Multidimensionality of behavioural engagement: Empirical support and implications

https://doi.org/10.1016/j.ijer.2016.02.007Get rights and content

Highlights

  • The findings support the presence of distinct dimensions of behavioural engagement.

  • Five dimensions are distinguished (participation, disruptive behaviours, etc.).

  • They are facets of a same common construct, namely behavioural engagement.

  • Implications of using a global measure vs. specific dimensions are discussed.

Abstract

Behavioural engagement refers to a large range of student behaviours, differing from one study to another. What are the advantages and disadvantages of using a global measure or specific behaviours? The aim of the present study was to test the multidimensionality of the construct of behavioural engagement (presence of distinct dimensions and relevance of grouping them). Five dimensions were distinguished: participation, following instructions, withdrawal, disruptive behaviour and absenteeism (explanatory factorial analyses, Sample 1). Confirmatory factorial analyses supported the grouping of these dimensions in a common construct (Sample 2). The links between correlates and a global measure of behavioural engagement or specific dimensions were generally consistent. The global measure hid differences in relations between dimensions and some correlates. Taking the multidimensionality of behavioural engagement into account appears crucial.

Introduction

Behavioural engagement is crucial for students’ schooling. Behaviourally engaged students reach higher achievement than disengaged students. The latter are also more at risk for school drop-out (Archambault, Janosz, Fallu & Pagani, 2009; Fredricks, Blumenfeld & Paris, 2004; Rumberger & Rotermund, 2012).

However, as stressed by Lawson and Lawson (2013), research on this concept is very extensive as it reflects various interests and theoretical approaches of engagement researchers. The meaning of « behaviourally (dis) engaged students » may differ largely from one study to another. According to some authors, behaviourally engaged students in the classroom context are those who take part in lessons, notably by asking questions to the teachers, by spending time on task, etc. (Hirschfield & Gasper, 2011; Ladd, Birch, & Buhs, 1999). In other studies, it consists in following teachers’ instructions (Fall & Roberts, 2012). Behaviourally disengaged students are considered by some authors as those who annoy others during lessons, who do not follow the classroom rules, who act defiant, etc. (Ladd & Dinella, 2009; Ponitz, Rimm-Kaufman, Grimm & Curby, 2009). For others, these students do not participate in activities, think about other things during the lessons, avoid help-seeking, or are absent (Roeser, Strobel, Quihuis, 2002; Shih, 2008).

All these behaviours differ from each other to some degree. For instance, acting defiant, displaying disruptive behaviours, annoying others, etc. refer to « active behaviours » which would have a potential effect on the classroom climate and other students. To the opposite, withdrawal, avoiding help-seeking, thinking about other things, etc. consist in more passive behaviours. Moreover, they were found to be associated with different outcomes. For instance, displaying disruptive behaviours was more related to anger while withdrawal was more related to sadness (Roeser et al., 2002). These behaviours seem therefore to refer to different student experiences in school. The concept of behavioural engagement assumes that, theoretically, all these kinds of behaviours have something in common and could be considered as indicators of a same construct (Appleton, Christenson, & Furlong, 2008; Fredricks et al., 2004). However, the relevance of grouping them still need to be empirically tested. Do the indicators used reflect different sides of behavioural engagement, supporting the multidimensional nature of this concept?

The answer to those questions has crucial implications for researchers as the findings of the studies may vary according to the behaviours investigated. Moreover, one could ask what are the advantages and the disadvantages of using a global measure of behavioural engagement compared with specific behaviours. The current study addresses these issues as the answer to those questions is critical to increase our understanding of the development of student engagement (Fredricks et al., 2004).

Engagement refers to “the quality of a student’s connection or involvement with the endeavour of schooling and hence with the people, activities, goals, values, and place that compose it” (Skinner, Kindermann, & Furrer, 2009, p. 494). It is a multi-components concept generally assumed to include behavioural, emotional and cognitive engagement (Fredricks & McColskey, 2012; Wang, Willett, & Eccles, 2011). Behavioural engagement refers to the students’ participation and involvement in school activities, academic, social or extracurricular (Fredricks et al., 2004). Emotional engagement consists in students’ affective reactions to the school, teachers, academics and classmates (Fredricks et al., 2004; Skinner, Furrer, Marchand & Kindermann, 2008). It refers notably to discrete emotions (such as boredom, anxiety, sadness, happiness, etc.; Fredricks et al., 2004, Skinner and Belmont, 1993) Cognitive engagement refers to the psychological commitment in learning (e.g., use of learning – such as surface and deep processing – strategies and self-regulation strategies; Greene, Miller, Crowson, Duke, & Akey, 2004; Patrick, Ryan, & Kaplan, 2007; Wang et al., 2011).

