Jan 102013
 

There has been a flurry of commentary in response to reports that Minister Shane McEntee’s suicide was predicated by cyberbullying. Implications of causality notwithstanding, what follows is intended to bring clarity to this phenomenon. By the nature of this blog, this post is largely aimed at public representatives, but anyone else who finds themselves receiving large amounts of unpleasant communications may also find it useful – again, causality of such communication notwithstanding.

At the outset, it should be noted that cyberbullying is understood to not only encompass behaviour on social media like Facebook and Twitter, but email, text messaging and any other form of electronic communication should also be borne in mind. The issue of who or what a public figure should or should not respond to is beyond the scope of this post: the point is simply to deliver some educational pointers as to how one can deal with what is perceived to be cyberbullying, or any kind of persistent unwanted electronic communication.

Personally, I rarely engage with any online accounts which are difficult to identify as ‘real’ individuals and I can’t see why anyone else should either. It is assumed that genuine cases of needing to use anonymity when contacting public representatives – such as whistleblowers – will make their justification clear in their communications.

I should also stress that I am of the opinion that legislative attempts to remove anonymity from cyberspace will be ultimately unsuccessful: human nature, such as it is, has aeons of experience in circumventing perceived obstacles to its freedom. That does not mean however, that an enhanced form of verification or recognition online, like some kind of digital passport, might be an idea worth examining in a voluntary capacity. In addition, it should be noted that there really is no anonymity online anyway – the same ‘exchange principle’ of forensic science, which asserts that a criminal will always both leave behind and take with them evidence of their crime, also applies online. In fact, anonymous harassment via email is likely to be much easier to trace than via letter.

However, the perception remains that cyberspace is a dark and awful place and this is not without foundation. The reality of this situation is at least partially as a consequence of a lack of enforcement of existing law in cyberspace – whether due to unwillingness or ignorance, in many jurisdictions all manner of illegal behaviour has endured long after it has been restricted in the offline sphere (though this seems to be changing). At the same time, this situation is also at least partially, and likely more so, as a consequence of our psychological naïveté of this novel and constantly evolving environment.

For example, a key distinction between cyberbullying and traditional bullying, and one which has not been sufficiently acknowledged by researchers in the field is that a considerable amount of the former occurs behind a facade of anonymity, something which is difficult to achieve in the latter. For example, the EU Kids Online survey final report reports that 56% of cyberbullies also bullied offline, and 55% of victims of cyberbullying were also victims offline. This obviously implies that 44% of cyberbullies only engage in this behaviour online, and 45% of victims are not bullied offline: a considerable chunk of this behaviour is only occurring in cyberspace, with no ‘traditional’ bullying analogue.

In addition, a number of studies have noted that a considerable number of victims of cyberbullying have no idea who their perpetrator is. Such individuals, whether ignorant of their traceability as already mentioned, do seem to be assured of the improbability of such ever taking place.

Moreover, the technology of cyberspace affords the cyberbully the ability to inflict harm or discomfort, maximising the effect-to-danger ratio of indirect aggression. It is easy to attack a person in this way because there is no obvious way that they will be able to fight back. However, why a person actually feels the need to deliver such messages in the first place – anonymous or nonymous – needs also be questioned.

As I have written before, much of our current affairs broadcasting is adversarial, argumentative and aggressive. Perhaps in the era of Web 2.0 and participatory culture, it is unsurprising that the members of the general public feel entitled to engage in such unseemly behaviour also. As has been said, “it seems clear that part of the human condition is to find a degree of entertainment in conflict...” and when viewing political conflicts, whether on television, radio, print or elsewhere, those who feel like participating can do so with minimal risk of exposure or accountability.

A cultural shift is needed, which involves a wider conversation about, and a deeper examination of, how people interact with each other – both face-to-face and online. It behoves public representatives to lead this conversation, not through scaremongering of online technology, but modelling better interpersonal reactions and demonstrating how interactions between citizens can be carried out in a civilised and considerate manner. Social media in particular is an excellent resource for promoting constructive communication – we must remember, at the heart of all of this, to positively reinforce good behaviour in cyberspace and politicians must take the lead on this. To begin with, however, it is essential that we try to understand what is actually occurring when someone decides to send what we shall term ‘hostile electronic communication’ to a public representative.

