This content is historic and has been archived for reference only. It does not necessarily represent the author's current thinking.

Challenges to higher education staff teaching online in handling disruptive behaviour

Alistair Knock // December 2009

Back to index


Increasing use of online learning environments may in turn mean an increasing incidence of disruptive behaviour online. There is much material discussing the advantages and disadvantages of online learning environments, and looking at the barriers faced and adjustments required for disabled students, but there is little research investigating the association between online disruptive behaviour, its causes, the declaration of disabilities, and staff attitudes to such disruptive behaviour. This study explores current staff attitudes to disruptive behaviour: whether disruptive behaviour in higher education is felt to be more or less prevalent online than in traditional settings; identification of the causes of disruptive behaviour; and how staff recognise and respond to disruptive behaviour. Through the available literature and results from a small scale survey, the study then suggests some common features and approaches in successfully dealing with disruptive behaviour.


Thanks to all those who contributed personal, emotional experiences to the study, and to the staff on the MSc programme for their support and innovation.

Table of Contents

  1. Abstract
  2. Acknowledgements
  3. Introduction
  4. Context and Literature
    1. Forms of disruption online
    2. Flaming
    3. Trolling
    4. Antagonism and polarisation
    5. Lurking
    6. Disability discrimination
    7. Models of disability
    8. E-learning As Enabling
    9. Academic Withdrawal
    10. Stigma
    11. Accessibility
    12. Staff development and managing disruptive behaviour
  5. Research Study
    1. Research design and method
    2. Findings
    3. Emotional themes
    4. Location
    5. Effectiveness of response
    6. Cause of incident
    7. Breadth of respondent experience
  6. Advice and tools
    1. What to consider when constructing the learning environment
    2. What to consider when teaching in the learning environment
    3. What to consider when dealing with a difficult situation
    4. How to reflect and review
    5. Using other resources
  7. Conclusion
  8. References
  9. Endnotes

Illustration Index

Index of Tables


Online computer-mediated communication presents an alternative option to traditional tutorial groups for both learners and teachers. Unlike traditional fixed time and space tutorials, online tutorials can be synchronous (e.g. chat sessions) and asynchronous (post-based discussion forums) and can be actively moderated or not: the tutor's presence is not necessarily a requisite. The online setting can present challenges for both learner and teacher in that a new etiquette must be assumed and applied, in the absence of physical cues such as tone of voice, facial and body language which can assist in facilitating a traditional tutorial. Concepts such as 'flaming' and 'trolling', which represent disruptive behaviour online, have become more recognised by visitors and moderators of online discussion forums on the web, to the extent that many large sites have established rules and procedures for taking action against disruptive behaviour. Some sites such as Wikipedia have been regarded as overly aggressive and censorial in their reactions to perceived 'breaches' of etiquette. Equally, it is important that the site provider takes responsibility for the content provision they are facilitating, and that there is sufficiently frequent moderation and checking of content. There is a fine line between protecting the reputation of the site and its content, and alienating and offending users by reacting too strongly.

E-learning, either as a way of complementing traditional provision in the form of blended learning, or as a complete package for facilitating distance learning, is growing in spread and importance across the United Kingdom higher education sector. Even five years ago, Britain and Liber (2004) found that 'uptake of virtual learning environments across HE is now practically universal', and separately, Browne and Jenkins (2006) put the figure at 86% of UK higher educational institutions. In an e-learning environment, the parameters are different in that students are expected to be positively motivated to achieving their pedagogical goals, and so tutors may assume they are less likely to encounter disruptive behaviour than in the wider setting of the internet. However, the causes of disruptive behaviour can often be related to mental health problems, of which there is a growing disclosure by students attending higher education in the United Kingdom, with a growth in real terms of 82% between 2003 and 2008 (UCAS, 2009). While this growth is significant, the number of students who have been diagnosed or regard themselves as having mental health problems is widely regarded to be far higher than those who formally disclose mental health problems . This is often a result of fear and doubt in the student as to how the disclosure will be received, particularly in a judgemental and competitive situation such as making an application to study at university, and the associated stigma from students and staff alike which is often formed through a lack of understanding of the causes and effects of different mental health problems. Stigma, though, is not the only reason for non-disclosure - Prior et al (2002) found that differing identification and classification of mental health problems across professions and different sections of society led to variability in disclosure.

This growth in disclosure and prevalence of mental health problems, which is part of an increasing diversity in higher education, coupled with assumptions that an e-learning - and particularly distance learning - environment may isolate students and exacerbate or amplify their social difficulties, means it is increasingly likely that tutors will have to deal with disruptive students whose behaviour impacts on both their own performance and that of their peers, in a technological setting which may not lend itself easily to dispute conciliation and avoidance. Consequently, this research sets out to investigate the following questions through an examination of the literature and through a small-scale research study:

This study distinguishes between two spheres of learning activity: online and offline. Online includes learning activity which is conducted through the internet, including virtual learning environments, email, websites, internet relay chat, and messaging services. Offline in this sense refers to 'traditional' settings, which include face-to-face communication in lectures, seminars, individual one-to-ones, and also telephone calls between students and between students and staff. The study is constrained to identifying disruptive behaviour in students and the impact this has on other staff and students; it does not intend to examine the impact of disruptive behaviour amongst educational staff.

In the context of this study, disruptive behaviour is defined as any deliberate action, whether intended to have negative consequences or not, which results in other planned actions being impeded and affected, or which results in a negative effect on the emotions of another person. Examples of direct offline disruptive behaviour include:

In addition, examples of indirect offline disruptive behaviour could include persistently remaining silent in seminars where every student is expected to and is encouraged to participate, or continuing to pursue a line of argument which had been already addressed in the discussion, and which had now moved on. It is more difficult to codify indirect behaviour, since it is typically the effect that several related actions have on other people which establish the behaviour, rather than a specific and tangible action.

Context and Literature

Forms of disruption online

Online communications are usually asynchronous and so allow time to elapse between each engagement. In some cases, this delay can mean that direct hostility may sometimes be more measured than a face-to-face argument. It is more likely however that misinterpretation of the message, the amplifying effect of the empty pause in discussion (where antagonism in a message received some time ago still has an effect on the recipient), and the anonymity or impersonality of online communications will lead to some anxiety or frustration for the recipients. The level of disruption depends on factors such as the size, proximity and authority of the 'audience' - a large group of people may dissuade a disaffected participant from voicing their complaint, or conversely, group dynamics and the deindividuation resulting from the 'herd instinct' may instead encourage other participants to join in with the antagonist, even where their initial strength of feeling about the complaint may be minimal. (Kraus et al, 2004; Williams, 2006) In the online world, levels of authority can also be perceived to be flattened by the more open methods of communication, such that participants have less respect for sovereign authority and the institutional distinction between learner and teacher. (O'Neil, 2009; Shirky, 2008)


Since most communication online continues to use asynchronous text through email and message board/blog postings, the most common forms of disruptive behaviour are those termed 'flaming' and 'trolling'. While definitions of these terms vary, flaming is generally held to be an action deliberately intended to provoke, alarm, or offend other people. Examples include swearing, use of capital letters (the online equivalent of shouting), inclusion of offensive material in quotes or images, and direct ad hominem attacks on others. A 'flame', as an incident of flaming is known, is normally easy to spot even without prior experience online because the intent to offend is strong. The incidence of flaming online as opposed to similar behaviour offline differs, likely due to psychological and scenario-linked factors discussed in more detail later.


'Trolling', on the other hand, is again intentional but rather than being overtly hostile, trolling is a more subtle form of antagonism, usually intended to generate feelings of frustration and to mislead less experienced users. This may mean intentionally asking questions that have already been answered many times on the list, or posting a message that they know will antagonise some participants/groups into continuing the discussion, perhaps ending up in a full-blown flame-war. Trolling is harder to monitor, because the poster will purposefully veil their intent and create confusion to the extent that it is difficult — particularly for new and inexperienced users — to know whether the post is genuine or not (Herring et al, 2002; Schwartz, 2008). Even at the 'end' of a trolling effort, it may still only be clear to the original troll that time has been wasted on a futile venture, but clearly in a time-limited environment such as study on an educational course, there is an knock-on effect that this can have on other students.

Antagonism and polarisation

In some cases, the unintended outcome of online antagonism can be tragic. Megan Meier, a 13 year old from Missouri, committed suicide after her online friendship with a fictitious user on the popular MySpace social network turned sour and the user resorted to bullying. According to local media (St. Louis Today, 2007), Meier had been diagnosed with attention deficit (hyperactivity) disorder (ADHD) and depression, and took her own life after being deceived repeatedly by a number of people known to Meier who jointly assumed a false MySpace user identity in retaliation to an earlier incident. In part due to this and similar incidents, many US states have laws which legislate against so-called 'cyberbullying', and in 2009, bill HR1966 was sent to Congress for consideration, with the aim of making cyberbullying a federal crime.

