Maladaptive
Cluster 3 – (n = 91, 26.7%) Moderate
Cluster 4– (n = 41, 12.0%) Adaptive
F p η²

M SD M SD M SD M SD

Constituting dimension

1. Enjoyment 4.01^a .39 3.06^b .50 3.50^c .36 4.74^d .27 189.996 .001 .628

2. Pride 4.19^a .53 3.43^b .67 3.67^c .41 4.89^d .18 90.148 .001 .445

3. Hope 3.93^a .46 3.23^b .55 3.45^c .43 4.64^d .40 98.916 .001 .468

4. Anxiety 2.23^a .48 3.26^b .52 2.89^c .41 1.68^d .41 146.988 .001 .567

5. Boredom 1.66^a .42 3.05^b .64 2.08^c .41 1.21^d .27 198.868 .001 .639

6. Anger 1.57^a .37 2.95^b .49 2.15^c .33 1.17^d .19 302.212 .001 .729

7. Shame 2.12^a .55 3.03^b .54 2.59^c .49 1.55^d .46 94.786 .001 .458

8. Hopelessness 1.85^a .52 3.09^b .66 2.64^c .52 1.26^d .35 149.010 .001 .570

Engagement

9. University intention 4.82^a .41 4.64^b .56 4.752 .37 4.95a .12 5.987 .01 .051

10. School Valuing 4.70 .49 4.53^a .55 4.62^a .37 4.87^b .27 5.819 .01 .049

11. Leave School 1.26^a .56 1.47^b .62 1.23^a .39 1.02^a .15 7.912 .001 .066

12. Affect to School 3.85^a .65 3.67^a .60 3.65^a .59 4.22^b .67 9.261 .001 .076

Achievement

13. Math grade 89.16^a 6.72 84.01^b 6.52 84.41^b 7.20 90.59^a 7.21 17.485 .001 .135

14. Average Math grade 88.13^a 5.72 84.89^b 5.65 85.10^b 5.75 86.70^a 7.34 10.746 .001 .087

Note: For each dependent variable, bolded means are the highest mean scores, means with different subscripts indicate a significant difference at p < .05 using Scheffe. Cluster means in the same row are significantly different if they have different subscripts.Cluster means with no subscript signify that there is no significant difference with any other cluster means. Constituting variable: F(24,996) = 46.573, pillais V = 1.086; partial η²= .362; engagement: F(12,1008) = 3.960, pillais V = .1135; partial η²= .045; achievement: F(6,674) = 8.238, pillais V = .132; partial η²= .068

The results indicated that Cluster 1(adaptive shame group) and Cluster 4 (adaptive group) had the highest scores on most of the engagement outcomes (i.e., university intention, school valuing, affect to school) and these two groups also had the highest mathematics grade. Cluster 2 (maladaptive group) and Cluster 3 (moderate group) had lower math achievement scores compared to the other clusters. Cluster 2 (maladaptive group) also had the highest score on the intention to leave school.

General discussion

The primary objective of the study was to assess the role of academic emotions in the educational context using a person-centered approach. Our results converged, to a large extent, with the extant literature and theoretical evidence regarding the adaptive role of positive academic emotions on a wide range of educational outcomes.

Our results were consistent with the findings of Fernando, Kashima, and Laham (2014) in terms of the likelihood that students would blend academic emotions as shown by cluster analysis. Contrary to H3 on the hypothesized differences on the cluster structures of academic emotions in a domain-general and domain-specific context, converging evidence has been detected across studies regarding the naturally-occurring constellation of emotions in the academic setting. Four clusters of emotional profiles emerged which involved the following: a) adaptive shame; b) maladaptive; c) moderate; and d) adaptive. These findings imply that students feel relatively complex patterns of academic emotions in school. To the best of our knowledge, this was the first investigation which adopted a person-centered approach in assessing the associations of academic emotions with key academic outcomes.

Across studies, our results offered support on H1 and H2, suggesting that the most adaptive profile was the one characterized by high endorsement of positive academic emotions and low endorsement of negative academic emotions. Students with this emotional profile have the highest levels of school functioning characterized by high levels of motivation, engagement, and achievement.

