Building an Evidence Base for Supporting Schools in Managing and Addressing Behavior Concerns

During the second half of this academic year, a high school asked me to work on a number of individual cases, each relating to behavioral concerns of pupils. The ‘problems’ as expressed to me by the SENCo were similar throughout the cases: pupils being defiant and non-compliant, pupils being disrespectful and/or verbally aggressive to staff and refusing to engage with punitive measures (i.e. pupils were often told to go to a type of ‘remove room’, separate from their class, and the pupils often refused). In all cases, the issues of concern appeared to have been increasing throughout the school year, as was the number of removals from class and fixed term exclusions. As such, these pupils were now seen as ‘At risk of exclusion’ (AROE).

This extended piece of writing will explore what the psychological theory and research literature can offer in informing my input in these situations, followed by an exploration of whether a sufficient evidence base can be gathered in this way. I will also reflect on how I use information and opportunities to which I have access, to support informed decision-making in offering advice and support in these contexts. First, I will explore the discourse surrounding pupil behavior in the UK, and show how such discourse shapes and is shaped by how professionals choose to respond in these situations.

With few exceptions, we all have our own experiences of education systems, and as such, the majority of us have opinions on education likely influenced by those experiences. It is easy to cherry-pick evidence to support any approach, but it appears much harder to agree on an approach that everyone can agree is justified and effective. In the following section, I will look at the research literature surrounding school behavior, and discuss how this might guide me in responding appropriately to such concerns from schools as highlighted in the introduction.

Research and Literature

School Psychologists (SPs) are in an appropriate position to work alongside schools to help address issues around pupil behavior (R. Hart, 2010) and are assumed to have knowledge and experience to support such problem-solving (Monsen & Frederickson, 2008). Given this, I have chosen to use a review of SP views on effective approaches to classroom behavior management as a framework through which to structure this section. R. Hart (2010) suggests that the literature regarding psychological approaches to classroom behavior largely falls into four categories: behavioral, psychodynamic, systemic and humanistic. This is in keeping with other reviews of the literature (Evans, Harden, & Thomas, 2004; Gulliford & Miller, 2015). I will now explore these four areas in turn, and critique how useful this categorisation is in guiding my research and informing my practice.

Behavioral Approaches

Behavioral approaches seek to reinforce desired behavior, and extinguish non-desired behavior through the use of positive or negative contingencies. These are the approaches that are most in line with traditional approaches to school discipline, and are perhaps concerned with pupil compliance above all else, with an emphasis on teacher control. Hart’s (2010) review found that behavioral approaches were the most popular for SPs to recommend. Despite their apparent popularity, behavioral approaches are relatively divisive in the education community. Lake (2004) suggests that ‘…controlling young children hinders their development of self-esteem and self-identity…The act of controlling children is the act of oppressing children (p. 14). In contrast, Bennett argues in his review that ‘Compliance is only one of several rungs on a behavioral ladder we hope all our students will climb, but it is a necessary one to achieve first’ (Bennett, 2017, p. 23). Undoubtedly, both writers have experiences and examples that reinforce their perspective.

For those who adhere to behavioral approaches and teacher control, and those who feel that these approaches are oppressive, are the end goals so different? Drawing on literature from both ends of the spectrum, both Parker, Rose, and Gilbert (2016) and Bennett (2017) speak of the importance of pupils developing self-regulation, and both imply that their ideological opponent’s approaches are not addressing this importance. As a relevant example in the field of school-based consultation, Gutkin (1999) suggests that the debate between collaborative approaches and directive approaches is a false dichotomy, and indeed that ‘“Collaboration” and “directive-ness” are not opposites of each other’ (p. 180). He proposes a different relationship between the two (Figure 1 – Directive vs Collaborative Consultation (Gutkin, 1999)), characterizing them as ‘discriminable continua’ (p. 180).
Although there are likely to remain strong differences over how to address pupil behavior, I believe Gutkin’s example could support clarification of approaches, particularly in breaking down what we mean by ‘directive’ and ‘collaborative’. Indeed, Peal’s polemic discussed earlier suggested that progressive approaches to behavior management largely involved avoiding discipline and letting pupils do whatever they liked. An understanding that a ‘collaborative’ approach is not necessarily a ‘non-directive’ approach may help to defend progressive approaches from such criticism, and support constructive real world conversations. In my experience, some education staff can adhere to the view that anything other than punitive responses to unwanted behavior is in some way “letting the child away with it”.