A more subtle, sometimes unintentional but equally insidious, form of disruption can be found in the 'brand wars' and in situations which lead to what Noelle-Neumann (1984) calls the 'spiral of silence'. These lead to a polarisation of views, which eventually can exclude and deny other valid viewpoints. Brand wars are typically fought between proponents of the products of commercial brands (e.g. Reebok and Nike, Microsoft and Apple) but have a parallel in the support of sociological, academic, ethical and political perspectives, such that participants tend to end up in two opposing camps and the debate becomes about volume rather than quality. This can be seen commonly in online political debates about capitalism and communism, where debate about the specific issue essentially ceases once the affiliation of each participant is known, and the debate descends into personal attacks - examples include the BBC's 'Have Your Say' and The Guardian's 'Comment Is Free' sections. In a guide to posting online, Sunny Hundal (2008) explains 'I start from the assumption that most readers are going to be somewhat hostile or won't read it through my worldview'. This hostile behaviour is disruptive because a bystander who is yet to participate may now feel threatened by the strength of the viewpoints held, and so is less likely to contribute their own thoughts, whether contradictory or aligned. The effect is that the discussion becomes one-way. This culminates in a spiral of silence, where an individual discovers that opinions are so strongly held by those around them that they are afraid to air an alternative view for fear of being insulted or attacked, and also that the degree of cognitive effort they would have to make for their viewpoint to gain any traction against the common view seems too great. The individual then typically says nothing, which helps to homogenise the argument, such that the next (perhaps as yet undecided) participant who arrives at the debate is even less likely to contribute, having assumed that those already involved have chosen a final position. Succinctly put, 'the crux of the spiral of silence is that people believe consciously or subconsciously that the expression of unpopular opinions will lead to negative repercussions' (Wikipedia, 2008). This is particularly disruptive to silent participants, since there are barriers to contributing and learning from the resulting discussion; it is also unconstructive for those with strongly held viewpoints, since the diversity of their knowledge lessens over time, leading to entrenched views. In their study of deviant opinions, Miller and Morrison (2009) conclude that 'Over time … the average group member's attitude can be expected to converge on the prototypical group attitude, as the latter will be more likely to be expressed and hence more likely to be taken as the average attitude'.

The net effect of these activities is to dissuade participants from participating. In an online educational environment where increasing credence is placed on the discourse, rather than the 'reading' of a course's material, a naïve adjudicator might therefore see inaction by a participant as a weakness on the participant's part, rather than an effect of the discourse itself.

It may be reasonable to assume that the incidence of disruptive behaviour in an educational setting may be somewhat less than in a wider public setting, due to the narrower sociological composition of the group, and the purposeful act of engaging in a learning setting. While this may be true, it is less clear whether the distinction between offline and online environments means that such behaviour will happen more or less frequently and with a different intensity online. For example, a hypothetical face-to-face argument about a contentious issue could quickly escalate into irrelevant and hostile attacks as a result of the synchronicity of each participant's response and the narrow time in which to consider a measured response, without taking so long as to leave the impression that the argument has been won by the opponent. The asynchronicity of an email list or discussion forum — where sometimes days elapse between responses — could mean that participants are more measured in their approach, and are more able to have an intellectual rather than emotional discussion about the issues. In many cases though, the opposite is true: the additional time offered by asynchronicity may allow participants to consider the minute detail of the opposition's argument, and to dissect it more thoroughly than would be possible in a flowing discussion. That this small, and often distantly related, point can then consume the next phase of the discussion is a characteristic much more common online, and is characterised in a tongue-in-cheek image showing the cycle of online debate (illustration 1).

New York Times cartoon diagram of a blog, from original thesis through reservations, disagreements, inpugning characters, criticism and denounciation, through to settlement, praise and ending in general ennui

Illustration 1: Diagram Of A Blog. Source: Paula Scher, The New York Times Company, 2007

Although satirical, this image shows the numerous flash points where disruption can arise, often due to misinterpretation/ misrepresentation rather than a deficiency in the argument itself.

In a sense, some forms of disruptive behaviour are relatively easy to moderate, but this depends on the perspective of the moderator as set against those of the users. What is offensive to one person may not be to the other, or the concept and acceptance threshold of offensiveness may vary within one person from day to day, and as Lea puts it, is 'radically context-dependent' (Lea et al., 1992; O'Sullivan and Flanagin, 2003b).


A form of behaviour not usually classed as actively disruptive, but which can nevertheless have an influence on group dynamics online, is lurking. A lurker is a read-only participant: someone who is present in the background of a group of users, who is consuming content and not visibly contributing to it, whether in the form of discussion boards, blog entries, wiki maintenance, or chat room sessions. Where a group has a high proportion of lurkers, it is sometimes felt that there is greater risk of the actively participating students overcompensating for this, and therefore dominating the conversation. This can cause lurkers to be even less likely to participate, with inexperienced online users' hesitancy to post amplified by the participants' overcompensation (Beaudoin, 2002). While the group dynamic may be influenced by the levels of participation, this does not however equate to a disrupted learning experience for participants. While there is evidence that the benefits of online learning are increased when learners function as a group and learn from and with one another, many students are comfortable with and better equipped to learn by consumption of content (Anagnostopoulou, Parmar and Priego-Hernandez, 2009). That the online group as a whole is dysfunctional does not prevent these learners from studying independently through engaging with the course readings, conducting their own research and using ideas from the group discussion. Fritsch (1997) describes this notion as 'witness learners' - that lurkers are passive in terms of their contributions to the pool does not preclude them from being active observers; as in the offline world, some people prefer to listen rather than talk. Indeed, in Beaudoin's study of 'invisible' online students (2002), 75% of those students identifed as being lurkers said that either they preferred to read rather than write, or that what they'd intended to write had already been said. In some cases, students can become forced lurkers: they regard their initial attempts to participate as unsuccessful, perhaps due to genuine or perceived lack of feedback from peers, such that although they had intended to participate, their early experiences are sufficiently unsatisfactory to prompt a return to lurking (Dickey, 2004). Regardless of the cause, Anagnostopoulou et al (2009) recommend that 'it is important to acknowledge lurking as both a personal preference and a legitimate method for learning'.

The origin of disruptive behaviour can sometimes be easy to identify. Prior knowledge of other factors in that person's personal life, suggestions from other colleagues that there is a likelihood that behaviour they have experienced might be repeated, or external influences such as current affairs relating to particular demographics, can all provide a guide to help explain specific instances of disruptive behaviour. Often though, these contributing factors alone are insufficient to explain the behaviour experienced, and may indicate specific mental health problems the individual is experiencing. The incident in question may be the first time these problems have presented themselves, both to the individual and to those affected by it. It is therefore important for the moderator to bear this in mind when approaching and handling the situation, both in terms of the safety of those involved, and the rights of the individual under scrutiny.

Disability discrimination

Various countries have equivalents to the United Kingdom's Disability Discrimination Act 1995 (DDA), which in 2002 was extended to cover education, all of which at their bases:

In the UK, the DDA defines a disabled person as a person with 'a physical or mental impairment which has a substantial and long-term adverse effect on his ability to carry out normal day-to-day activities' (EHRC, 2009). The Act then defines two main types of disability discrimination:

  1. unjustified less favourable treatment for a reason related to a person's disability; and
  2. unjustified failure to take reasonable steps.

Public attention focused closely in the latter half of the twentieth century on physical disabilities, such as mobility impairments, blindness, and hearing impairments, and much improvement has been made both in terms of access to buildings and services, and in public awareness of the fact that disabled people are no different than non-disabled people, other than the mode and means by which they access information, venues and services. The realm of cognitive impairments has had less attention and resource allocated to it — Stanley and Manthorpe (2000) use the term 'invisible disability' to describe the broad attitude in education to mental health problems - partly as the discrimination is mostly related to awareness rather than physical barriers, but also as research being conducted in both areas of specific learning difficulties and mental health problems in education (though not necessarily higher education; see Taylor et al, 2008) has grown significantly in scale in recent years. Despite efforts by mental health organisations, and the impact of legislation, a stigma is still strongly felt in many organisations and the wider public around the causes and effects of mental health problems, particularly amongst employees and managers who may regard their colleague's mental health problems as a deficiency (Robinson, Martin and Thompson, 2007; Thornicroft, 2006). This is often demonstrated through ignorance in identifying the existence of a mental health problem, and subsequently in the consequences felt by the individual of any official actions, which are often unconstructive.

Models of disability

A key distinction in the modern approach to disability is in separating the two main models used: medical and social. The medical model regards the disabled person and the specific characteristics of their disability as the cause for any barriers that they encounter - for example, the reason the person cannot get to the third floor of a building is because their limited leg movement means they cannot climb stairs, or the reason a person cannot participate in a tutorial is because their hearing impairment means they cannot hear questions. In contrast to this negative model, the social model views the environment itself, negative attitudes, and systemic factors as the primary barriers to a disabled person's activities. In this view, the reason the person cannot get to the third floor is because the stairs are not designed to accommodate wheeled vehicles, and because the building was designed without a lift or other level access to the third floor. The reason the deaf person cannot participate in the tutorial is because it is ill-designed: it doesn't provide written notes or facilitate the presence of a sign language interpreter.

The social model was originally proposed by Mike Oliver in 1983; 25 years later, the concept is beginning to take hold in higher education but is far from embedded. A key barrier to acceptance is the apparent conflict with 'academic standards': the idea that students should be assessed on their abilities, and distinguished between by awarding grades objectively based on uniform marking standards. For some staff, this necessity to have standard benchmarks against which to measure student performance conflicts with the Disability Discrimination Act's requirement to make reasonable adjustments for disabled students, particularly with more esoteric adjustment requests where the adjustment would appear to be more favourable to the student than necessary. The guidance issued by the UK government's Equality and Human Rights Commission (2007) in the Code of Practice for Providers of Post-16 Education and Related Services preserves these benchmarks in the concept of 'competence standards', which are:

an academic, medical, or other standard applied by or on behalf of an education provider for the purpose of determining whether or not a person has a particular level of competence or ability

However, if a particular measure can be legitimately shown to be a competence standard, the education provider still has a duty to make reasonable adjustments for a student when demonstrating attainment of this competence standard. (Simpson, 2009) So, if one of the competence standards in a political debate course was the 'ability to discuss and debate ideas lucidly and with rapidity', a disabled student or student with mental health problems would need to demonstrate this ability; they would not, however, be required to do this in a traditional face-to-face aural debate scenario, if it was shown that this scenario would put them at a disadvantage - for example, because of anxiety over participating vocally in group settings. Instead, the competence standard could be equally demonstrated through written discourse, asynchronously, perhaps online.