Above those distinctions, authors stressed the need to distinguish engagement from motivation (Hospel and Galand, 2011, Skinner et al., 2008). Engagement refers to the way students feel, think and behave in classroom or at school, to the level of energy or effort they put in school. Motivation, antecedent of engagement, consists in the perceptions, beliefs, and motives that fuel those reactions.

Common in almost all definitions and measurement of engagement (Fredricks et al., 2011), behavioural engagement is a key construct. Only this component significantly predicts drop-out when all components of engagement are considered simultaneously (Archambault et al., 2009). It is a stronger predictor of long-term achievement than emotional engagement (Ladd & Dinella, 2009) and mediates the link between academic competence and emotional engagement (Li, Lerner, & Lerner, 2010).

But what is meant by “behavioural engagement”? Authors define this concept as “the behaviours students engage in that involve them in the activities of the classroom and school. (…) (It) included the social tasks of school, for example, attending classes and school, following classroom rules, interacting positively and appropriate with teachers (…)” (Finn & Zimmer, 2012; p.100). In the literature, this concept has been used to refer to a large range of behaviours and has been measured through very different ways (Fredricks et al., 2011). Some authors focused on specific students’ behaviours and, investigated separately: for instance, effort (Hughes, Luo, Kwok & Loyd, 2008), acting-out/disruptive behaviours (Finn, Pannozzo and Voelkl, 1995; Hughes et al., 2008, Roeser et al., 2002), inattentive behaviours (Finn et al., 1995), withdrawal (Roeser et al., 2002); participation (Buhs, Ladd & Herald, 2006; Gregory, Allen, Mikami, Hafen, & Pianta, 2014; Ladd et al., 1999), time on task (Gregory et al., 2014, Hirschfield and Gasper, 2011, Lan et al., 2009), compliance to classroom norms (Fall and Roberts, 2012), school avoidance (Buhs et al., 2006). Other authors mixed several types of behaviour and merged them in a global measure: participation, effort, concentration, and persistence (Hughes, Wu & West, 2011; Hoglund, 2007, Smalls, 2010); attention and compliance (Wang et al., 2011); school attendance and discipline/following the rules (Archambault et al., 2009, Li and Lerner, 2011); class attendance, engagement in other activities than class work (e.g., chatting, texting; Elffers, 2013); participation and disruptive behaviours (Ladd & Dinella, 2009); involvement, persistence, avoidance, withdrawal and participation (Shih, 2008); etc. (see also e.a. Conner and Pope, 2013, Darensbourg and Blake, 2013, Li and Lerner, 2013; Virtanen, Lerkkanen, Poikkeus & Kuorelatti, 2014). As stressed by Lawson and Lawson (2013), this diversity expressed the different interests and theoretical approaches that guide research on behavioural engagement. For instance, research on self-efficacy theory or on perceived control distinguish behaviours such as active attempts, effort, persistence from passivity, giving up, etc. (see Skinner et al., 2009).

Yet, most of the time, this diversity of behaviour has not been taken into account in past research, either in measuring of behavioural engagement (as no authors included all kind of behaviours) or in discussing the results. Beyond the differences in behaviours measured, almost all authors have operationalized this construct as a single underlying continuum, from engagement to disengagement (Archambault et al., 2009, Hughes et al., 2008, Ladd et al., 1999, Lanza and Taylor, 2010, Li et al., 2010). Behaviours cited above are quite different and authors have sometimes studied only some specific dimensions, but it seems to us that they may be organized along this continuum. Behaviours such as participation, effort, time on task may indicate the strongest engagement as students really take part in schooling by displaying such active behaviours. Compliance/following the rules is less active and may indicate a weaker engagement. Disruptive/acting-out behaviours, annoying others, etc. are indicators of disengagement. They are active and inappropriate behaviours. Withdrawal doing other activities than class work, avoiding to participate in activities, giving up, etc. are also indicators of disengagement but are more passive. Through both kinds of behaviours, students are still connected in some ways with their classroom environment. Yet, it is difficult to order these behaviours as they could indicate the same level of engagement but could refer to different ways for students to disengage from classroom activities. Finally, school avoidance, absenteeism, etc. may be the strongest indicator of disengagement as it consists in physical withdrawal from school activities. The relevance of considering all these behaviours as a same global construct is a central question. Including or merging different indicators of behavioural engagement from one study to another is not very problematic if they simply reflect variation in the intensity of engagement. However, if some relations are specific between some kinds of behaviour and some correlates, differences in indicators of behavioural engagement between studies could change the pattern of results. In the next section, we checked if consistent results were found in previous studies using different indicators of behavioural engagement.