Generally, the seemingly advantageous profile associated with the adaptive cluster (high positive emotions-low negative emotions profile) corroborates the key tenets of the control-value theory of achievement emotions (Pekrun, 2000; Pekrun et al., 2002). Indeed, experiencing frequent positive emotions (i.e., enjoyment) and occasional negative emotions (anxiety) may be strongly tied to activities or outcomes that characterized achievement activities and outcomes in the educational setting. To some extent, this is because students who experience positive emotions are likely to practice effective self-regulation (Villavicencio & Bernardo, 2013).

On the other hand, however, our study demonstrated that students who belonged to the adaptive cluster (high positive emotions-low negative emotions) had the highest mean on controlled motivation. Understanding this seemingly surprising result may have cultural overtones. It is documented that Asian learners are most motivated when authority figures made the choice for them (Iyengar & Lepper, 1999; King & McInerney, 2014). This appears to contradict the central argument of self-determination theory (Deci & Ryan, 2000) which posits that personal choice is the main facilitator of intrinsic motivation and engagement. In other words, as extrinsic forms of motivation may not always be detrimental for students in collectivist societies, it is likely that students who are embodying the most optimal academic emotion profile may find value in studying for other-oriented reasons.

Interestingly, students who belong to the adaptive shame group (Cluster 1) had the highest mean values on autonomous motivation (Study 1) and average mathematics grade (Study 2). These results indicate that even the feelings of shame may be associated with positive outcomes in the educational context. A plausible explanation for these findings may point to the relative salience of shame in collectivist contexts (Crystal, Parrott, Okazaki, & Watanabe, 2001; Kitayama, Markus, & Matsumoto, 1995). In collectivist cultures, face is an important factor in interpersonal relations (Hwang, 2011; Lee, Leung, & Kim, 2014). When students do not perform well in academic tasks, they may feel loss of face before their parents, teachers, and classmates. Therefore, they may be especially motivated to work harder in their academic tasks to avoid losing face. This corroborates previous studies which showed that collectivist students’ desire to avoid showing their lack of competence before significant others can actually lead to adaptive educational outcomes (Chan & Lai, 2006; King, 2015, 2016; Lau, Liem, & Nie, 2008).

Furthermore, previous studies have also pointed out that shame functions as a negative but activating form of academic emotion (Pekrun, 2000; Pekrun et al., 2002). As such, shame may not be as harmful as other types of negative academic emotions such as boredom and hopelessness because it may actually spur students to work harder next time. This is especially evident in collectivist societies where motivation has been found to be more socially-oriented (Bernardo, 2008; Cheng & Lam, 2013; King & Ganotice, 2015; King, Ganotice, & Watkins, 2014; King & McInerney, 2014; King, McInerney, & Watkins, 2013).

The present research has a number of limitations. First, we involved a different set of participants in two studies from various secondary schools which may offer limited insights about the possibility of students’ switching of academic emotions profiles from one subject domain to another. This can be addressed by utilizing within-subject designs to examine if students switch academic emotions as a function of the subject domain. Second, we utilized self-report assessments which are prone to common method variance so it is recommended that future studies utilize alternative approaches in assessing academic emotions. Third, this study is cross-sectional in nature. Future studies may use a longitudinal strategy to allow investigation of whether the emotion clusters would display different educational trajectories over time and whether some students might change to a different cluster over time.

Despite these limitations, the findings obtained in these studies underscore the importance of assessing the theoretical linkage of academic emotions to key educational outcomes through a person-centered approach. Generally, our study indicates that students who experience frequent positive emotions and sporadic negative emotions are likely to achieve desirable academic outcomes. In terms of practice, psychologists, counselors, educators, and administrators are encouraged to plan and implement interventions that are sensitive to the needs of students with distinct emotional characteristics so that they can optimize students’ success in the academic context.

School mental health practitioners (i.e., guidance counselors and school psychologists) are recommended to collaborate with teachers in designing instructional approaches or learning activities that can enhance students’ positive feelings and reduce negative emotions when working on academic tasks. Practitioners can also work towards developing a positive classroom climate which could facilitate positive emotions. It would also be important to nurture positive relationships between students and school staff which may reduce negative feelings and increase positive emotions in school.