Once known about, an individual's disability, specific learning difficulty, or mental health problem cannot be used to discriminate against them in an educational institution. The low level of disclosure of mental health problems though, means that many students already receive a differential level of treatment and while this may not necessarily be discriminatory under the law, it is in most educational provider's interests to create an accessible environment by default, rather than having to make specific reactive adjustments for individual students. (Burgstahler et al, 2004)

E-learning As Enabling

It is important to note that while there are documented accessibility barriers within many e-learning environments, notably for students with visual and auditory impairments, some disabled students can benefit hugely from the provision of learning environments online. For many students, the perceived benefits of a traditional bricks and mortar university can be barriers to their participation in higher education, in ways not limited to physical access. For instance, students with agoraphobia might prefer to work in quiet environments that they control and hence don't attend university. It may be more suitable for them to study in an online environment where they can define their own hours of attendance and can screen out activities which they feel uncomfortable with. Phipps and Kelly (2006) outline a holistic approach to encouraging accessibility of e-learning environments, where each learner's needs are the driver to generating an accessible environment, in contrast to the accessibility-by-rote result of standards and policies such as the Web Content Accessibility Guidelines, which are intended to cover the majority, but by extension still exclude the needs of the minority. For students with anxiety issues, difficulties adjusting to fast-paced social environments, and for students who learn best outwith a daily 9-5 routine, the asynchronicity and 'distance' of e-learning can open up new avenues for participation in what is — for them — an inclusive learning environment.

Different environment configurations can enable participation in different ways: although synchronous activities may be a particular barrier for students with dyslexia (Woodfine et al, 2008), it is conceivable that the fast-paced, collaborative nature of such an activity, where all participants contribute smaller parts to the whole rather than each participant being expected to develop their own independent contribution, could provide an enabling environment for students with low self-esteem, both educationally and socially. Indeed, Sapp and Simon (2005), adding to similar views expressed by Overbaugh and Lin (2006) and Palloff and Pratt (1999), point out that the benefits can extend wider than the field of disability:

On the other hand, distance learning may provide opportunities for some writing students to compete on more equal footing than they experience in face-to-face classrooms. Introverted or shy students, for example, may actually feel empowered by using a mediator such as technology. The invisibility of students in online writing courses can break down barriers to learning such as racism, sexism, homophobia, and the social obstacles related to popularity and fashionability.

Academic Withdrawal

Disruptive behaviour is not solely an issue for the direct recipient and those indirectly affected in the surrounding group. Since disruptive behaviour is often triggered by feelings of disaffection in the source student, their own emotional state can be worsened by their behaviour and its consequences. This can often lead to diminished participation in the course, or to formal withdrawal from the programme and from higher education. The issue of withdrawal and its mitigation is of current interest in UK higher education, and new approaches to student retention are being developed (Waggoner and Goldman, 2005). As retention of non-failing students benefits both the student and the institution, it is clearly important for the causes of difficult behaviour, including those related to mental health problems, to be identified early and, where possible, addressed before a tipping point is reached. Parmar and Trotter (2004) indicate that the first 6-8 weeks are the key decision-making period for students who are likely to withdraw, giving a relatively narrow window for intervention by course staff, particularly given the tendency for a prolonged 'introductory' period in e-learning to allow students to familiarise themselves with the tools used. That withdrawal is a problem more likely to arise as a result of mental health problems was noted in a US study by Kessler, Foster and Saunders (1995), which found that 'eighty-six percent of students with mental illnesses withdraw from college before completing their degree, a figure that is much higher than the approximately 37% withdrawal rate for the general student population.'


Findings by Salzer et al (2008) in their survey of the use of reasonable adjustments in the United States indicate that stigma remains a major barrier to starting discussions with academic staff: 56% of students with mental health problems were embarrassed and afraid of being stigmatized, and 42% of this group then found that staff were 'uncooperative or unreceptive'. Conversely, anecdotal evidence from online courses run by a different institution indicates that in some cases, academic staff may go beyond the adjustments recommended by central services and 'over-accommodate' based on other difficulties the student might be having, which are not directly related to their disability (Lindsay, 2004). While the quality of adjustments provided may have some bearing on the success of the student, it is not the only factor, and disabled students often overachieve, in comparison to their peer group, because they have strong self-determination and commitment to successfully completing their studies (Brockelman, 2009).


Literature directly related to the field of accessibility also often overlooks mental health problems. A 2003 report on the accessibility of virtual learning environments makes little reference to mental health problems, other than in defining disability (Dunn, 2003). Previous work by this author as recently as 2005 covers some of the negative online effects that can be provoked through mental health problems, but does not identify this as a specific cause. (Knock, 2006)

The otherwise-excellent Access All Areas report by TechDis (2002) mentions mental health problems in passing, but concentrates more on the technical aspects of ensuring that assistive technologies can access information, and that content is displayed in an accessible format. Bohman and Anderson (2005) are among few web accessibility promoters who have attempted to address accessibility and cognitive impairments, but their narrow definition does not cover the majority of mental health problems. As with the delay in identifying and implementing solutions for students with specific learning difficulties such as simpler language construction, visual aids, and consistent typography, the management of online environments is not covered in accessibility literature to any significant extent. While disruptive behaviour may be regarded as a standard feature of e-learning, not necessarily related to mental health, it nevertheless is an area where the accessibility of the environment can play a key role in further inflaming situations. In Disabled students in higher education, Riddell et al (2006) write:

For people with mental health difficulties it is essential to examine how the environment is creating or exacerbating difficulties, as well as looking at ways to support people to deal effectively with the environment within which they are operating.

Staff development and managing disruptive behaviour

A key step in changing perceptions and reducing stigma is to provide increased training for staff about the causes of disruptive behaviour, mental health problems, disability in general, and approaches to creating a more inclusive learning environment. Collins (2005) reports that '42% reported limited knowledge or experience, and 28% had no knowledge or experience' in formal supported education in relation to mental health problems.

In general, discussion of mental health problems in literature specific to e-learning and online communication is infrequent. Although the social and cultural aspects of online communication are implicit through Salmon's 2004 book E-Moderating: The Key to Teaching and Learning Online, the section on users with disabilities makes reference only to visual and physical impairments, and dyslexia, not to mental health problems or other cognitive impairments.

Less information appears to be available on the perceptions of staff when working with students who demonstrate disruptive behaviour. While it is assumed that online mediators are aware of the appropriate tactics to use in different settings in order to pacify students and handle volatile situations such that students are less affected, it is less clear what the subsequent impact on the tutor is in having to intervene and assess their next steps. For example, having diffused a difficult situation, does the tutor feel their work has been done, or does the tutor harbour concerns about the student's state of mental health and worry about what the student may do next? (Hara & Kling, 2000) Guidance issued by the University of California, Berkeley, suggests:

Certain signals that distressed students give out may go unnoticed for a variety of reasons. And even when we do notice them, it can be very difficult to intervene. We may feel we are 'in over our heads,' or we may have competing concerns, such as other students waiting to see us. It is important to know that it is quite likely that the problem will not go away unless there is an intervention. Part of a good intervention requires knowing how to act during these incidents and what resources to call upon.

Daniels, in 'How To Manage Disruptive Behavior in Inclusive Classrooms' (1998), lists ten questions she suggests that tutors ask themselves after a disruptive incident which may have been related to a disability, specific learning difficulty or mental health problem, and where possible, prior to reacting:

  1. Could this misbehavior be a result of inappropriate curriculum or teaching strategies?
  2. Could this misbehavior be a result of the student's inability to understand the concepts being taught?
  3. Could this misbehavior be an underlying result of the student's disability?
  4. Could this misbehavior be a result of other factors?
  5. Are there causes of misbehavior that I can control?
  6. How do I determine if the misbehavior is classroom based?
  7. How do I teach students to self-regulate or self-manage behavior?
  8. How do I determine what methods of control are appropriate without violating the rights of students with disabilities mandated under P.L. 105-17? [the US's Individuals with Disabilities Education Act]
  9. How do I use reinforcement strategies to reduce disruptive behavior?
  10. Is it appropriate for me to use punishment?

Several of these questions relate directly to the individual and disabilities or difficulties they may have, but several make it clear that the environment is also a significant factor - that the curriculum and teaching method, influences from other students in the class, and factors from outwith the classroom are all elements to consider in ascertaining the source of disruptive behaviour and how to attend to it.

Online, an additional and far more common provocation for disruption can be simple misunderstanding. Most communication on the web is still text-based and asynchronous, and e-learning conversations can involve a diverse range of participants from different backgrounds and who communicate at different levels. The fact that e-learning is typically used to facilitate distance learning courses amplifies this diversity, in that students can be engaging from different countries, cultures and time zones, and will often span a much wider range of age groups than a typical undergraduate course where participants fill a narrow age range of 17-24 (Dutton, 2002). The perception of seemingly simple textual augmentations such as emoticons or smileys, originally intended to help the perception of tone and avoid misunderstandings in heated discussions, may even be a cause for complication (Walther & D'Addario, 2001) — research by Krohn (2004) into how different generations perceive emoticons concludes:

It is recommended that recipients who are Traditionalists (born before 1946) should not be sent e-mail with emoticons; those who are Baby Boomers (those born between 1946 and 1964) probably should not be sent e-mail with emoticons; those who are Generation Xers (those born between 1964 and 1980) may be sent e-mail with some of the more common emoticons; and those who are termed Millenials (born after 1980 and coming of age after 2000) may be sent e-mail with generous use of emoticons.

While this is clearly a generalisation, the use of emoticons and internet-age terminology, combined with the need to use and configure an increasing amount of technology in order to fully participate in an e-learning course rich in multimedia and immersive environments such as Second Life, does mean that those with more relevant experience are obviously more likely to be comfortable in an online learning setting, and less sensitive to the misinterpretations, abrupt actions and apparent cliquishness which can quickly evolve in discussion fora or instant messaging.