Are These Approaches Effective?

Another important consideration is when and with whom are these approaches successful. English education, like much across the Western world, is predominantly built upon approaches such as rules, sanctions and rewards (Parker et al., 2016). The most optimistic accounts of pupil behavior in England have suggested that pupil behavior is at least satisfactory in 99.7% of schools (Education Standards Analysis and Research Division, 2012), although others have argued that this figure ‘seriously underestimate[s] the extent to which deficits in classroom climate and poor pupil behavior are a problem in English schools’ (Haydn, 2014, p. 58). Are these areas of poor behavior due to a lack of behavioral approaches, a weakening of teacher control, as some would claim?

Some of the basic tenets of a behavioral approach seen in English schools are the strategies of using rules, praising positive behavior, ignoring negative behavior and the use of punishments (Gulliford & Miller, 2015). Gable, Hester, Rock, and Hughes (2009) performed a literature review to ascertain whether such approaches may still have a place in modern education: ‘The accumulated evidence shows that rules, praise, ignoring, and reprimands continue to represent sound classroom management strategies, but with several caveats’ (p. 202). They included the following caveats as to the positive use of these approaches:

There should be a small number of expectations (4-5), and they should be ‘necessary, reasonable, easy to understand and enforceable’ (Gable et al., 2009, p. 196).
Those expectations should be systematically taught, with clear opportunities for pupils to engage in and receive positive feedback for achieving.
Low intensity behavior may be addressed with cues and prompts, and verbal behavior reminders should have a clear message regarding both start and stop behaviors.
Praise should be paired with physical proximity and pupils must have opportunities to respond correctly and display preferred behavior.

A literature review such as this may support the stance taken by some researchers, namely that ‘It is fair to say that such a system can work … for most of the children, for most of the time. But what happens when it does not work? (Parker et al., 2016)’. Perhaps the most controversial area regarding behavioral approaches, is the concept of punishment, or reprimand.

Gable et al.’s (2009) strongest caveat was based around reprimands, where they concluded that negative, coercive approaches such as threats, nags and reprimands ‘can increase the probability that students will engage with escape-motivated behavior (e.g. defiant acts, noncompliance with teacher requests)’ (p. 201) and can ‘alienate students, undermine the integrity of the teacher/pupil relationship, and often exacerbate an already difficult situation’ (p. 201).

‘The assumption is that responding to repeated problem behavior with increasingly severe consequences will teach students that their unruly behaviors are unacceptable and will not be tolerated. Eventually, it is assumed (hoped) that the student will “get it” and stop the displays of irresponsible behaviors. Unfortunately, evidence indicates that students with the most severe problem behaviors are the least likely to be responsive to these consequences, and the intensity and frequency of their behavior is likely to get worse instead of better…’ (Sugai & Horner, 2006, p. 246).

An appropriate conclusion may be that behavioral approaches can be an appropriate foundation for providing basic structure and clear behavioral guidelines for many pupils, but they may not be sufficient (or necessary) when used reactively in response to displays of behavior that the teacher perceives as disruptive or disrespectful. When used in a preventative manner they can correlate with pro-social behavior, but when used reactively, they can be associated with teacher stress (Gulliford & Miller, 2015), as well as the fraught situations described above. Lake (2004) suggests that teachers can still receive respect and be the final decision-maker in a classroom, without resorting to a context of child compliance and teacher dominance (perhaps incorporating Figure 1). She argues that such contexts are in fact a barrier to teaching pro-social behavior skills, whereas Bennett (2017) appears to suggest that such contexts are the first step toward teaching pro-social behavior skills.

An overuse of behavioral responses may lead to increased compliance from many pupils, but at what cost? ‘It is absurd to think that the same schools that promote conflict resolution skills, character education programs and cooperative learning also support classroom environments that clearly communicate teacher dominance and child compliance’ (Lake, 2004, p. 568). Does an emphasis on compliance distract us from the goal of pupil learning? ‘The research indicates that there are times when teachers feel that they can prevent low level disruption by planning lessons around ‘control’ rather than focusing purely on pupil learning’ (Haydn, 2014, p. 52). A number of researchers (Assor, Kaplan, Kanat-Maymon, & Roth, 2005; Deci, Koestner, & Ryan, 1999; Niemiec & Ryan, 2009) have also questioned the use of extrinsic motivators in schools, arguing that an emphasis on rewards and punishments, whilst showing short-term gains in compliance, over time can actually harm a pupil’s intrinsic motivation to learn and engage.