	M	SD	α	Skewness	Kurtosis	1	2	3	4	5	6	7	8	9	10
Academic emotion
1. Enjoyment	4.73	.99	.75	−.54	−.08	–	.48***	.65***	−.01	−.29***	.−30***	.02	−.04	.73***	.39***
2. Pride	4.60	1.05	.61	−.63	.21		–	.540**	.03	−.14***	−.13***	.03	−.03	.52***	.31***
3. Hope	4.49	1.06	.83	−.53	.38			–	.01	−.14***	−.20***	.02	−.03	.57***	.34***
4. Anxiety	3.17	1.30	.62	.10	−.56				–	.41***	.47***	.58***	.44***	−.01	.12***
5. Boredom	2.42	1.38	.83	.78	−.27					–	.69***	.38***	.51***	−.29***	−.10**
6. Anger	2.50	1.33	.88	.63	−.42						−	.37***	.42***	−.31***	−.10**
7. Shame	3.29	1.35	.68	.06	−.63							−	.52***	.01	.11***
8. Hopelessness	3.10	1.39	.74	.18	−.65								–	−.01	.08**
Motivation
9. Autonomous	4.65	.94	.72	−.43	.05									–	.48***
10. Controlled	4.25	1.06	.84	−.51	.15										–

	Cluster1– (n = 419, 36.53 %) High shame group	Cluster 2 – (n = 290, 25.28%) Moderate	Cluster 3 – (n = 228, 19.88%) Maladaptive	Cluster 4– (n = 210, 18.30%) Adaptive	F	p	η²
Constituting dimension
1. Enjoyment	5.21^a	.67	3.94^b	.93	4.36^c	.92	5.27^a	.75	185.257	.001	.327
2. Pride	5.02^a	.70	3.66^b	.89	4.39^c	1.08	5.29^d	.66	209.218	.001	.354
3. Hope	4.91^a	.80	3.58^b	.83	4.31^c	1.03	5.12^d	.86	176.003	.001	.316
4. Anxiety	3.28^a	1.11	2.78^b	1.09	4.37^c	.97	2.19^d	1.10	164.134	.001	.301
5. Boredom	1.97^a	.97	2.23^b	.99	4.37^c	.95	1.47^d	.68	445.789	.001	.539
6. Anger	2.07^a	1.01	2.39^b	1.03	4.21^c	.97	1.66^d	.86	306.909	.001	.446
7. Shame	3.78^a	1.00	2.80^b	1.00	4.46^c	1.05	1.74^d	.71	352.633	.001	.481
8. Hopelessness	3.30^a	1.19	2.60^b	.96	4.37^c	1.00	2.00^d	1.33	184.893	.001	.327
Motivation
9. Autonomous	4.50^a	1.00	3.79^b	1.00	4.21^ab	.99	4.43^ab	1.15	29.61	.001	.072
10. Controlled	5.04^a	.73	4.03^b	.84	4.27^c	.92	5.12^a	.81	128.30	.001	.252

	M	SD	α	Skewness	Kurtosis	1	2	3	4	5	6	7	8	9	10	11	12	13
Academic emotion
1. Enjoyment	3.71	0.66	0.87	−0.15	0.06	–	0.69***	0.71***	−0.57***	−0.68***	−0.65***	−0.43***	−0.58***	0.13*	0.18**	−0.15**	0.30***	0.30***
2. Pride	3.93	0.69	0.85	−0.45	0.21		–	0.65***	−0.47***	−0.45***	−0.46***	−0.37***	−0.53***	0.20***	0.25***	−0.13*	0.26***	0.26***
3. Hope	3.70	0.65	0.81	−0.37	0.15			–	−0.47***	−0.54***	−0.53***	−0.39***	−0.51***	0.19***	0.23***	−0.20***	0.27***	0.32***
4. Anxiety	2.61	0.71	0.89	0.01	−0.39				–	0.59***	0.69***	0.74***	0.77***	−0.06	−0.05	0.11*	−0.14**	−0.35***
5. Boredom	2.08	0.79	0.86	0.81	0.49					–	0.78***	0.48***	0.56***	−0.15***	−0.20***	0.24***	−0.19***	−0.25***
6. Anger	2.04	0.73	0.89	0.55	−0.23						–	0.65***	0.69***	−0.15**	−0.13*	0.18**	−0.15**	−0.27***
7. Shame	2.42	0.70	0.78	0.13	−0.33							–	0.72***	−0.18***	−0.10	0.16**	−0.06	−0.29***
8. Hopelessness	2.32	0.83	0.84	0.39	0.05								–	−0.18**	−0.12*	0.17**	−0.16**	−0.34***
Engagement
9. University intention	4.77	0.43	0.81	−0.79	0.64									–	0.77***	−0.54***	0.32***	0.18**
10. School valuing	4.65	0.47	0.86	−0.49	0.62										–	−0.50***	0.40***	0.08
11. Leave school	1.28	0.52	0.72	−0.44	0.09											–	−0.18**	−0.12*
12. Affect to school	3.79	0.65	0.73	−0.06	−0.06												–	0.17**
Achievement
13. Mathematics grade	86.70	7.34	–	−0.22	−0.90													–