Research Study

Research design and method

University staff have much experience in handling the behaviour of students in a traditional face-to-face setting, with established security protocols to protect both staff and students should dangerous situations arise. The immediate threat posed online by a student may initially appear to be more virtual and less damaging as a result. However, virtuality is still grounded in a real world presence somewhere, and the ambiguity of place and lack of direct influence over the student has the potential to be even more damaging than when in the classroom — it is more likely that a student who has a high probability of causing harm will be noticed through the subliminal cues present in a face-to-face encounter with a staff member, than in attempting to read equally deeply into the words and tone used in an email conversation or chat session (Parkinson, 2007).

The approach taken in this study was to gather qualitative feedback from a range of academic staff across the United Kingdom. This feedback was intended to cover both online and offline situations, particularly as the use of online environments as a primary teaching tool is a relatively recent innovation, and so the number of examples of disruptive behaviour available would be expected to be significantly fewer. The study was designed around three research questions:

Secondly, the study was intended to summarise the guidance offered by existing 'toolkits' in how to approach and address disruptive behaviour online and offline, and whether these expertise and experiences could be shared to produce new or revised recommendations for use by disability services and academic staff in higher education institutions.

As the research is primarily qualitative and was likely to require participants to carefully consider how much detail they provide, an online survey format was used to give participants as much time as they needed to answer questions. The participants could choose to complete the survey anonymously, and were asked not to include any information in their responses which would allow individuals to be identified. The core of the survey centred on participants' experiences. The survey allowed for up to 7 experiences to be entered, each with five prompts to which the participant could type text freely. These prompts asked the participant to:

  1. describe where the situation occurred
  2. describe the situation and how they identified it as being disruptive
  3. explain what action they took to try and resolve the situation
  4. reflect on how effective they thought their response to the problem was
  5. describe any indicators they may have observed which they felt may have been the cause of the problem.

These prompts were designed to give sufficient context to align them with the wider research questions - the description of where the situation occurred places the instance either online or offline, while prompt #5 hopes to establish whether the member of staff had considered the possible causes, suspected the possible existence of mental health problems and connected this with the incident. These prompts, using language similar to that used above, were designed to be neutrally posed in order to avoid influencing the participant's emotional response to avoid leading questions. The prompts were designed to be clearly separable into three phases of detection, response, and reflection, in order to try to demarcate the emotional aspects of each phase and ensure the language used by participants was related specifically to that phase, rather than as an overall feeling about the whole experience. For instance, a single question asking about the experience may describe the behaviour as 'terrifying' and this emotionally charged use of language could dominate the rest of their response, whereas the participant's response and reflection on the situation could actually be quite positive. By separating their feedback into three areas it was hoped to ensure greater accuracy on the change in their emotions throughout the experience.

In addition, participants were asked to separately estimate the number of times they had experienced disruptive student behaviour in the past 3 years, in both a traditional 'offline' learning environment and an online learning environment. They were also asked to provide optional demographic data, including their subject area, geographic region, role in relation to students, age, gender and the number of years they had worked in higher education.

The rationale was to investigate staff perceptions of disruptive behaviour, and not necessarily how students or third parties saw the situation. This limits the coverage of the survey to only those directly affected by the incident, and so gives a distorted and subjective view of the situation, but since one of the key areas being investigated was that of attitudes, which are inherently subjective, it is less important to maintain a balanced viewpoint in the data set and more important to obtain true measures of how respondents felt.

The survey was conducted using the Bristol Online Survey tool, which was selected for its recognised reliability within the academic community and so to minimise any technical issues in using or developing an alternative survey tool. The survey was open from the beginning of October 2008 to the end of January 2009. The survey was open to any academic member of staff in UK higher education institutions, but a number of pre-selected programmes were specifically invited to participate. The initial intention was that these programmes were to be identified as having entirely online, partially online, and entirely offline delivery methods. Eight programmes were to be selected from each group, meaning 24 programmes were to be targeted in total. This was an attempt to ensure data was gathered from both offline and online participants, i.e. to avoid responses solely relating online experiences.

When research began to identify these programmes, it became difficult to clearly identify the stratification of specific courses into online and offline components - those that were wholly online advertise this fact well, but courses offering the combined blended learning approach were harder to find. As a result, this part of the research design was modified so that 10 mostly online and 10 offline programmes were selected from UK institutions. The selection process was semi-random in that an institution name was specifically chosen and a programme was randomly selected from their programme list. The programme was then analysed and added to either the online or offline list; if either list reached the maximum of 10 items, the programme was discarded and another search attempted.

The contact for each of the targeted programmes was the course organiser in most cases, and the departmental secretary where this could not be ascertained. The email also invited contacts to circulate the survey to other colleagues in their department. In addition to the targeted programmes, an invitation to complete the survey was circulated using various JISCMail mailing lists, including those dealing specifically with disabled students and students with mental health problems.

The expected response rate was assumed to be relatively low. The lists targeted have a combined membership of around 1500-2000 people, and assuming a typical response rate of 10%, this would yield a potential 150 respondents. However, the subject of the survey is of very narrow scope, in that participants most likely to participate are those who have experienced disruptive behaviour. One would expect participation to decrease even more when the demands of the survey (particularly that free text was to be entered as the main part of the survey) became clear to the participant. As a result of these factors, it was estimated that around 70 people would take part, each with variable engagement and with the number of experiences they contribute tending towards one.

The study was purposefully qualitative, with demographic quantitative data collected solely to isolate any particular biases that may be evident within the participant pool. The qualitative nature of the study means that the number of responses did not necessarily need to be high, so long as the detail collected in the response text was rich enough to provide some indication of whether the research questions could be answered adequately through the means of a survey, or whether further research would need to be conducted to show stronger correlations.

The initial design of the study, and the subsequent analysis, was in the grounded theory approach, where an open set of questions would be asked without specifying any expected outcomes, from which the answers participants provided would help in compiling theories. The responses from participants would be collated to create a corpus and assessed by coding different aspects of their responses. The range of behaviours and experiences of a disparate group of individuals in a wide variety of different - and difficult - situations makes it near impossible to preconceive a theory of the expected results, and so the design was for conclusions to be drawn inductively from the corpus.

The collection of additional incidental data as a supplement to the main survey was initially considered but rejected. This would have taken the form of an open online web poll with a simple 'Have you experienced disruptive behaviour online?' question, or a slightly more detailed Mechanical Turk Human Intelligence Task where participants would be recompensed for their participation. This approach was rejected for several reasons: technical (Amazon do not operate Mechanical Turk outside the US, and although there are techniques to submit tasks by proxy, participants would be likely be North American and a different demographic from the main survey), skewed participation (Mechanical Turk participants who are paid have additional extrinsic motivation to participate) and methodological (even if the poll did give a reasonable response rate, it would not be possible or necessarily helpful to relate this to the main survey).


Over 3000 words were obtained from 38 specific experiences identified by 23 respondents. Nearly all (22) respondents provided at least one experience, with seven of these relating two incidents, three describing three experiences each, and one giving five examples. In total, the 23 respondents reported that they had encountered 169 incidents of disruptive behaviour in their experience in higher education. This was split dramatically between traditional and online environments, with 162 taking place in a face-to-face setting, and just 7 emerging online. In the traditional category, 11 respondents had experienced 1-4 incidents, and 8 had experienced 10-14 incidents, with the remaining 4 each reporting zero incidents, 5-9, 15-19 and 20-24 incidents respectively.

The survey tool used, Bristol Online Survey, also records participants who started, but did not complete, the survey. Analysis of this larger set, which includes an additional 20 participants, has been excluded from this study as the data is incomplete and in all but one case the available data only extends to the first two 'have you experienced...' questions and does not give additional qualitative data. For these first two questions, the inclusion of the non-completion participants indicates that their experience of disruptive behaviour online is significantly less notable than offline - 17% of the core sample experienced online disruptive behaviour (26% in the wider data set) while 96% experienced offline behaviour (81% in the wider set). Note however that these data do not allow us to validly make any direct observations, for reasons discussed later.

With incidents online, 2 respondents had experience of 1 incident, 1 recalled 2 encounters, and the fourth had had 3 encounters with disruptive behaviour. Illustration 2 shows the comparison between the quantity of respondents' experiences in online and offline environments.

Bar chart showing the number of respondents on the vertical axis, with the horizontal axis showing grouped of the number of incidents encountered (0 incidents, 1-4 incidents, 5-9 etc.) and a set of bars for online and offline incidents.  The bars for online show the majority of respondents (19) had experienced no incidents with a further 4 experiencing 1-4.  For offline incidents there is more of a spread, with 13 experiencing 1-4 and 8 experiencing 10-14 incidents; other categories hold 1 respondent or fewer

Illustration 2: Disruptive incidents encountered online and offline

The number of respondents is not sufficient to draw any conclusions from the demographic of respondents against the data itself; as a general indicator of the type of respondent, of those who answered each demographic question:

Of 22 respondents who answered the question asking about their role in relation to students, 6 described themselves as support staff, with the remainder (72%) selecting roles of an academic nature. Illustration 2 illustrates the split of association between different roles; note that since respondents could select multiple roles, this is not a precise one-to-one mapping of role to respondent, but more an estimation of the percentage of time the respondent pool spends on each role.

Bar chart

Illustration 3: Staff roles that respondents associated themselves with

Emotional themes

The responses for each of the 5 questions were combined into 5 individual corpora and one overall corpus, which were then analysed. Stop-words were removed (using a list provided by the University of Glasgow's Information Retrieval Group) and commonly occurring descriptive words (students, student, situation, lecture, class) were also removed where necessary. In order to approximate any similarity in respondents' experiences, the basic word frequencies were then calculated using a Python script and ordered by frequency. Excluding stop words and common words, 965 unique words were used in 1608 instances, indicating very little crossover between responses. The top ten words and their frequency are shown in table 1:

Table 1: Frequency of words in complete corpus - ten most frequent
word frequency
effective 16
asked 15
shouting 12
staff 11
group 10
work 10
library 9
stop 9
behaviour 7
leave 7

The corpus was then analysed using Yoshikoder against the English Regressive Imagery Dictionary in order to quickly isolate words of an emotional nature. Of 1256 words (including stop-words), 94 were identified as within the emotional categories, with 52 related to 'aggression', 16 to 'expressive behaviour', and 16 to 'affection'/'positive affection'. Four were classed in the 'glory' category, and three each in 'anxiety' and 'sadness'. These results are not surprising given that the survey explicitly sought experiences of a negative nature, though the number of emotive words encountered in the corpus only accounts for 7% of unique words identified.