Technological

There is both a technological and a cyberpsychological strand to this phenomenon, and consequently two approaches to dealing with it, neither of which should be neglected. I will deal with the former first, explaining a few technical features which you may or may not be aware of but which are worth noting. I am the of (largely unfounded) belief that careful attention to the architecture of the medium in which the unwanted communication is being received is likely to forestall it in the majority (including recent high profile cases).

  1. Mobile phone:
    • for better sleep patterns and peace of mind, reduce screentime of all kinds for at least an hour prior to sleep
    • it has been argued that mobile phone usage is like the transitional objects which young children have (‘security blankets’, cuddly toys etc.) and consequently we need to think about weaning ourselves off them to some extent
      • no-one should ever feel irrationally attached to their phone number, however neither should anyone feel like they have to change it
    • there are arguments to be made for not answering ‘private number’ callers, and discourse has begun on the ‘end of voicemail’
    • upon receiving malicious/hateful messages, save them, do not engage and report immediately to the Gardaí. Unfortunately, while it is possible to report suspected illegal internet/email content online to Hotline.ie, no such service exists for text messages.
  1. Email:
    • whatever email application you use should be capable of being configured so that any message containing particular words can be sent to a designated folder.
      • for example, you can filter all words containing profanities to the ‘junk’ folder
      • you can also filter all emails titled or containing policy keywords to specific folders which will allow you to accurately be able to report on the volume of communication you are receiving on it (which is important for certain types of individuals, see below)
    • email content which constitutes incitement to hatred can be reported to both the Gardaí and via Hotline.ie
  1. Facebook:
    • as a public representative, you should be using a Fanpage, not a Friend account – simply unprofessional in 2013.
      • your personal assistant, or whoever is managing your social media presence should know how to convert (and/or merge) from one to the other.
    • filter the content which can be posted on your Fanpage
      • use ‘Edit Page’ -> ‘Manage Permissions’ to block posts with profanities, and whatever keywords you like.
      • you can also restrict your Fans by age and location
      • this could save your team a lot of hassle in responding to or deleting unwanted posts
    • there is a considerable amount of detail on how to report abusive content on Facebook here: facebook.com/report
  1. Twitter
    • while you will learn a lot more about social media by spending some time using them, as with everything else, avoid usage late at night when concentration levels are likely to slip
    • it is up to you whether or not you engage with pseudonymous accounts
    • use a free service like Hootsuite to mute or filter conversations which you do not wish to engage in
      • you can also use those services to schedule Tweets (and Facebook updates) to appear at a later time
    • again, there is a considerable amount of information on how to report abusive content on Twitter here: support.twitter.com

 

Cyberpsychological

But what is arguably more important is to recognise that communication via electronic media is quite different to traditional communication; however, even the most erudite experts in this field still struggle to clearly conceptualise this area. In addition, the majority of research on cyberbullying has occurred ‘largely in the absence of theory’, as scholars are happy to accept a Procrustean solution of cyberbullying being simply another form of bullying.

What follows is not only an attempt to enlighten as to how this is not likely to be the case, but additionally an exercise on my part to understand senders of ‘persistent unwanted communication’ (which will inevitably stray beyond cyberbullying per se and into broader psychology). While admittedly much of what is suggested may require more time and effort than is available, it is hoped that it will serve as a resource in dealing with this phenomenon.