So if behavioral approaches are not enough when faced with contexts of negative cycles of interaction with pupils, how should we proceed? I will now discuss psychodynamic approaches.

Psychodynamic Approaches

These approaches are often based on attachment theory (Ainsworth & Bowlby, 1991; Geddes, 2003), and emphasise ‘the importance to children of stable, caring and trusting relationships’ (R. Hart, 2010, p. 358). Attachment Theory proposes that a child’s early relationship with their primary caregiver leads to the development of an ‘internal working model’ that provides the child with expectations and practiced responses for interacting with others. A child who has not had their social and emotional needs met in their early years may develop an insecure internal working model, causing them to perceive the world as ‘comfortless and unpredictable and [the child] responds either by shrinking from it or doing battle with it’ (Bowlby 1973, p. 208; as quoted in Geddes (2003, p. 233)).

Whilst Attachment Theory has many critics, with some perceiving it as ‘risking deterministic views of child functioning’ (Gulliford & Miller, 2015, p. 246) and being too blaming of women (Birns, 1999), there are few who would disagree that relationships are important to children (Cooper & Jacobs, 2011). ‘For too long we have thought that vulnerable students could ‘do it if they wanted to’, when, in fact, they can only ‘do it’ within a relationship with a trusted adult’ (Bombér & Hughes, 2013, p. 3; original emphasis). Although there are many factors to consider in regard to causality, I feel it is relevant to note that in all of my ‘behavior’ cases, the young person in question has had very negative perceptions of their relationships with education staff. As an example of anecdotal evidence, one pupil reported that they only ever enjoyed school during Y6. With consent, I contacted his Y6 teacher, who told me that his positive time only came after considerable work from classroom staff to build a positive, non-judgmental relationship with the pupil above all else.

Are These Approaches Effective?

Bergin and Bergin (2009) make the claim that teachers can be far more effective if they have an understanding of attachment issues, and they state that ‘Children’s socioemotional well-being is critical to school success, and attachment is the foundation of socioemotional well-being’ (p. 141). Their review of the literature suggests that children who are labelled as having insecure parental attachments are likely to fare much worse across a number of factors relevant to a school context (academic achievement, frustration tolerance, social competence, emotional regulation and attention). An emphasis on parental attachment and historical factors beyond our control is where one can likely begin thinking deterministically. However, Geddes (2003, p. 232) believes that our understanding of a child’s attachments ‘can help us to make sense of their behavior and so open up the possibilities of change for these children and their opportunities in education’. Bergin and Bergin (2009, p. 154) also looked at teacher-pupil relationships and found that secure relationships predicted ‘greater knowledge, higher test scores, greater academic motivation, and fewer retentions or special education referrals’ than did insecure relationships, although they acknowledged the research is correlational, and their interpretation of causation from relationship to school adjustment is based on their logical and theoretical assumptions. They argue that by developing nurturing relationships, teachers can disconfirm a pupil’s insecure internal working models. Roorda, Koomen, Spilt, and Oort (2011) meta-analytic review also found positive correlations between teacher-pupil relationships and pupil engagement and achievement.

Based on their research, Bergin and Bergin (2009) make six recommendations for how teachers can improve teacher-pupil relationships:

Increase positive, warm, sensitive interactions with pupils.
Hold high expectations for pupils, and be well prepared for class.
Provide choice where possible.
Explain reasons for rules, and avoid coercive discipline.
Model kind, helpful, accepting behavior and encourage pupils to do likewise.
Use specific interventions for difficult relationships, e.g. allocate time 1:1 with the pupil, conveying ‘acceptance, interest and safety to the child’ (p. 159).