As the study was exploratory in nature, to attempt to identify key themes the individual corpora for each question were loaded into Wordle and a visual representation of the weighting of each word obtained. These images are displayed below along with an indication of the size-frequency measurement:

Word cloud showing a large number of words, with most frequently occurring including lecture, library, classroom, class, office

Illustration 4: Question A, 'Describe where the situation occurred' (measure: lecture = 12 counts, classroom = 6; only 'student' and 'students' excluded due to the situational nature of the question)

Word cloud showing a large number of words, with most frequently occurring including shouting, staff, another, disruptive, talking, particular

Illustration 5: Question B, 'Describe the situation and how you identified it as being disruptive' (measure: shouting = 8 counts, staff = 6)

Word cloud showing a large number of words, with most frequently occuring including asked, explain, leave, follow

Illustration 6: Question B, 'Describe the situation and how you identified it as being disruptive' (measure: shouting = 8 counts, staff = 6)

Word cloud showing a large number of words, with most frequently occuring including very-effective, effective, situation, good, worked, given, think

Illustration 7: Question D, 'Reflect on how effective you think your response to the problem was' (measure: very-effective = 6 counts, worked = 3. Additional modifications to this wordset were to replace 'not ' with 'not-', ' effective' with '-effective' and remove 'response' in order to give a better indication of the direction of effectiveness)

Word cloud showing a large number of words, with most frequently occuring including unknown, drugs, one, behaviour, lack, enough

Illustration 8: Question E, 'Describe any indicators you may have observed which you feel may have been the cause of the problem' (measure: unknown = 5 counts, drugs = 4)

In addition, each response was manually ascribed codes, selected by the researcher based on the content of the responses, to indicate:


As would be expected given the results from the quantitative questions, the degree to which disruptive behaviour was experienced online was found to be low when analysing the responses, with two responses specifically mentioning 'email', and three classed as ambiguous since they occurred over the telephone. The majority of incidents occurred in lecture/classroom settings (17 of 37) with a large number of people, with a further 6 occurring in small groups (laboratory/tutorial), 4 in a library, and 3 in a one-to-one setting in the respondent's office. It is not surprising that more incidents occurred in semi-public settings, given the amplifying effect on behaviour that groups have over individuals (see Dyer, Ioannou, Morell et al, 2007), and this is further explained by looking at the type of behaviour experienced. The responses were coded using the description of the incident, into three categories: direct, where the respondent or a third party was the specific target of behaviour; indirect, where the behaviour was observable but did not have a specific target; and passive, where the behaviour was not notable as an incident but did have consequences. The responses were mainly split between indirect (19) and direct (16), with a slight tendency toward indirect, and two classed as passive.

Effectiveness of response

In terms of whether the respondent's action was effective, this is a more subjective measure both in terms of the coding and the respondent's perception of the situation: it could clearly be possible for the respondent to believe the situation had been resolved, but for the student (and their peers) to believe the situation had not been tackled and for problems to still exist. This difference in perspective is addressed by Sullivan and Flanagin (2003b) in the combinations they propose in 'An Interactional Reconceptualization of 'Flaming' and Other Problematic Messages', which presents a matrix and examples for the eight combinations of 'message interpretation from multiple perspectives': whether an action is perceived as appropriate or a transgression, by the sender, recipient, and third-party. This is a useful approach to determine the impact of an action; however, in the case of the data available within this survey, only that of the 'recipient' is available, and so only their interpretation of the success of the action has been used. Perhaps unsurprisingly, the majority of interventions were perceived to be successful, with 26 culminating in successful or partially successful intervention. Seven incidents were recorded where an attempt at intervening was made, but ultimately did not resolve the problem. In three cases, no attempt at intervening was made.

Cause of incident

Respondents were asked to suggest what they felt the cause of the incident may have been. This question was again asked neutrally with no examples of causes, in order to avoid tainting responses. Similarly, there is clearly a large dependence on respondent's perception and judgement, and a smaller margin in coding this into a small set of categories. Here, the outcomes are less polarised, and spread over a wider range of possible causes:

Table 2: Possible causes of incident based on interpretion of participant's response
Cause Number of incidents
unknown 9
student confidence / immaturity 7
student cliques and group dynamics 4
deficiencies in programme structure 3
student drug abuse 3
student disability / mental health / specific learning difficulty 3
deficiencies in programme delivery 2
student (absence of) discipline 2
student beliefs / uncharacteristic activity 2
cheating / plagiarism 1

There is no evidence to indicate that the three suspected cases of drug abuse were substantiated, or that they should be considered to have affected the student's state of mental health. However, combined with the 'confidence/immaturity' and 'disability/mental health/specific learning difficulty' categories, these cases amount to 13 incidents where the student's personality and cognitive state have been felt to contribute most significantly to the disruptive behaviour. In the cases where mental health and specific learning difficulties are mentioned, respondents noted:

Incident: Worked her actions and jokes into the lesson content, and after the class challenged and analysed her attention seeking behaviour. Response/cause: Very effective. Giving appropriate attention led to the ultimate resolution of the underlying problem, which was undiagnosed dyslexia
Incident: the student was complaining about a mistake that a colleague of mine had made in the past, during which time i was still working in another institution. This was also something that i had found out recently about and had already done whatever was necessary Response/cause: The student has explained to me in the past that he is bipolar and that he had to change his medication recently and that doctors warned him that this would affect his mood.
Incident: student invading others' privacy; eg student lying under cafe tables.Complaints from other students about being distracted. Response/cause: cause unknown; likely illness

Those responses which referenced drugs described the cause as:

he blamed others for his failing on courses. Later I thought he was on drugs.
here drugs were certainly involved
Word has it that this student was on drugs and it was allegations of drug dealing and abusive behaviour to other students which had resulted in his being evicted from the halls

The respondents clearly indicate these are hearsay or assumptions made on the part of the respondent. Those which reference confidence problems or immaturity describe this as:

I suspect that being home sick may have played a part - although encouragement to be disruptive by another student undoubtably played a part.
lack of confidence (indicator??), lateness
the quiet or good behaved children feel that the less able or naughtier children dominate teachers time therrfore they rebel to gain the attention back.
One male ego amongst 27 females
The immaturity and personality of the individual concerned.
Wanting to hear his own voice and/or low self esteem?

The responses given by participants describing how they attempted to resolve the situation are more varied and so are more difficult to codify. However, some broad, common-sense themes are visible:

Breadth of respondent experience

As previously noted, this type of survey is unrepresentative as part of its nature. Due to the title and description of the survey, it is highly unlikely that recipients who have not encountered disruptive behaviour either online or offline would participate; the core will have experienced such behaviour and so the data gathered is principally of a qualitative nature. For this reason - that the study sought to discover people who had experienced behaviour - it is likely that there is no major distortion effect in pre-selecting specific programmes, and inviting participants to invite their colleagues, since the data was intended to be cumulative rather than representative.

However, it is important to gather some quantitative data in order to put the responses into perspective, which was achieved by asking for simple metrics on the number of offline and online incidents experienced over the past 3 years. The data here initially point to a substantive bias toward offline incidents, but a significant flaw in the structure of the survey is subsequently revealed. The main quantitative questions asked how many times the participant had experienced disruptive behaviour in a traditional offline environment, and in an online environment. They did not identify the non-disruptive experience the participant has had: whether or not they have ever used online and offline environments. This additional knowledge - coupled with the gathering of a simple split of teaching time between online and offline - would have given important context to the original questions, allowing them to be weighted properly and allowing a determination on the validity of these statistics. This does not affect the textual responses, but it does mean we lose important insight into the underlying drivers in the research questions: whether disruptive behaviour happens more or less online than in traditional environments, and whether it is noticed less or more online.

Essentially, had the respondents who have experienced no disruptive behaviour online also never taught online, there would be no correlation to investigate and their response could be declared void. Since this question was not asked, all of the quantitative data recorded is partially ambiguous and so allows no concrete conclusions to be drawn. In a sense this is acceptable, since the participant sample is a low number, but the design of this section of the study should be reconsidered should it be revisited in future.

The technical and aesthetic design of the survey was clearly a deterrent in some cases. Of 43 participants who completed the first relatively simple page, only 23 went on to complete the more demanding, qualitative portion. Part of this attrition is natural adversity to free text responses amongst time-stretched participants, but it is also likely that the length of the page - 7 questions, each with 5 text boxes - will have dissuaded some initially committed participants from continuing. This perhaps could have been averted by means of a more dynamic system whereby participants only add, and are only presented with, as many experiences as they require.