Various aspects of student experience in classrooms and schools have been documented as correlates of behavioural engagement (e.g., achievement, motivation, etc.). The aim of this section is to compare the results of previous studies in order to see whether they are consistent or different according to the indicators of behavioural engagement used. The second case would suggest the presence of specific dimensions. To do this, we focused on some well-known correlates of behavioural engagement. The results of previous studies focusing on the links between behavioural engagement and these correlates have produced inconsistent findings. For instance, some authors found significant associations between behavioural engagement and achievement (Buhs et al., 2006, Gottfried, 2010, Ladd and Dinella, 2009; Petrides, Chamorro-Premuzic, Frederickson, & Furnham, 2005). Yet, other authors found no links between behavioural engagement and achievement (Finn and Zimmer, 2012, Hughes et al., 2008). These inconsistencies may be due to the behaviours used as indicators of engagement. Buhs et al. (2006) measured students’ participation, Gottfried (2010) measured attendance, Ladd and Dinella (2009) measured compliance with rules and directives, and Petrides et al. (2005) focused on truancy. On the other hand, Finn and Zimmer (2012) and Hughes et al. (2008) focused on disruptive behaviours. Moreover, among the behaviours significantly related to achievement cited above, it appears that the strength of the associations vary according to the specific behaviours investigated within the same study. For instance, classroom participation is more strongly related to achievement than school avoidance (Buhs et al., 2006). Unexcused absences were more strongly associated with low achievement than excused absences (Gottfried, 2009). Therefore, it may be that some aspects of behavioural engagement such as participation and withdrawal are more crucial for achievement than others (i.e. attendance and disruptive behaviours).

Regarding motivation, self-efficacy was negatively related to withdrawal and giving-up (Pintrich, 1999, Roeser et al., 2002, Wang, 2005), while perceived task value was negatively related with disruptive behaviours (Borders, Earleywine, & Huey, 2004; Roeser et al., 2002). Effort, persistence and active participation were positively linked with both motivational variables (Cox and Whaley, 2004, Linnenbrink and Pintrich, 2003, Patrick et al., 2007).

Regarding the relations between dimensions of behavioural engagement and others components of engagement (emotional and cognitive engagement), studies also showed some differences. As stressed in “Engagement as multi-components” section, emotional engagement refers to discrete emotions (Fredricks et al., 2004). Previous research showed negative relations between effort and negative emotions (Dettmers et al., 2011), between participation and boredom (2005Ainley, Corrigan, & Richardson, 2005; Pekrun, Goetz, Daniels, Stupnisky, & Perry, 2010). Persistence was mainly associated with interest (Ainley et al., 2005). Acting-out behaviours were more linked to anger, while withdrawal was more related to sadness (Roeser et al., 2002) or anxiety (Wang, 2005). Skipping classes was documented to be positively linked with anger and a combined measure of sadness and hopelessness (Roeser & Eccles, 1998). As described previously, cognitive engagement refers to the use of learning and self-regulation strategies (Greene et al., 2004). Students who used better self-regulated strategies spent less time on the task (van Den Hurk, 2006). Students’ attention and problem behaviours were negatively linked (Taylor & Lopez, 2005).

These results show that students’ behaviours may reflect specific students’ experiences at school as they are linked with specific outcomes (for instance, some students who display acting-out behaviours may be more angry while others who display more withdrawal behaviours may feel more sadness). It raises some key questions. Are all these behaviours the reflection of a same underlying construct? What are the consequences of using different ways to measure behavioural engagement (e.g., different kinds of behaviours vs. a global scale)? If the behaviours measured vary from one study to another, inconsistent findings may emerge.

Section snippets

Research questions and hypotheses

The variations in the content of the measures used to assess behavioural engagement question the construct validity of this concept. The goal of this paper was not to propose a new measure of behavioural engagement, but to address the concerns raised above and to contribute to the improvement of the measurement of this construct. The first aim of the present paper is to investigate the multidimensional nature of behavioural engagement by testing if specific dimensions of behavioural engagement

Study 1

Study 1 aimed at testing the multidimensional nature of behavioural engagement, as a step towards testing its construct validity. We tested if, among the large variety of indicators used in previous research, some dimensions of behaviour emerged.

Study 2

First, Study 2 aimed at testing the construct validity of the measure of behavioural engagement by examining if the presence of specific behaviours highlighted in Study 1 could be replicated and related to a higher-order construct, behavioural engagement. Second, it aimed at investigating how each of these dimensions, as well as a global measure of behavioural engagement, were related to expected correlates: emotional and cognitive engagement, task-value, self-efficacy, achievement and gender.

General discussion

The variety of behaviours used in previous studies under the label “behavioural engagement” challenged the relevance of considering all these behaviours as indicators of a same underlying construct, namely behavioural engagement (Aim 1). It also raised questions related to the implications of using different measures of this concept (specific behaviours vs. a global measure; Aim 2).

Our results extend prior research on behavioural engagement by bringing empirical support for its multidimensional

Acknowledgements

This research was supported by a grant from the National Fund for Scientific Research (F.R.S.-FNRS) of the Fédération Wallonie-Bruxelles awarded to the first author.