  1. Need for cognition (Cacioppo and Petty, 1984)
    • people differ in their preference for information: some people think and enjoy thinking more than others – such individuals want more information than others.
    • such individuals, who you may infer by the means and manner in which they communicate with you, are more likely than others to prefer a reply which is more rational/logical than emotional/expressive
    • As such, in response it might be wise to:
      • include measure of volume in email (‘so far this week I have received 100 emails on this subject…’)
      • direct to press releases
      • (NB I recommend all politicians use Dáilwatch as a matter of course)
      • include a wide variety information and resources: don’t simply direct to your website, link to as many relevant instances of sought information as possible
    • however, these tactics may not be useful in all cases
  1. Fundamental attribution error (Jones and Harris, 1967)
    • essentially human beings are more likely to interpret another person’s actions as being due to their character, rather than their context
      • this is related to the actor/observer effect – when we are late, it is because of heavy traffic, when someone else is late, it’s because they’re lazy
    • this can clearly lead to a misunderstanding of political actions and a tendency to overvalue a public representative’s personality rather than the constraints they are under.
    • consequently, in dealing with such communication, it is important to remind the sender of the various constraints under which politics and government operate
    • however, additional research suggests that this effect can be experimentally overcome when people are expected to justify their impressions
      • it therefore might be useful to, in cases where this might apply, to remind the individual of the ultimate accountability of the electorate in a democratic society and ask them to justify their opinions (the utility of this tactic may vary according to whether or not the communication is public or private)
  1.  Hyperpersonal model of computer-mediated communication (Walther, 1996; 2007)
    • people take advantage of certain aspects of online communication in order to enhance their interpersonal goals. In other words, people communicate online in ways that they cannot do normally, so as to be able to create effects which they could not do otherwise.
    • This is normally understood in a positive emotional context – e.g. internet dating – but obviously can be used negatively also, as in cyberbullying.
    • According to this model, there are four major features of online communication which people make the most of:
      • ability to edit messages: allows people to change mistakes, think out responses carefully, maximise impact, overcome l’esprit d’escalier
      • asynchronicity: even the most real-time online experiences are no match for face-to-face interaction, there is always a pause of some duration, which can be manipulated for dramatic effect.
      • lack of leakage: while this works both ways, when communicating online, no-one can see how shocked, frightened, smug or happy we are – we only release those social cues which we type out ourselves (such as those that hint at deception)
      • cognitive resources: in parallel to the latter, when we communicate online, we do not have to pay as much attention to the other person as when we would face-to-face, and this mental energy can be focussed instead on message construction
      • consequently, when communicating online, and manipulating all of these features at once, people can create a message which is not only far more emotionally forceful than what they might be able to say in person, but also more hurtful than they might have conceived during its composition.
        • the wisdom of enquiring of a person as to whether or not they would actually state such things were they facing you is debatable, but being conscious of this feature while reading communication is invaluable
  1. Online disinhibition (Joinson, 2001; Suler, 2004)
    • some of the most unusual and also most established research findings in cyberpsychology are to do with a person’s own sense of self. In comparison to when in the real world, when interacting online people
      • have less respect for status and authority
      • feel themselves to be a ‘different person’
    • these effects are due at least in part to individuals’
      • acting anonymously/pseudonymously
      • lack of eye-contact
      • invisibility
      • being in a text-heavy environment
    • difficult to see how this phenomenon can be overcome, however some pointers are pertinent
      • reminding the person of their ‘real identity’
      • direct the message sender to an offline meeting or clinic hours
      • suggesting a video Skype call or Google hangout
    • however it should be also noted that the features of cyberspace which give rise to the ‘toxic online disinhibition’ effect (cyberbullying, harassment etc.), also give rise to a benign or positive effect, wherein people reveal more personal information about themselves than they would offline
    • consequently, the most mature politician’s response to receiving abuse online is, in the same way as one would do while canvassing, is to ‘make notes, not war’, e.g.
      • ignore the brickbats and ask them where exactly they are coming from
      • attempt to understand the person’s reasons for communicating in such a manner
      • enquire of their personal circumstances and empathise with them
  1. Empathy gap (Lowenstein, 2000)
    • a finding from behavioural economics: it is difficult to accurately imagine the consequences of being in a particular emotional state when one is not in that emotional state
      • e.g. when you are sitting at home, warm, comfortable and well-fed, it is difficult to correctly predict how you would feel act when cold, hungry and tired
      • additionally it is even more difficult to understand how someone else would act in those circumstances
    • consequently, you should bear in mind that when someone contacts you in an emotional state, it is difficult for you to truly envisage and empathise with what they are experiencing
    • the only real solution to this effect is to allow your interlocutor time to disengage from that emotional state and delay replying for some time
  1. Just world fallacy (Lerner and Miller, 1978)
    • people like to believe that the environment in which they live is a just and orderly place, where people generally get what they deserve, whether for good or ill
    • this means that when they see a politician making what they believe to be an unfair decision, they expect to see the same politician eventually meet their fate
    • in the event that they do not see this occur, it is not too much of a stretch of the imagination to envisage such a person taking matters into their own hands, using a perceived cloak of anonymity to engage in personally vindicated cyber attacks against their perceived overlords
    • there is no psychological strategy for dealing with this fallacy: politicians are elected largely to deal with public disappointment with an unjust world