Advice and tools

When asked to reflect on their response to the incidents, some of the answers given by several of the respondents give some clear and grounded advice:

In all situations, I aim to:
  • ask students to calm down,
  • explain, in a calm, but firm, manner why I find such behaviour inappropriate,
  • ask students to have a minuted meeting and/or follow this up in writing.
  • highlight relevance of utilising peaceful and mutually respectful resolution.
  • listen to students (aiming to identify why they do that)
  • - offer solutions for issues which are of concern,
  • explain and follow all procedures which regulate our contact and bind it.
  1. gave students a 10 min break
  2. when we came back, I calmly, but firmly, explained the students the requirement that we talk and behave respectfully to each other and that we jointly resolve any issues they may face.
  3. My colleague got the relevant written guidance (which students had all along)which clearly stated students obligations (but they didn't read it and, apparently, we didn't stress it enough during preparation.
  4. Asked the student who shouted to meet me later on
  5. In the meeting, asked the student what was the problem and reiterated why shouting is not acceptable.
  6. Responded promptly (verbally, in the meeting, and in a follow up e-mail) to all of the concerns student raised. Some of the follow up correspondence included forwarding of evidence (i.e. previous communication) that her concerns have been dealt with. Reiterated relevance for students to read carefully relevant guidance.
  7. Asked student for an appology.
I took him into an open area and spoke very very quietly and sympathetically and explained what I could do to help. Meanwhile cos he was shouting other staff walked by wondering whether to intervene but then they decided I was handling it well
[regarding an inappropriately dressed student] I had already started the lecture. I invented a problem solving activity that required the students to go outside and then once they had started to move out, approached the offending student and instructed him to leave the lecture room and not return until he was appropriately dressed. He argued that he had been at lectures all that day and no-one had queried him, but I insisted and he left with minimal fuss. About an hour later he returned wearing jeans, a shirt, and shoes.
[regarding plagiarism] Students were penalised and required to meet with me and then dean to explain their motivation for cheating. The students were required to complete a parallel task which was rated one band lower than achieved. The matter was mentioned at the college without students being named. I have not experienced this sort of behaviour since.

From these experiences, and from many of the experiences and case studies supplied in the literature, it is clear that the intent of the disruptive student is normally not hostile. Their intention - if indeed it is a conscious decision - tends to be attention seeking, socially related, or through frustration with a particular inability to engage with a part of the subject or the way in which it is provided. This is not to say that there are not students who have a specific intent to cause problems; these students exist but are usually more identifiable, and institutions have a long history of using and developing disciplinary policies to accommodate this overt behaviour.

The likely causes of disruptive behaviour are widespread (Brockelman, 2009; James, 2006), ranging from a real educational problem where the student is unhappy with the way the course is being developed, to an unrelated problem in the student's personal life perhaps relating to a family problem, which is then channelled into an academic setting by virtue of it being a generally more open and discursive environment. It is impossible for academic staff to plan for all of these scenarios and aim to prevent them; disruptive behaviour will still occur in the most inclusive of environments, and it is important to have strategies to mitigate the impact of these events, and for staff to know how to safely protect their students while retaining a balanced perspective. The techniques used in this reactive phase are important tools to keep alongside; far in advance of this, however, is the important phase of creating a learning environment which is less likely to engender disruption, and the later conduct of teaching within this environment in a way which remains inclusive (Nilson, 2003). Usually, if the environment is designed well and is administered sensibly, with awareness of all the different types of personas and problems that may exist, it is possible to prevent serious problems from occurring. The next section discusses some simple tools that help in both these preparatory and reactive phases; one final critical phase, though, is that of reflection and review. Some of the participants in this study gave examples of changes that they had made to the way they provided the course, in direct response to the disruptive behaviour. For example:

I introduced the idea of written and unwritten ground rules, and adopted a check in and check out process. These mechanisms seem to have empowered students to let their peers know when they do behaviours that annoy them or detract from their ability to concentrate etc. The amount of irrelevant noise in the class has reduced to almost nil and few students have their phones ringing during lectures now. I also use less lecturer talk and more group work, problem solving activities etc., and also have changed the assessment schedule so that poster presentations are on the topic covered in the lecture the previous week, and use peer as well as lecturer assessment, so these strategies together mean that students need to attend to be able to do the tasks required.

This change could be perceived as unnecessary mitigation in order to prevent another difficult situation, or it could be regarded as a positive enhancement to the learning environment: the fact that students - aside from the disruptive individual - appear to be more involved in their studies may indicate that in this case that the enhancement has been felt across the cohort. After a disruptive incident, it is helpful for staff to reflect on both the incident, and the likely cause of it, and their response to it. If the cause is preventable through a minor change in their provision of the learning environment, then it is likely to be beneficial; as with many accessibility enhancements, making a particular feature more available and accessible for one person can often make it more usable for the wider population (Disability Rights Commission, 2004; W3C, 2005). Equally, the staff member's response should also be reviewed; it is often difficult to judge how best to deal with a difficult situation, and staff should be under no expectations to behave perfectly in each scenario. It would, however, be wasted effort if staff did not introspect and consider the effectiveness of their behaviour, so that they can improve in the future, and continue to contribute to the inclusivity of the learning environment.

What to consider when constructing the learning environment

When developing a new online course, a range of pedagogical and accessibility considerations should be taken into account. For students with mental health problems, these tend to be more of a social nature, although as with any student, complex technical problems can cause frustration - particularly as they are the sole access route into an activity - and so exacerbate any existing problems. Developing an inclusive framework for an online course in advance is difficult without knowing the spread of difficulties that students may have, so often it is necessary to build on prior experience and implement solutions that have been required in the past, and then be prepared to adjust to new problems. For students with mental health problems, when a particular issue arises, availability of information, availability of another person (staff or student) to speak to outwith the formal environment, and other tools which limit the feelings of isolation and frustration, can be key to minimising their negative experience. (Tinklin, Riddell, Wilson, 2005) Interestingly, the available literature tends to neglect the initial course design phase, focusing more closely on how to prevent disruption during teaching, and how to react to it. Yet creating a good teaching environment allows both student and teacher to relax, developing a closer bond and less likelihood of friction, and so potentially lessening the burden later in the course. Some key points to consider in the course design are:

What to consider when teaching in the learning environment

When it comes to the day-to-day activities in encouraging participation in an online course, and dealing with the inevitable frustrations, more guidance is available. Hailey, Grant-Davie and Hult (2001) discuss a range of difficult problems and solutions they have experienced in online classrooms, as do Sapp and Simon (2005) in their review of online versus face-to-face writing courses. The National Institute of Adult Continuing Education (NIACE, 2007) also gives credible advice on a larger, organisational level, and in combination they recommend:

What to consider when dealing with a difficult situation

When actually dealing with a difficult scenario, some of the best advice comes from outwith the educational environment, in corporate culture where difficult and disruptive scenarios are far more prevalent due to stress. Gray (2006) contributes a particularly strong set of considerations which can easily be translated into the e-learning world:

  1. Take a minute. Cool down. Try to see how the affront could have been unintentional.
  2. Pick up the phone, or walk over to the person's desk. If they're not there come back later. Whatever you do, don't send an email.
  3. Tell them what you are concerned about. Don't assume or judge them! Most of the time people are not aware that they stepped over a line. If something really bothers you or is unacceptable, you can say this -- but say it in a calm tone of voice.
  4. Ask them what you can do to ensure that such things don't happen in the future. That's right, what you can do.
  5. Listen. You're opening the door to a conversation. You will find that you are often surprised. Being open to the other point of view can enlighten your understanding of the situation and maybe even help you see a larger system dynamic.

On point #2, a particular challenge for e-learning arises, in that the only contact a teacher may have for a student is an institutional email address, but this may be something to consider in the planning phase. If an email is the only option, then an email which opens up other discursive opportunities, such as asking them to give you a call if they feel comfortable with this (and providing your phone number), may give the student enough space to defuse the situation and encourage them to tackle it offline. In general though, as both Gray and Hailey et al suggest, sending a terse email which attempts to 'end' the disruption will be futile and exacerbating.

There are clearly situations where tackling the situation single-handedly would be disadvantageous, and possibly dangerous. The safety of the staff member and their students is paramount, and while some staff members may feel experienced enough to deal with problems on their own, they should always first consider whether they could be better resolved with the assistance of a colleague, or whether security or emergency services should be silently (through panic buttons) or actively (raising an obvious alarm) contacted.

Many difficult situations can be defused to some extent, through the authority of the staff member and the combined authority of the student group, or perhaps simply through distraction in the form of joking or talking tangentially about an unrelated topic; while in retrospect the staff member may not feel that their approach was ideal, if the situation did not escalate to a more hostile level then their intervention can be considered a success, and lessons can be learned from the experience.

How to reflect and review

Coupled with Gray's advice on dealing with flames generally, while the individual will always be the most important indicator, these suggestions point to some clear ground rules for creating and sustaining an environment which is less likely to contribute to the causes of disruption. When situations do, inevitably, become difficult and disruptive, while the staff member's reaction may be adequate at that moment and in the short term, it is crucial to follow up and review the incident, for the benefit of the disruptor, the third parties involved, and particularly the staff member affected. Failing to do so can leave a lasting impression on each individual's performance and satisfaction, potentially leading to further disruption and an increased likelihood of developing mental health problems in the future. As a staff member reflecting on the situation, it may be useful to:

Using other resources

Key to leading an online course within an institution is recognising that the rest of the institution is there to help you and your students. While some of the participants in this survey felt that their teaching environment was in some way deficient and the cause of the problems they encountered, others recognised that the environment would be perfectly adequate when combined with some additional support, either through making reasonable adjustments for particular students, or through advising students to take advice from the institution's student support services.

Floyd and Casey-Powell (2004) set out a number of phases through which disabled distance learners may require some level of student support from their institution, loosely following Miller and Prince's (1976) student development process model. The main phase for academic teaching staff to consider is the Learner Support Phase, where the majority of 'handholding', advice and assistance is offered in specifically relation to the student's studies (as opposed to future career plans etc.). They note the policies set out by Colorado Community Colleges Online (2009) for making adjustments for disabled students, which are set out simply, and assign clear responsibilities. They conclude that 'Distance learning students expect to receive online support services that will allow them to succeed in their educational endeavors.'

This is true, and while support services may be perceived as being behind the field in respect to online learning, their activity in this field may be influenced by an increased volume of online learners who are encouraged to use their services. While ad-hoc solutions created between teacher and learner can have a remedial effect on one specific issue, the problem is that this learning experience is then not shared across the wider community and allowed to gradually influence teaching practice. Continual feedback of experience - whether good or bad - to the central student support services can only help to equip them with more information on how to address and improve provision across the campus, leading to a better experience for all.