The authors want to thank Mostafa Azahaf, Noémie Baudoin, Julie De Meyer, Céline Georges, Anne Gilet, Arnaud Maréchal, Abdel Oulad Ben Taïb, and Sandra Robinet for their help in data collection. They also thank Noémie Baudoin for her helpful comments on a previous version of this article.

References (69)

  • M.-T. Wang et al.

    The assessment of school engagement: examining dimensionality and measurement invariance by gender and race/ethnicity

    Journal of School Psychology

    (2011)
  • J.J. Appleton et al.

    Student engagement with school: critical conceptual and methodological issues of the construct

    Psychology in the Schools

    (2008)
  • N.J. Blunch

    Introduction to structural equation modelling using SPSS and AMOS

    (2008)
  • A. Borders et al.

    Predicting problem behaviors with multiple expectancies: expanding expectancy-value theory

    Adolescence

    (2004)
  • E.S. Buhs et al.

    Peer exclusion and victimization: processes that mediate the relation between peer group rejection and children’s classroom engagement and achievement?

    Journal of Educational Psychology

    (2006)
  • J. Cohen

    Statistical power analysis for the behavioral sciences

    (1988)
  • J.O. Conner et al.

    Not just robo-students: why full engagement matters and how schools can promote it

    Journal of Youth and Adolescence

    (2013)
  • A.E. Cox et al.

    The influence of task value, expectancies for success, and identity on athletes’ achievement behaviors

    Journal of Applied Sport Psychology

    (2004)
  • A.M. Darensbourg et al.

    Predictors of achievement in African American students at risk for academic failure: the roles of achievement values and behavioral engagement

    Psychology in the Schools

    (2013)
  • J.S. Eccles et al.

    In the mind of the actor: the structure of adolescents’ achievement task values and expectancy-related beliefs

    Personality and Social Psychology Bulletin

    (1995)
  • L. Elffers

    Staying on track: behavioral engagement of at-risk and non-at-risk students in post-secondary vocational education

    European Journal of Psychology of Education

    (2013)
  • J.D. Finn et al.

    Disruptive and inattentive-withdrawn behavior and achievement among fourth graders

    The Elementary School Journal

    (1995)
  • J.D. Finn et al.

    Student engagement: what is it? Why does it matter?

  • J.A. Fredricks et al.

    School engagement: potential of the concept, state of the evidence

    Review of Educational Research

    (2004)
  • J. Fredricks et al.

    Measuring student engagement in upper elementary through high school: a description of 21 instruments

    (2011)
  • J.A. Fredricks et al.

    The Measurement of student engagement: a comparative analysis of various methods and student self-report instruments

  • B. Galand

    Le rôle du contexte scolaire et de la démotivation dans l’absentéisme des élèves [The role of school context and motivation in student absenteeism]

    Revue des sciences de l’éducation

    (2004)
  • B. Galand et al.

    Style motivationnel des élèves du secondaire: Développement d’un instrument de mesure et relations avec d’autres variables pédagogiques [Secondary school student's motivational style: developing a measurement instrument and relationships with other educational variables]

    Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement

    (2002)
  • B. Galand et al.

    L’école telle qu’ils la voient: Validation d’une mesure des perceptions du contexte scolaire par les élèves du secondaire [School as they see it: Validation of a measure assessing student perceptions of school context]

    Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement

    (2005)
  • B. Galand et al.

    Engineering students’ self-regulation, study strategies, and motivational believes in traditional and problem-based curricula

    International Journal of Engineering Education

    (2010)
  • M.A. Gottfried

    Excused versus unexcused: how student absences in elementary school affect academic achievement

    Educational Evaluation and Policy Analysis

    (2009)
  • M.A. Gottfried

    Evaluating the relationship between student attendance and achievement in urban elementary and middle schools: an instrumental variables approach

    American Educational Research Journal

    (2010)
  • A. Gregory et al.

    Effects of a professional development program on behavioral engagement of students in middle and high school

    Psychology in the Schools

    (2014)
  • P.J. Hirschfield et al.

    The relationship between school engagement and delinquency in late childhood and early adolescence

    Journal of Youth and Adolescence

    (2011)
  • Cited by (45)

    • Need-supportive teaching and student engagement in the classroom: Comparing the additive, synergistic, and global contributions

      2021, Learning and Instruction
      Citation Excerpt :

      Students rated their behavioral, emotional, and cognitive engagement in math lessons on a scale ranging from 0 (never) to 4 (very often). Behavioral engagement was measured using a 10-item scale (7th grade α = 0.85; 8th grade α = 0.86; Hospel, Galand, & Janosz, 2016). The measure included items tapping into student compliance, participation, and withdrawal (e.g., “When the teacher asks a question during a lesson, I try answering.”).

    View all citing articles on Scopus
    View full text