One of the biggest research areas in this field has been the use of ‘nurture groups’ in schools – designated areas for pupils with social and emotional difficulties within which positive relationships and social development are emphasised (Boxall, 2002). Research suggests these approaches can be effective in developing children’s social and emotional skills and in empowering teachers to meet these children’s needs (Sanders, 2007). However, the majority of this work is done within primary schools, and setting up a separate area for a minority of pupils can be resource intensive. Doyle (2003) describes using a successful nurture group as the basis for developing a ‘social development curriculum’ to employ the principles of nurture across the whole school, ‘to provide inspiration and starting points for mainstream teachers to use alongside other planning tools to encourage the active promotion of social development opportunities within mainstream classrooms’ (p. 258). She writes of this being instrumental in transforming the school into a positive learning environment.

Nurture groups in high settings are much less common, but research suggests that they can have an impact here also:

‘We are keeping students in mainstream education that might otherwise fail. We are giving vulnerable students the space to feel more confident in themselves and to therefore become active participants in society. This might not have been the case had we not given them that nurturing support’ (Trace (2009), pers. comm; as quoted in Colley (2009)).

As psychodynamic approaches place great stock in the relationship between pupil and teacher, so too does the next area of focus: humanistic approaches.

Humanistic Approaches

There are many approaches and theoretical strands that would fall under the umbrella of ‘humanistic’ approaches, but their commonality tends to be an emphasis on teacher-pupil relationships, and a belief that pupil motivation to engage is underpinned by basic psychological needs (R. Hart, 2010). As an example, Self Determination Theory would categorise these needs as a self-perception of competence, autonomy and relatedness (Ryan & Deci, 2000). These approaches often come under the label of a ‘learner-centred approach’ (Cornelius-White, 2007; McCombs, 2001), which states that learning must ‘support diverse learners, provide time for reflection, and offer opportunities for teachers and students to co-create practices that enhance learning, motivation, and achievement’ (McCombs, 2001, p. 186). Relative to the approaches described above, humanistic approaches are more likely to subscribe to a non-directive style, based on a belief that people have ‘inherent actualizing and organisational tendencies’ (Ryan & Deci, 2000, p. 76) that will be discouraged and impaired in controlling contexts.

Are These Approaches Effective?

Humanistic approaches tend to be used as a way to structure learning environments and teacher-pupil relationships, rather than a specific intervention to address behavior concerns, although as R. Hart (2010) found, advice based on humanistic thinking is common in SP recommendations. However, this advice is in regards to creating an environment, rather than responding reactively to a pupil’s behavior. For example, ‘Listening to the views of all (and be obviously considering them)’ (R. Hart, 2010, p. 365).

Cornelius-White (2007, p. 131) performed a meta-analysis of learner-centred teacher-pupil relationships and concluded that ‘Reduction in dropout (r = .35), disruptive behavior (r = .25), and absences (r = .25) seem to be associated with a learner-centered environment’, as well as noting positive correlations with academic outcomes and learning skills. Similarly, Shogren, Faggella-Luby, Bae, and Wehmeyer (2004) meta-analysis suggested that choice-making opportunities reduced incidents of ‘acting aggressively, engaging in noncompliance, leaving the area, exhibiting off-task behavior, and destroying property’ (Shogren et al., 2004, p. 229), a finding seen in other research (Stephanie, Cyndi, & Amy Jo, 2001).

As Ryan and Deci (2000) sagely remind us, not all school curriculum activities, or behavioral expectations, are always going to inspire intrinsic motivation in pupils. On these occasions, they once again highlight the importance of teacher-pupil relationships, and the pupil’s sense of relatedness. ‘…the primary reason people are likely to be willing to do the behaviors is that they are valued by significant others to whom they feel (or would like to feel) connected…’ (Ryan & Deci, 2000, p. 64). It is a very relevant point, that in today’s education system there is a very centralised control over curriculum delivery and a strong drive towards standards (Z. Brown & Manktelow, 2016). This structure places barriers on what strategies a teacher is likely to implement to address behavior concerns, noting as well that a teacher may be limited by whole-school behavior policies and the ethos of senior management. This brings us to our next area of exploration: systemic approaches.