In response to the rise in e-learning in the UK, several general resources have arisen to assist teachers in providing a good learning experience for disabled students. Some of these include:

In addition to these general courses, most institutions will have on-campus student support services which provide training on disability awareness and equality, and many should have more technical accessibility related training, including e-learning focused courses. Where these do not exist, their development should be encouraged, and academic staff have a key role to play in developing course content which is tailored to the institution.


Online learning is set to increase over the next 5 years (Information Week, 2007), and is likely to be a major growth area for many institutions which have until now only experimented with pilot and small-scale e-learning projects. That online learning has previously addressed only a comparatively narrow section of the higher education audience (aside from the Open University) may in part help to explain the lack of reported online incidents received through the small sample of respondents in this study. Deficiencies in the study itself make it difficult to draw strong conclusions from the data, but the increasing availability of advice and guides on dealing with difficult behaviour online indicates a need for institutions to prepare staff in the different ways of approaching disruptive behaviour online as opposed to the offline environment.

The survey used in this study intentionally did not provide a definition of what 'disruptive behaviour' might mean, in order to avoid influencing respondents' descriptions of their experiences. It is possible that this is a contributing factor to the lack of responses relating to online experiences: academic staff may not yet recognise or regard some types of online behaviour such as trolling as being particularly 'disruptive', as the asynchronicity of online communication can alleviate the immediacy of disruption, or at least defer the impact.

There are observable flaws in the survey design which affect both the qualitative and quantitative results. Firstly, the quantitative measure of the number of incidents experienced offline and online lacks a control indicator showing whether participants had ever been involved in educating students in these environments. This limits confidence in the difference in the data between offline and online; in any case, the number of participants is not sufficiently high to draw conclusions on this characteristic of the survey. This survey question also limits participants to the previous three years, which may be too confined, and does not enquire as to the frequency of incidents: whether they are increasing in proportion to the rise in online learning or remain static. Again this is a theoretical concern given the data collected, but should be addressed in any further study.

However, as previously noted it is clear from the available literature that the potential causes of disruptive behaviour, and specifically attitudes toward mental health problems, either are not a central part of the discourse in discussing accessibility of e-learning environments and in making adjustments for disabled people, or are negatively held, seeing such behaviour and mental health problems as an issue for the student and other health professionals to overcome. This attitude overlooks the role that the member of teaching staff, and the design of their teaching environment, can have in influencing the underlying causes of disruptive behaviour, both positively and detrimentally. In a traditional setting, Suler (2003) suggests that presence can be identified through:

In the online environment, some of these factors can be limited and suppressed depending on the student's previous experience of and confidence with technology. (Conole et al, 2006) While the fast changes in generational uptake of technology will mean that many students will quickly become more proficient than teachers in the use of online learning tools, this apparent experience gap is not merely a short-term temporal issue: the learning styles, confidence levels, and particular abilities of each and every student will always be different, and will always be maximised by participating in an environment which is flexible and understanding of their needs. The tutor's role is still key to ensuring greater success for learners, and their own background, experience and early intervention play a crucial role in creating a welcoming and dynamic environment. (Parmar and Trotter, 2004; McPherson & Baptista Nunes, 2004) In summarising student's perspectives in how higher education teaching staff handled the consequences of the academic experience and the student's mental health problems, Riddell et al (2006) note:

Lack of understanding among lecturers, a culture in which it was difficult to admit to having difficulties, a lack of support for learning and badly designed learning experiences had all contributed to the students' distress.

Orton-Johnson (2007), in identifying the need for more research into the learner experience online, speculates that, 'If a student does not actively participate in the online discussion, he [sic] does not exist.' This may not be true in every circumstance, but having mechanisms in place that will help tutors identify - at a very early stage - whether students are having any difficulties with the technological, pedagogical or social elements of the course, and acting quickly on this knowledge, is likely to be crucial to sustaining learner's early engagement with the course in order that they can built up confidence and become self-sufficient (Boaz, 1999; Rekkedal & Qvist-Eriksen, 2004; Tobin, 2001).

Learning, whether online or offline, is geared around the exploration of the subject and the understanding of knowledge, and good communication between tutor and learner is fundamental to success. The experiences contributed in this study show that disruptive behaviour, although by no means wholly avoidable, can be mitigated by sufficient preparation and creation of a welcoming, accessible environment. The experiences and literature also suggest that some disruptive behaviour can be positive, in drawing attention to facets of the environment or the teaching provision which could be improved upon. Perhaps the core message identified is that while computer-mediated communication may have the advantages of being fast, cheap, asynchronous and global, the technical superiority, structure, visual aesthetic, and detailed pedagogical content provided through the learning environment are not in themselves enablers of learning. Continued high quality and high frequency communication, a measure of successful traditional learning which serves to actively engage and invigorate the recipient as well as the sender, is still key in the online world to helping students achieve their full academic potential.