	Cluster1– (n = 119, 34.9%) High shame	Cluster 2 – (n = 90, 26.4%) Maladaptive	Cluster 3 – (n = 91, 26.7%) Moderate	Cluster 4– (n = 41, 12.0%) Adaptive	F	p	η²
Constituting dimension
1. Enjoyment	4.01^a	.39	3.06^b	.50	3.50^c	.36	4.74^d	.27	189.996	.001	.628
2. Pride	4.19^a	.53	3.43^b	.67	3.67^c	.41	4.89^d	.18	90.148	.001	.445
3. Hope	3.93^a	.46	3.23^b	.55	3.45^c	.43	4.64^d	.40	98.916	.001	.468
4. Anxiety	2.23^a	.48	3.26^b	.52	2.89^c	.41	1.68^d	.41	146.988	.001	.567
5. Boredom	1.66^a	.42	3.05^b	.64	2.08^c	.41	1.21^d	.27	198.868	.001	.639
6. Anger	1.57^a	.37	2.95^b	.49	2.15^c	.33	1.17^d	.19	302.212	.001	.729
7. Shame	2.12^a	.55	3.03^b	.54	2.59^c	.49	1.55^d	.46	94.786	.001	.458
8. Hopelessness	1.85^a	.52	3.09^b	.66	2.64^c	.52	1.26^d	.35	149.010	.001	.570
Engagement
9. University intention	4.82^a	.41	4.64^b	.56	4.752	.37	4.95a	.12	5.987	.01	.051
10. School Valuing	4.70	.49	4.53^a	.55	4.62^a	.37	4.87^b	.27	5.819	.01	.049
11. Leave School	1.26^a	.56	1.47^b	.62	1.23^a	.39	1.02^a	.15	7.912	.001	.066
12. Affect to School	3.85^a	.65	3.67^a	.60	3.65^a	.59	4.22^b	.67	9.261	.001	.076
Achievement
13. Math grade	89.16^a	6.72	84.01^b	6.52	84.41^b	7.20	90.59^a	7.21	17.485	.001	.135
14. Average Math grade	88.13^a	5.72	84.89^b	5.65	85.10^b	5.75	86.70^a	7.34	10.746	.001	.087

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biographies

Fraide A. Ganotice Jr, PhD, is a Lecturer at the Institute of Medical and Health Sciences of the University of Hong Kong. His research interests include team-based learning, teacher commitment, student motivation, and cross-cultural test development, translation and adaptation.

Jesus Alfonso D. Datu, PhD cand, is a Doctoral candidate at the Division of Learning, Development, and Diversity of the University of Hong Kong. His research foci include positive psychology, cross-cultural psychology, and quantitative validation of psychological scales.

Ronnel B. King, PhD, is an Assistant Professor at the Department of Curriculum and Instruction, The Education University of Hong Kong. His research interests are in student motivation and well-being.

References

Artino

A. R.

Jones

K. D.

(2012) Exploring the complex relations between achievement emotions and self-regulated learning behaviors in online learning. The Internet and Higher Education 15: 170–175.

Bernardo, A. B. I. (2008). Individual and social dimensions of Filipino students achievement goals. International Journal of Psychology, 43, 886–891.