Systemic Approaches

Although there is inconsistent use of the term ‘systemic’ within the literature (Pellegrini, 2009), these approaches tend to relate to taking a broader perspective on social interactions, taking into account the wider context in which the behavior occurs, emphasising the environmental factors at play and seeking to ‘address the interplay between multiple influences’ (R. Hart, 2010, p. 358). Such thinking is not necessarily in contrast with approaches discussed above, and indeed much of psychodynamic and humanistic approaches are concerned with interactions with the environment around the child (e.g. relationships with education staff). However, they do tend to move the focus away from within-child framing of the problem, and towards a view of the behavior as a product of complex interactions in which all involved parties are likely contributing (Cooper & Upton, 1990). Interventions can be addressed at the different levels of individual, classroom, school and family. Daniels and Williams (2000) discuss a systemic approach to behavior, in which the first response is an audit of whole school policies and classroom management approaches, rather than a focus on the child themselves.

Are These Approaches Effective?

Pellegrini (2009) and Evans et al. (2004) notes that there is limited published research concerning systemic approaches within educational psychology literature, but it is suggested that if SPs can successfully adopt such approaches, it would lead to ‘much creative work’ (Pellegrini, 2009, p. 282). An example of a systemic alternative to individual casework is Circle Time, which has shown success in bringing about ‘marked positive change in the behavior of children’ (Kelly, 1999, p. 44), particularly when used as a whole-class approach, although it should be noted there is limited literature in this area. Tyler and Jones (2000) presented case study evidence of a systemic approach based on teachers ‘reframing’ their views on the pupils unwanted behavior. They found that the majority of teachers, although initially sceptical, were able to employ the techniques successfully and saw positive change in their classrooms. Solution-focused approaches used in a systemic way has also been seen to support positive behavior change (Fernie & Cubeddu, 2016).

Interestingly, when R. Hart (2010) gather views on approaches recommended by SPs, and retrofitted them to align with underlying theory, they only fit under behavioral, psychodynamic and humanistic labels. None of the SP recommendations was seen as systemic. This may suggest a gap between the psychological theory of which SPs have awareness and understanding, and the actions we feel we can take according to our role. A number of barriers to working systemically have been noted, such as limited training, limited time and the view of the SP role as conducting assessments and giving advice (Pellegrini, 2009).

Having explored the four areas suggested by R. Hart (2010) and others, I will now explore whether approaching the literature in this ‘top-down’ way is an appropriate way to inform myself and gather an evidence base, or am I at risk of missing potentially important gaps of knowledge and ways of thinking, before going on to conclude what I have learned that will guide me in supporting schools. First I will explore the current context of UK public services in relation to evidence-based practice.

Is this a Sufficient Evidence Base? Is an Evidence Base Sufficient?

School services find themselves in a cultural context of increased accountability and School Psychology is no different. ‘Increasingly there is a belief that evidence-based guidelines on best practice are the cornerstone of providing professional services to the public’ (Fox, 2003, p. 93). Fox (2003) addresses these concerns within a context of SP services increasingly having to justify their unique contribution and to protect themselves (in a legal sense) from providing unwarranted advice. This approach primarily comes from a health context, in which ‘Qualitative studies (and expert opinion) come down the bottom of the research hierarchy’ (Fox, 2003, p. 93). However, the social sciences are very different from medical sciences and we cannot assume a consistency of intervention to result as can be seen in physiological contexts. Bryceland and Stam (2005) discuss how advocating empirically supported treatments (ESTs) within ethics guidelines can lead to the marginalisation of alternative approaches, that may, by their very nature, not fit the restrictive EST criteria. Although writing from a context of Clinical Psychology, there are clear parallels to be made with School Psychology, particularly when considering the use of postmodern approaches, such as constructivist, narrative, and social constructionist approaches. Shortcomings of a rigid empirical evidence-based approach include ‘the preference among constructivist practitioners for more individualized change criteria focused on personal meanings rather than observable actions’ (Bryceland & Stam, 2005, p. 140). It is with this in mind that I draw attention to the notable absence of these approaches in the research I have drawn on thus far.

Based on Hart’s (2010) article, one might assume that the approaches he discussed cover the gamut with regard to psychological approaches to behavior. Indeed, Gulliford and Miller’s (2015) chapter ‘Managing Classroom Behavior’ in an School Psychology textbook offers a similar review (behavioral, psychodynamic) with the addition of cognitive approaches. The book purports to ‘encourage students to integrate their understanding of core psychological disciplines, as well as to consider what ‘evidence based practice’ really means’ (School Psychology, 2015, back cover). If it is not contained within, as is the case with postmodern approaches, is it not a core psychological discipline? Does it not have an evidence base? Or rather, do we need to challenge our understandings of ‘evidence’ and broaden our approach to engaging with research literature. Bryceland and Stam’s (2005) critique of ESTs reminds us that cultural and professional understandings of evidence-based practice can have a real impact on how alternative approaches can be marginalised within dominant narratives and across the wider research community, a view that is shared elsewhere in the literature (Moore, 2005).