 

Nov 192012
 

In which I begin a cross-over between my interests of political intrigue and cyberpsychological methodology. 

Just after a brief trawl of the internet, I’ve found a few cases of a certain class of online behaviour with headline consequences:

  1. The Taliban emailed everyone and visibly included their email address. Roughly 9:20pm local time on a Saturday evening
  2. A Manitoba MP emailed everyone and visibly included their address. Or her team, to be fair. 5:03pm on a Friday evening
  3. And apparently Arthur Spring TD has just emailed everyone and visibly included their email address. Probably his team too, but 5pm on a Monday evening

I’ve found a few others, like the London Metropolitan Police Service emails everyone and includes their addresses, but have no confirmed time for them yet. I am looking for more, and if you have any others (with confirmed time of sending) please comment below.

What is the cyber psychology behind this? Well, first-and-foremost, computer-mediated communication of any kind requires that a user is, to some degree, aware of the fact that, while they are most likely at a computer alone, what they are writing will be seen by many people. It’s a private act that is instantly public.

Now that’s a conceptual leap which most of us are learning to handle reasonably well, but it is still a cognitive transformation which requires some effort, which is why sometimes we post things online which we might wish we hadn’t. We don’t always remember that every message typed online is effectively a speech to potentially millions of people. Conjuring up that image, even unconsciously, requires brainpower which we might not always have available.

But sending the same message to a group of people, who may not know each other, and may not want to know each other, is another stretch entirely. I don’t even know how to conceptualize that! what is it like? is there an offline/realworld equivalent? you can post the same letter to many people, but not instantaneously – that takes time and effort, with stamps and envelopes. Is it like giving a speech to hundreds of people, all of whom are listening to you from individual, blinkered boxes, so they can only see you and not each other? soundproofed, so all they can hear is you, through magic headphones? or something…

I don’t know if there is a real-life equivalent of a BCC email: where a person can instantaneously communicate personally with countless number of people who are anonymous to each other. And that’s what makes it difficult –  it’s hard to be aware of the consequences of that single action: pressing send.

Consequently that’s why there are several examples of this behaviour and related errors (e.g. reply all) and why they seem to occur late in the day and the week, when we’re tired and lacking in ability to concentrate.
How to avoid the CC/BCC error:

  1. Try to only send in am hours, and early in the week.
  2. Construct the list and draft the email separately, combining with care.
  3. Ask someone else to review, and or even send them for you: if possible, institute a two-pass system for all BCC emails.
  4. Save, pause, walk, talk, then edit and check again all communications which are going to multiple recipients.
  5. Use a draft template of regularly used BCC email lists so you don’t have to do this every time.

 

Any other suggestions for maintaining awareness of the invisible audience?

Nov 032012
 

I’ve started blogging this (31st Dáil on Facebook and Twitter and Party flagship accounts) and more to come over the next few days, but here’s the full thing:  Mid-Term Report 2012

In summary:

  • Main party ‘flagship’ accounts:
    • Sinn Féin still way ahead on both Facebook & Twitter accounts support
    • Fine Gael most talked about on Facebook, Labour least
    • Sinn Féin flagship Twitter account most popular, most tweets
    • Fine Gael flagship Twitter account least popular of major parties, fewest tweets too
  •  24thSeanad:
    • Senators more popular on Twitter than on Facebook
    • 67%  on Twitter, average followers 948; updated on average 143 days ago
    • 70% on Facebook, average friends/fans 538; updated on average 151 days ago
    • Labour Senators hold most Facebook support
    • Sinn Féin’s Facebook updates most recent, Fianna Fáil’s least
    • Fine Gael & Fianna Fáil Senators least likely to be on Twitter, Labour & Sinn Féin most
    • Independent Senators take 62% of total Seanad Twitter support (Norris effect!)
    • Sinn Féin’s Twitter updates most recent, Fianna Fáil’s least
    • despite promises of its abolition, both Fine Gael and Labour Senators still reasonably active on Facebook and Twitter, though some have switched off
  • 31stDáil:
    • TDs more popular on Facebook
      • 85% on Twitter average followers 1697; updated in the last 94 days
      • 85% on Facebook, average friends/fans 2342; updated in the last 87 days fractious
    • FG have most total friends/fans, then Independents followed by Sinn Féin
    • On average, SF have most friends/fans per TD, Labour fewest
    • Most recently active Facebook accounts SP/ PBP accounts, least recent FF
    • FG have most total followers, then Labour, followed by Independents (as a whole)
    • On average, Independent TDs have most Twitter followers, FF fewest
    • Most recently tweeted accounts are Socialist/PBP TDs, least recent FF
  • Since the General Election
    • there is no sign of major momentum or threat to the current government
      • FF have held ground but no evidence of growing support on either network
      • FG have strengthened on Facebook, weakened on Twitter
      • SF have weakened on Facebook, strengthened on Twitter
      • but biggest shift is Labour, losing ground on both Facebook and Twitter
      • Independents show strong growth on Twitter, and some growth on Facebook
        • this suggests there is fertile ground for an urban-based, possibly centre-right party
  • Despite huge activity during the General Election, which has eased off to a more regular level, it still doesn’t seem like any party ‘gets’ Facebook or Twitter…

 

Nov 032012
 

One hundred and sixty-six Teachta Dálaí entered Leinster House shortly after the General Election of February 25th, 2011. As has been previously noted in a variety of locations, that election was the most active on social networks, and could be the last where it will be possible to get elected without a presence there.

 

Facebook

Of the 166 TDs, 140 could be identifying as having an account on Facebook, whose total support can be represented as follows. A full directory of those accounts can be found here.

 

It should be clear from this picture why this is one of my preferred measures for estimating party support levels, given how closely the order of support matches recent opinion poll results better than the flagship or Seanad data.

91 of these were regular ‘Friend accounts’, while the remaining 49 were Fanpages. 12 of the Friend accounts did not have a visible total friend count.

 

Deputies Accounts Percentage
Fianna Fáil 19 17 89%
Fine Gael 76 66 87%
Independents 15 13 87%
Labour 38 28 74%
People Before Profit 2 2 100%
Sinn Féin 14 12 86%
Socialist 1 1 100%
WUAG 1 1 100%
Total 166 140 84%

The major outlier in this table is Labour, which is curious, given their showing in previous data. Does this represent the lack of social networking activity by Deputies who have already decided not to contest another election?

Visible count

Total

Average

 Recency

Fianna Fáil

15

33473

2231.53

254.65

Fine Gael

61

129123

2116.77

85.42

Independents

12

41277

3439.75

37.77

Labour

25

39145

1565.80

41.93

People Before Profit

2

5290

2645.00

0.00

Sinn Féin

11

40058

3641.64

35.33

Socialist

1

4819

4819.00

0.00

WUAG

1

4267

4267.00

110.00

Total

128

297452

2342.14

94.56

Again, we see that smaller and left-wing groups are more active – Sinn Féin, PBP, Socialist and Independents. Fine Gael are less current than their government partners in their postings, though with 130k supporters, they will hardly be too worried. Labour may be concerned that their level of support is lower than that of Sinn Féin. On the whole though, the most striking aspect must be high recency, or lack thereof, of Fianna Fáil’s deputies postings. This is unusual internationally for an opposition party, indicating that

 

 

Twitter

Of the 166 TDs, 138 could be identifying as having an account on Twitter. Again, those accounts can be found here and there is also a Twitter list here.