Black, A. E., & Deci, E. L. (2000). The effects of instructors' autonomy support and students' autonomous motivation on learning organic chemistry: A self-determination perspective. Science Education, 84, 740–756.

Cerasoli, C. P., Nicklin, J. M., & Ford, M. T. (2014). Intrinsic motivation and extrinsic incentives jointly predict performance: A 40-year meta-analysis. Psychological Bulletin, 140, 980–1008.

Brown

Gonzalez

Zagefka

Manzi

Cehajic

(2008) Nuestra Culpa: Collective guilt and shame as predictors of reparation for historical wrongdoing. Journal of Personality and Social Psychology 94: 75–90.

Chan

K. W.

Lai

P. Y.

(2006) Revisiting the trichotomous achievement goal framework for Hong Kong secondary students: A structural model analysis. The Asia-Pacific Education Researcher 16: 11–22.

Cheng, R. W.-Y., & Lam, S.-F. (2013). The interaction between social goals and self-construal on achievement motivation. Contemporary Educational Psychology, 38, 136–148.

Crystal

D. S.

Parrott

W. G.

Okazaki

Watanabe

(2001) Examining relations between shame and personality among university students in the United States and Japan: A developmental perspective. International Journal of Behavioral Development 25: 113–123.

Daniels

L. M.

Stupnisky

R. H.

Pekrun

R. H.

Haynes

T. L.

Newall

N. E.

& Perry

R. P.

(2009) A longitudinal analysis of achievement goals: From affective antecedents to emotional effects and achievement outcomes. Journal of Educational Psychology 101: 948–963.

10.

Datu, J. A. D. King, R. B., & Valdez, J. P. M. (2016). The academic rewards of socially-oriented happiness: Interdependent happiness promotes engagement. Manuscript submitted for publication.

11.

Deci

E. L.

Ryan

R. M.

(2000) The ‘what’ and ‘why’ of goal pursuits: Human needs and the self-determination of behaviour. Psychological Inquiry 11: 227–268.

12.

Dettmers

Trautwein

Ludtke

Goetz

Frenzel

A. C.

& Pekrun

(2011) Students’ emotions during homework in mathematics: Testing a theoretical model of antecedents and achievement outcomes. Contemporary Educational Psychology 36: 25–35.

13.

Fernando

J. W.

Kashima

Laham

S. M.

(2014) Multiple emotions: A person-centered approach to the relationship between intergroup emotion and action orientation. Emotion 14: 722–732.

14.

Finney

S. J.

DiStefano

(2006) Nonnormal and categorical data in structural equation models. In: Hancock

G. R.

Mueller

R. O.

(eds) A second course in structural equation modeling, Greenwich, CT: Information Age, pp. 269–314.

15.

Hastie

Tibshirani

Friedman

(2009) The elements of statistical learning, (2nd ed.). New York, NY: Springer.

16.

Hwang

K.-K.

(2011) Foundations of Chinese psychology, Singapore: Springer.

17.

Iyengar

Lepper

(1999) Rethinking the value of choice: A cultural perspective on intrinsic motivation. Journal of Personality and Social Psychology 76: 349–366.

18.

King

R. B.

(2010) What do students feel in school and how do we measure them?: Examining the psychometric properties of the S-AEQ-F. Philippine Journal of Psychology 43: 161–176.

19.

King

R. B.

(2015) Examining the dimensional structure and nomological network of achievement goals in the Philippines. Journal of Adolescence 44: 214–218.

20.

King, R. B. (2016). Is a performance-avoidance achievement goal always maladaptive? Not necessarily for collectivists. Personality and Individual Differences, 99, 190–195.

21.

King, R. B., & Areepattamannil, S. (2014). What students feel in school influences the strategies they use for learning: Academic emotions and cognitive/meta-cognitive strategies. Journal of Pacific Rim Psychology, 8, 18–27.

22.

King, R. B., & Gaerlan, M. J. (2014). High self-control predicts more positive emotions, better engagement, and higher achievement in school. European Journal of Psychology of Education, 29, 81–100.

23.

King, R. B., & Ganotice, F. A. (2015). Does family obligation matter for your motivation, engagement, and well-being? It depends on your self-construal. Personality and Individual Differences, 86, 243–248.