The discussion above would suggest to me that SPs need to construct their relationship with research and theory in such a way that it exposes them to alternative ideas and approaches in areas such as behavior management. An awareness of these ideas can then prompt an SP to perform research in a more precise way, specifically searching for, as an example, ‘narrative approaches’, and accessing research that may not immediately present itself when exploring the literature in a ‘top-down’ fashion, as I have done above. An appropriate way to remain informed may be through discussion and idea-sharing with other SPs. Within my Local Authority SP service, SPs have voluntarily set up interest groups around such areas as ‘narrative approaches’. I am also a member of the Associational of School Psychologists (AEP), which includes time for professional development in their half-termly meetings. Recent meetings have included SPs discussing and providing examples of how they use Video Interaction Guidance (VIG) in their work. Ensuring that I allocate time to partaking in these idea-sharing communities is a vitally important way to remain aware of approaches that may not always be part of the dominant narrative in the research literature. Narrative psychology, for example, rejects structuralism, ideas of internal states, associated deficits and attempts to reduce behaviors (Hobbs et al., 2012). Instead the focus is on how stories such as ‘behaving badly’ are socially constructed, and how other, alternative narratives are available through language. This process can allow those involved to make space between themselves and the ‘problem’, create new perspectives, and allow for narratives and opportunities more in keeping with their values (Hobbs et al., 2012).

Having reflected on the process of gathering an evidence-base, I will now consider how to integrate the information I have gathered and decide whether I consider myself informed to support the school appropriately with these cases.

Discussion

What is immediately apparent to me is that there is no ‘correct’ response in regards to situations where there is concern about a pupil’s behavior. However, there might be some ‘incorrect’ responses. I believe that I have researched enough that I am confident I would not advocate the use of behavioral, punitive responses to a pupil’s behavior. There is unlikely to be a situation where this would be effective or appropriate. That appears to be a theme of relationship and respect throughout most of the approaches discussed above, and I suspect that the lack of positive impact of coercive discipline is that it undermines a positive relationship.

Given that each case and context is unique, I would hesitate to become overly prescriptive is designing a response to ‘disruptive’ pupil behavior, but I believe my research above allows me to draw out some general principles that I would want to have as a starting point. As a basis for discussing behavior concerns with school staff, I would likely wish to have a collaborative discussion around the following principles:

Rules are useful in providing a structured foundation to the classroom, but should be reasonable, easy to understand, and young people should be involved in discussions about them.
There should be an emphasis on building and protecting positive relationships and interactions with the young person, with opportunities provided to allow this. To this end, I would reinforce the narrative psychology principle that the problem is the problem, not the person (Hobbs et al., 2012).
Within school there should be room for pupil voice, and opportunities for autonomy and choice.
Encourage awareness of whether the work provided is of a level that the pupil feels competent to engage with.
Avoid negative, coercive discipline.

If I am involved in individual cases, I would likely gather views and perspectives of the pupil too, in the understanding that this information would help guide appropriate next steps. It is during this consultation with pupils (as well as staff and family) that I would be most likely to draw on narrative approaches if appropriate, engaging people in conversations that externalise problems and allows them to see a richer description of their lives.

One concern is whether I should work pragmatically in this way of being informed by different epistemological paradigms. Hart’s (2010) review suggested that practicing SPs often don’t concern themselves with which paradigm they draw from. I believe I can be guided by the above research, but without committing to these understandings as representing an objective truth. My understanding and awareness of social constructionism need not force me to discard approaches based on more objectivist epistemologies, but rather can allow me to engage with them with a sense of criticality. ‘In this sense, social constructionism might be more accurately referred to as a meta-theory that allows experts to consider how other theories they use develop and are, or could be, applied in practice, and so offering an alternative and even a potential liberation from a wholly technical rationality’ (Moore, 2005, p. 111).