 

Notice how Fine Gael have managed to maintain a similar level of market share dominance at 40% on Twitter as on Facebook. This is even more so impressive as the other major parties support levels show a preference for one or the other network – Labour for Twitter, Sinn Féin and Fianna Fáil for Facebook, as further demonstrated below.

Deputies Accounts Percentage
Fianna Fáil 19 13 68%
Fine Gael 76 67 88%
Independents 15 12 80%
Labour 38 32 84%
People Before Profit 2 2 100%
Sinn Féin 14 10 71%
Socialist 1 1 100%
WUAG 1 1 100%
Total 166 138 83%

Activity on these Twitter accounts can be summarised as follows:

Totals

Averages

Tweets Following Followers Tweets Following Followers Recency
Fianna Fáil

1989

1518

9941

153.00

116.77

764.69

234.58

Fine Gael

42865

17723

92785

639.78

264.52

1384.85

141.61

Independents

4166

3408

43076

347.17

284.00

3589.67

3.11

Labour

26766

15130

52707

836.44

472.81

1647.09

90.39

PBP

1223

314

3429

611.50

157.00

1714.50

2.50

Sinn Féin

4287

1632

19074

428.70

163.20

1907.40

113.89

Socialist

1129

595

13004

1129.00

595.00

13004.00

1.00

WUAG

374

193

170

374.00

193.00

170.00

5.00

Totals

82799

40513

234186

599.99

293.57

1697.00

122.50

Clearly, the smaller parties are maximising their support levels by being very active and posting frequently, whereas the larger parties ostensibly rely on more traditional broadcasting media. At the same time, both Fine Gael and Labour TDs have built up a considerable following by, unsurprisingly, posting a considerable number of tweets. Similarly, Sinn Féin TDs, while this does not seem to be their preferred medium, and are not far behind on over four hundred tweets each on average. What is most striking from this table though, and which explains the meagre slice of this pie which they claim, is the small number tweets posted by Fianna Fáil, not to mention how long ago they were made.

 

Again, it should be noted that there are more Facebook accounts (297k) than Twitter accounts (234k) attached to TDs, but not by much. Relative to the total reach of these networks (2m vs 500k) this implies that Twitter is more politicised, but has a smaller reach. In sum, politicians, both elected and aspiring, should try to understand the differences between the two, and adjust their strategies accordingly.

Oct 312012
 


A separate report will be issued later analysing the activity and interaction patterns on these accounts in greater detail, but for our present purposes we are mainly interested in the support levels on these accounts, as a reference point for the sections to follow. Although there are several parties registered with the Oireachtas, only parties with elected representation therein are included.

 

 

Noticeably, Sinn Féin maintain an impressive lead over all other parties, which is in line with their longstanding appeal in this network, as noted at the General Election (more below). However, when the numbers Facebook reports as ‘talking’ about the parties is included, the picture is somewhat different. The two bottom placed parties, Fine Gael and, interestingly, Fianna Fáil, can some solace from such figures. On the other hand, the two parties with the highest levels of support have the lowest levels of chatter. This could be evidence of the sort of ‘soft support’ traditional pollsters talk about, but also falling support: supporters of a party account who stop interacting with party accounts before they ‘unlike’ it.

  Total Talking % of support
Fianna Fáil

4186

243

6%

Fine Gael

6898

634

9%

Labour

7365

168

2%

Sinn Féin

11708

338

3%

Total

30157

 

 

Twitter
Again we see a significant Sinn Féin presence here, and the sort of Labour showing to be expected. Surprisingly, Fianna Fáil beat Fine Gael into last place, though this may be explained by the table below – the former are tweeting on a much larger scale than the latter. The only point of note however, is that, with little else to go on, considering the number of tweets, and accounts that Labour is following, one might expect more other accounts to be following them.
  Followers Following Tweets
Fianna Fáil

8807

2822

4177

Fine Gael

7510

4002

562

Labour

9855

5674

4642

Sinn Féin

11944

604

6087

Total

38116

It is worth noting, for the time being, the total number of accounts attached to these flagships accounts in Facebook (30k) and Twitter (38k), which suggests, at least slightly, that Twitter is more of a politicised atmosphere than Facebook. This is not a ratio which is repeated, however. More analysis along these lines to follow…