24.

King, R. B., Ganotice, F. A., & Watkins, D. A. (2014). A cross-cultural analysis of achievement and social goals among Chinese and Filipino students. Social Psychology of Education, 17, 439–445.

25.

King

R. B.

McInerney

D. M.

(2014) Culture’s consequences on student motivation: Capturing cross-cultural universality and variability through personal investment theory. Educational Psychologist 49: 175–198.

26.

King

R. B.

McInerney

D. M.

Ganotice

F. A.

Villarosa

(2015) Positive affect catalyzes academic engagement: Cross-sectional, longitudinal, and experimental evidence. Learning and Individual Differences 39: 64–72.

27.

King, R. B., McInerney, D. M., & Watkins, D. A. (2013). Examining the role of social goals in school: A study in two collectivist cultures. European Journal of Psychology of Education, 28, 1505–1523.

28.

Kitayama

Markus

H. R.

Matsumoto

(1995) Culture, self, and emotion: A cultural perspective on ‘self-conscious’ emotions. In: Tangney

Fischer

K. W.

(eds) Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride, New York, NY: Guilford, pp. 439–464.

29.

Larsen

J. T.

McGraw

A. P.

Cacioppo

J. T.

(2001) Can people feel happy and sad at the same time? Journal of Personality and Social Psychology 81: 684–696.

30.

Lau

Liem

G. A. D.

Nie

(2008) Task- and self-related pathways to deep learning: The mediating role of achievement goals, classroom attentiveness, and group participation. British Journal of Educational Psychology 78: 639–662.

31.

Lee

H. I.

Leung

A. K.-Y.

Kim

Y. H.

(2014) Unpacking East-West differences in the extent of self-enhancement from the perspective of face versus dignity culture. Social and Personality Psychology Compass 8: 314–327.

32.

Liljedahl

P. G.

(2005) Mathematical discovery and affect: The effect of AHA! experiences on undergraduate mathematics students. International Journal of Mathematical Education in Science and Technology 36: 219–235.

33.

McInerney

(1992) Cross-cultural insights into school motivation and decision making. Journal of Intercultural Studies 13(2): 53–74.

34.

McInerney

D. M.

Ali

(2006) Multidimensional and hierarchical assessment of school motivation: Cross-cultural validation. Educational Psychology 26: 595–612.

35.

McLeod

D. B.

(1994) Research on affect and mathematics learning in the JRME: 1970 to the present. Journal for Research in Mathematics Education 25: 637–647.

36.

Martin

A. J.

(2003) How to motivate your child for school and beyond, Sydney: Bantam.

37.

Muis

K. R.

Psaradellis

Lajoie

S. P.

Leo

I. D.

Chevrier

(2015) The role of epistemic emotions in mathematics problem-solving. Contemporary Educational Psychology 42: 172–185.

38.

Nurmi

Aunola

(2005) Task-motivation during the first school years: A person-oriented approach to longitudinal data. Learning and Instruction 15: 103–122.

39.

Pekrun

(2000) A social-cognitive, control-value theory of achievement emotions. In: Heckhausen

(ed.) Motivational psychology of human development, Oxford: Elsevier.

40.

Pekrun

(2006) The control–value theory of achievement emotions: Assumptions, corollaries, and implications for educational research and practice. Educational Psychology Review 18: 315–341.

41.

Pekrun

(2009) Global and local perspectives on human affect: Implications of the control-value theory of achievement emotions. In: Wosnitza

Karabenick

S. A.

Efklides

Nenniger

(eds) Contemporary motivation research: From global to local perspectives, Cambridge, MA: Hogrefe, pp. 97–115.

42.

Pekrun, R., Elliot, A. J., & Maier, M. A. (2006). Achievement goals and discrete achievement emotions: A theoretical model and prospective test. Journal of Educational Psychology, 98, 583–597.

43.

Pekrun

Elliot

A. J.

Maier

M. A.

(2009) Achievement goals and achievement emotions: Testing a model of their joint relations with academic performance. Journal of Educational Psychology 101: 115–135.

44.