  1. Allen, C. 2006, 'Flames: Emotional Amplification of Text', viewed: 20 June 2009 (
  2. Anagnostopoulou, K., Parmar, D. & Priego-Hernandez, J. 2009, 'An exploration of perceptions of learning and e-learning held by students who withdraw and those who persist with UK higher education', Brookes eJournal of Learning and Teaching, Volume 2, Issue 4
  3. Beaudoin, M. 2002, 'Learning or lurking?: Tracking the 'invisible' online student', The Psychology of Cyberspace, Volume 5, Issue 2, Summer 2002, Pages 147-155
  4. University of Berkeley Health Advice, 'How to Assist the Distressed Student', viewed: 20 June 2009 (
  5. Boaz, M. 1999, 'Effective Methods of Communication and Student Collaboration', in Teaching at a distance: A Handbook for Instructors, Harcourt Brace
  6. Bohman, P., Anderson, S. 2005, 'A conceptual framework for accessibility tools to benefit users with cognitive disabilities', paper presented to the ACM 2005 International Cross-Disciplinary Workshop on Web Accessibility, Chiba, Japan
  7. Britain, S. & Liber, O. 2004, 'A Framework for the Pedagogical Evaluation of Virtual Learning Environments', Joint Information Systems Committee, viewed: 20 August 2009 (
  8. Brockelman, K. 2009, 'The Interrelationship of Self-Determination, Mental Illness, and Grades Among University Students', Journal of College Student Development, Volume 50, Number 3, May/June 2009, pp. 271-286 (
  9. Browne, T., Jenkins, M. & Walker, R. 2006, 'A Longitudinal Perspective regarding the use of VLEs by Higher Education institutions in the United Kingdom', Interactive Learning Environments, Volume 14, Issue 2
  10. Burgstahler, S., Corrigan, B., McCarter, J. 2004, 'Making distance learning courses accessible to students and instructors with disabilities: a case study', The Internet and Higher Education, Volume 7, Issue 3
  11. Cheng, J. 2009, 'Trolling someone online? Bill would slap you with jail time', Ars Technica, viewed: 20 June 2009 (
  12. Collins, M., Mowbray, C. 2005, 'Higher education and psychiatric disabilities: National survey of campus disability services', American Journal of Orthopsychiatry
  13. Colorado Community Colleges Online, 2009, 'ADA Information', viewed: 20 August 2009 (
  14. Conole, G., de Laat, M., Dillon, T., Darby, J. 2006, 'An in-depth case study of students' experiences of e-Learning — how is learning changing?', paper presented at 23rd annual ascilite conference: Who's learning? Whose technology?, Sydney, Australia
  15. Daniels, V. 1998, 'How to Manage Disruptive Behavior in Inclusive Classrooms', Teacher Vision, viewed: 20 June 2009 (
  16. Dickey, M. 2004, 'The impact of web-logs (blogs) on student perceptions of isolation and alienation in a web-based distance-learning environment', Open Learning: The Journal of Open and Distance Learning, Volume 19, Issue 3 November 2004 , pages 279 — 291 (
  17. Disability Rights Commission (DRC), 2004, 'Formal investigation report: The Web: Access and Inclusion for Disabled People', viewed: 3 October 2009 ( )
  18. Dunn, S. 2003, 'Return to SENDA? Implementing accessibility for disabled students in virtual learning environments in UK further and higher education', viewed: 20 June 2009 (
  19. Dutton, J., Dutton, M., Perry, J. 2002, 'How Do Online Students Different From Lecture Students?', Journal for Asynchronous Learning Networks, Volume 6, Issue 1 — July 2002 (
  20. Epley, N., Kruger, J. 2004, 'When what you type isn't what they read: the perseverance of stereotypes and expectancies over e-mail', Journal of Experimental Social Psychology, Volume 41, Issue 4
  21. Equality and Human Rights Commission (EHRC), 2006, 'Disability Discrimination Act Guidance on matters to be taken into account in determining questions relating to the definition of disability', EHRC, viewed: 20 August 2009 (
  22. Equality and Human Rights Commission (EHRC), 2007, 'Code of Practice (revised) for providers of post-16 education and related services', EHRC, viewed: 20 August 2009 (
  23. Equality and Human Rights Commission (EHRC), 2009, 'Disabled students and learners in post-16 education', EHRC, viewed: 20 August 2009 (
  24. Floyd, D., Casey-Powell, D. 2004, 'New roles for student support services in distance learning', New Directions for Community Colleges, Volume 2004, Issue 128
  25. Fritsch, H. 1997, 'Host Contacted, Waiting for Reply', in Final report and documentation of the virtual seminar for professional development in distance education, viewed: 3 October 2009 ( )
  26. University of Glasgow Information Retrieval Group, 'List of stop words', viewed: 20 June 2009 (
  27. Gray, D. 2006, 'Go direct', Communication Nation, viewed: 20 June 2009 (
  28. Hailey, D. E., Grant-Davie, K., & Hult, C. A. 2001, 'Online education horror stories worthy of HALLOWEEN: a short list of problems and solutions in online instruction', Computers and Composition, Volume 18 Number 4, DOI: 10.1016/S8755-4615(01)00070-6 Hara, N., Kling, R. 2000, 'Student distress in a web-based distance education course', Information, Communication & Society, 3:4
  29. Herring, S., Job-Sluder, K., Scheckler, R., Barab, S. 2002, 'Searching for Safety Online: Managing "Trolling" in a Feminist Forum', The Information Society, Volume 18, Issue 5
  30. Hundal, S. 2008, 'The beginner's guide to writing for Comment is Free', Pickled Politics, viewed: 3 October 2009 ( )
  31. Information Week, 2007, 'Worldwide E-Learning To Draw Nearly $53 Billion By 2010', Information Week, viewed: 20 August 2009 (
  32. James, K. 2006, 'Supporting learners with mental health difficulties', Learning and skills development agency, viewed: 20 June 2009 (
  33. Kessler, B., Foster, C. & Saunders, W. et al, 1995, ' Social consequences of psychiatric disorders. I: educational attainment', American Journal of Psychiatry, 152:1026-1032
  34. Knock, A. 2006, 'Open Doors? The virtual learning life-cycles of students from contrasting backgrounds', viewed: 20 June 2009 (
  35. Kraus, R., Zack, J., Stricker, G. 2004, Online counseling: a handbook for mental health professionals, Academic Press
  36. Krohn, F. B. 2004, 'A generational approach to using emoticons as nonverbal communication', Journal of Technical Writing and Communication, 34(4)
  37. Lea, M., O'Shea, T., Fung, P., Spears F. 1992, ''Flaming' in computer-mediated communication: Observations, explanations, implications' in Contexts of computer-mediated communication, eds M. Lea, 1992, Harvester Wheatsheaf, New York
  38. Lindsay, E. 2004, 'Distance Teaching: Comparing Two Online Information Literacy Courses', The Journal of Academic Librarianship, Volume 30, Issue 6, November 2004, Pages 482-487
  39. McPherson, M., Baptista Nunes, M. 2004, 'The Role of Tutors as an Integral Part of Online Learning Support', European Journal of Open, Distance and E-Learning, 2004 Volume 1
  40. Miller, D., Morrison, K. 2009, 'Expressing deviant opinions: believing you are in the majority helps', Journal of Experimental Social Psychology, Volume 45, Issue 4
  41. Miller, T., Prince, J. 1976, 'The future of student affairs — a guide to student development for tomorrow's higher education', Jossey-Bass, San Francisco
  42. Mott, N. 2004, 'Organizational and Ethical Concerns in Providing Academic Accommodations for Learning Disabled Students in Higher Education', New Foundations, viewed: 20 June 2009 (
  43. New York Times, 2007, 'Diagram Of A Blog', viewed: 20 June 2009 (
  44. NIACE, 2007, 'Top Ten Ways to Promote Positive Mental Health in Adult Education', National Institute of Adult Continuing Education, viewed: 20 June 2009 (
  45. Nilson, L. 2003, Teaching At Its Best: A Research-Based Resource for College Instructors, Jossey-Bass, San Francisco
  46. Noelle-Neumann, E. 1984, The Spiral of Silence: Public Opinion — Our Social Skin. University of Chicago Press, ISBN: 0226589366
  47. Oliver, M. 1990, 'The Individual and Social Models of Disability', viewed: 20 June 2009 (
  48. O'Neil, M. 2009, Cyberchiefs: autonomy and authority in online tribes, Pluto Press, London
  49. Orton-Johnson, K. 2007, 'The Online Student: Lurking, Chatting, Flaming and Joking', Sociological Research Online, Volume 12, Issue 6 (
  50. O'Sullivan, P. 2003a, 'Reconceptualizing 'flaming' and other problematic messages', New Media & Society, Vol. 5, No. 1, 69-94 (
  51. O'Sullivan, P. & Flanigan, A. 2003b, 'An Interactional Reconceptualization of 'Flaming' and Other Problematic Messages', viewed: 20 June 2009 (
  52. Overbaugh, R., Lin, S. 2006, 'Student characteristics, sense of community, and cognitive achievement in web-based and lab-based learning environments', Journal of Research on Technology in Education, Volume 39, Number 2
  53. Palloff, R., Pratt, K. 1999, 'Building learning communities in cyberspace: Effective strategies for the online classroom', viewed: 3 October 2009 ( )
  54. Parkinson, B. 2007, 'Emotions in direct and remote social interaction: Getting through the spaces between us', Computers in Human Behavior, Volume 24, Issue 4
  55. Parmar, D. & Trotter, E. 2004, 'Keeping our students: identifying factors that influence student withdrawal and strategies to enhance the experience and retention of first-year students', Learning and Teaching in the Social Sciences, Volume 1, Number 3
  56. Phipps, L. & Kelly, B. 2006, 'Holistic approaches to e-learning accessibility', ALT-J, Research in Learning Technology, Vol. 14, No. 1, pp. 69—78 (
  57. Prior, L., Wood, F., Lewis, G., Pill, R. 2002, 'Stigma revisited, disclosure of emotional problems in primary care consultations in Wales', Social Science & Medicine, Volume 56, Issue 10
  58. Rekkedal, T., Qvist-Eriksen, S. 2004, 'Support Services in E-Learning - an Evaluation Study of Students' Needs and Satisfaction', European Journal of Open, Distance and E-Learning, 2004 Volume 1
  59. Riddell, S., Tinklin, T., Wilson, A. 2006, 'Support for students with mental health difficulties in higher education: the students' perspective', Disabled students in higher education: perspectives on widening access and changing policy, Routledge
  60. Robinson, C., Martin, J., Thompson, K. 2007, 'Attitudes towards and perceptions of disabled people', Disability Rights Commission, viewed: 3 October 2009 ( )
  61. St. Louis Today, 2007, "My Space" hoax ends with suicide of Dardenne Prairie teen', viewed: 20 August 2009 (
  62. Salmon, G. 2004. E-Moderating: The Key to Teaching and Learning Online, Routledge, ISBN: 0415335434
  63. Salzer, M., Wick, L. & Rogers, J. 2008, 'Familiarity with and use of accommodation and supports among postsecondary students with mental illnesses', Psychiatric Services, Volume 59, Number 4
  64. Sapp, D. & Simon, J. 2005, 'Comparing grades in online and face-to-face writing courses: Interpersonal accountability and institutional commitment', Computers and Composition, Volume 22, Issue 4
  65. Schwartz, M. 2008, 'The Trolls Among Us', The New York Times, August 3 2008, viewed: 3 October 2009 ( )
  66. Sharpe, R. and Benfield, G. 2005, 'The Student Experience of E-learning in Higher Education: A Review of the Literature', Brookes eJournal of Learning and Teaching, Volume 1, Issue 3
  67. Shirky, C. 2008, Here comes everybody: the power of organizing without organizations, Penguin, London
  68. Simpson, A. 2009, 'Competence Standards', SIG Inclusion e-bulletins on Inclusive Practice, viewed: 3 October 2009 (
  69. Stanley, N. & Manthorpe, J. 2000, 'Responding Effectively to Students' Mental Health Needs', The University of Hull, viewed: 20 August 2009 (
  70. Stanley, N. & Manthorpe, J. 2001, 'Responding to students' mental health needs: Impermeable systems and diverse users', Journal of Mental Health, Volume 10, Issue 1 February 2001 , pages 41 — 52 (
  71. Sternberg, J. 2000, 'Virtual Misbehavior: Breaking Rules of Conduct in Online Environments', Proceedings of the Media Ecology Association, Volume 1
  72. Suler, J. 2003, 'Presence in cyberspace', The Psychology of Cyberspace, viewed: 20 June 2008 (
  73. Taylor, M., Baskett, M., Duffy, S., Wren, C. 2008, 'Teaching HE students with emotional and behavioural difficulties', Education + Training, Volume 50, Number 3
  74. TechDis eds L. Phipps, A. Sutherland & J. Seale, 2002, 'Access All Areas: disability, technology and learning', TechDis, viewed: 20 June 2009 (
  75. Tinklin, T. Riddell, S. & Wilson, A. 2005, 'Support for students with mental health difficulties in higher education: the students' perspective', 2005, British Journal of Guidance & Counselling, Volume 33, Issue 4 November 2005 , pages 495 - 512 ( )
  76. Thornicroft, G. 2006, 'Actions speak louder... Tackling discrimination against people with mental illness', Mental Health Foundation, London
  77. Tobin, T. 2001, 'Dealing with Problem Students and Faculty', Online Journal of Distance Learning Administration, Volume 4, Number 3
  78. Turnage, A. 2007, 'Email flaming behaviors and organizational conflict', Journal of Computer-Mediated Communication, 13(1)
  79. UCAS, 2009, 'Breakdown of applicant and accepted applicant disability data over six years', viewed: 20 June 2009 (
  80. Waggoner, D. & Goldman, P. 2005, 'Universities as communities of fate', Journal of Educational Administration, Volume 43, Number 1
  81. Walther, J., D'Addario, K, 2001, 'The Impacts of Emoticons on Message Interpretation in Computer-Mediated Communication', Social Science Computer Review, Volume 19, Number 3
  82. Wikipedia, 2008, 'Spiral of silence), viewed: 20 June 2009 (
  83. Williams, K. 2006, 'On-Line Anonymity, Deindividuation and Freedom of Expression and Privacy', Penn State Law Review, 687 (2005-2006)
  84. Wolcott, H. F. 1990, Writing up qualitative research. Qualitative Research Methods Series, Vol. 20, Newbury Park, CA: Sage Publications
  85. Woodfine, B. P., Baptista Nunes, M. & Wright, D.J. 2008, 'Text-based synchronous e-learning and dyslexia: Not necessarily the perfect match!', Computers & Education, Volume 50, Issue 3, April 2008, Pages 703-717
  86. W3C (World Wide Web Consortium), 2005, 'Introduction to Web Accessibility', Web Accessibility Initiative, viewed: 3 October 2009 ( )