Pekrun, R., Goetz, T., & Frenzel, A. C. (2005). Achievement Emotions Questionnaire—Mathematics (AEQ-M): User’s manual. Unpublished manual, University of Munich, Department of Psychology.

45.

Pekrun

Goetz

Frenzel

A. C.

Barchfeld

Perry

R. P.

(2011) Measuring emotions in students’ learning and performance: The Achievement Emotions Questionnaire (AEQ). Contemporary Educational Psychology 36: 36–48.

46.

Pekrun

Goetz

Perry

R. P.

Kramer

Hochstadt

(2004) Beyond test anxiety: Development and validation of the Test Emotions Questionnaire (TEQ). Anxiety, Stress, and Coping Vol 17: 287–316.

47.

Pekrun

Goetz

Titz

Perry

R. P.

(2002) Academic emotions in students’ self-regulated learning and achievement: A program of qualitative and quantitative research. Educational Psychologist 37: 91–105.

48.

Pekrun

Hall

N. C.

Goetz

Perry

R. P.

(2014) Boredom and academic achievement: Testing a model of reciprocal causation. Journal of Educational Psychology 106: 696–710.

49.

Pekrun

Linnenbrink-Garcia

(2012) Academic emotions and student engagement. In: Christenson

S. L.

(eds) Handbook of research on student engagement Vol 343: 259–281.

50.

Pelletier

L. G.

Fortier

M. S.

Vallerand

R. J.

Brière

N. M.

(2001) Associations between perceived autonomy support, forms of self-regulation, and persistence: A prospective study. Motivation and Emotion 4: 279–306.

51.

Ratelle

C. F.

Guay

Vallerand

R. J.

Larose

Senécal

C. B.

(2007) Autonomous, controlled, and amotivated types of academic motivation: A person-oriented analysis. Journal of Educational Psychology 99: 734–746.

52.

Roeser

R. W.

Eccles

J. S.

Sameroff

A. J.

(1998) Academic and emotional functioning in early adolescence: Longitudinal relations, patterns, and prediction by experience in middle school. Development and Psychopathology 10: 321–352.

53.

Roeser

R. W.

Galloway

M. G.

(2002) Studying motivation to learn in early adolescence: A holistic perspective. In: Urdan

Pajares

(eds) Academic motivation of adolescents: Adolescence and education, Greenwich, CT: Information Age Publishing, pp. 331–372.

54.

Ryan

R. M.

Connell

J. P.

(1989) Perceived locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and Social Psychology 57: 749–761.

55.

Ryan

R. M.

Deci

E. L.

(2000) Self-determination theory and the facilitation of intrinsic motivation, social development and well-being. American Psychologist 55: 68–78.

56.

Schiefele

(1999) Interest and learning from text. Scientific Studies of Reading 3: 257–279.

57.

Schutz

P. A.

Pekrun

(2007) Introduction to emotion in education. In: Schutz

P. A.

Pekrun

(eds) Emotion in education, San Diego, CA: Academic Press, pp. 3–10.

58.

Shahar

Henrich

C. C.

Blatt

S. J.

Ryan

Little

T. D.

(2003) Interpersonal relatedness, self-definition, and their motivational orientation during adolescence: A theoretical and empirical integration. Developmental Psychology 39: 470–483.

59.

Skinner

E. A.

Kindermann

T. A.

Furrer

C. J.

(2009) A motivational perspective on engagement and disaffection: Conceptualization and assessment of children’s behavioral and emotional participation in academic activities in the classroom. Educational and Psychological Measurement 69: 493–525.

60.

Turner

J. C.

Meyer

D. K.

Cox

K. E.

Logan

DiCintio

& Thomas

(1998) Creating contexts for involvement in Mathematics. Journal of Educational Psychology 90: 730–745.

61.

Vansteenkiste

Lens

Dewitte

DeWitte

Deci

E. L.

(2004) The ‘why’ and ‘why not’ of job search behavior: Their relation to searching, unemployment experience and well-being. European Journal of Social Psychology 34: 345–363.

62.

Villavicencio

F. T.

Bernardo

A. B. I.

(2013) Positive academic emotions moderate the relationship between self-regulation and achievement. British Journal of Educational Psychology 83: 329–340.

63.

Ward

J. H.

(1963) Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association 58: 236–244.