Backstage peformances


Backstage by DanielaNob on flickr

I’ve been re-reading Erving Goffman, who wrote a seminal wee book called The Presentation of Self in Everyday Life, in which Goffman wrote about our daily life as a series of performances – and indeed, a constant shifting between formal performances on the big stage, and informal ‘backstage’ chitchat.

I’ve been working on a social media project for the last few weeks, and one thing that strikes me is the loss of backstage in internet-mediated conversation.    It used to be that you had a bad day and you moaned about it to your sympathetic friends over a pint.  Job done, and no (particular) danger that your friends will tell all their other friends.

It’s getting complicated now.  Whether you lead an online life in public, or pseudonymously, or buried under 130 Facebook privacy settings, your online buddies can often check out what you said unguardedly 6 months ago.   I think we’re aware of this, that there is a large audience who may read our artless thoughts on Lady Gaga or Seth Godin and who may not always approve of our critical stance. ^^

Beyond the response of our known online audience, there is the business of large-scale analysing, scraping, searching, mining, and aggregating.     I link to your post (moaning about your unreconstructed opinions), you read your pingbacks and come over to complain.   Perhaps Seth Godin and Lady Gaga get bored one day and hunt down our complaining reviews, and set their internet pack-dogs free.   Someone finds our angry, stupid comment and complains to the police.   Maybe our social network helpfully allows our text to be searched and aggregated, so that the world can know  what we think of our boss.

We know this, I think.   Perhaps we learn it painfully.  We take care of our public performances, even when they look private.  We might look like we’re merely chatting idly to the make-up artist, but we’re also keeping one eye on the mirror.

Our unvarnished opinions aren’t even backstage.  They’re in the interstices, the cloakrooms, the whispered conversations.    And if you’re analysing free-flowing conversation openly available on the internet, you’re looking at a hell of a lot of minor performances.

The Dark Art of the Telephone Depth Interview

Who's out there?

Old telephone by macinate, on Flickr

People aren’t very keen on telephone interviews.    The phone interview tends to be seen as the (deeply) impoverished relation of the face-to-face interview.  There are plenty of obvious negatives: on the phone, you have no eye contact, no body language, and precious little context.

It’s true that the lack of visual information is a distinct drawback.   There are advantages, though.  Personally, I love telephone interviews.     They can be direct, freeing, and incredibly stimulating to conduct.  People have told me things on the phone that they probably never would in a face-to-face interview.

However, to make phone interviews work as proper, high quality conversations, you need to be highly prepared and work that little bit harder.  Here are my 9 top tips to turn you from shaky salesman into smooth late-night radio host.

1. Have a simple interview guide ready

You need to know your subject inside out before you start contacting people.   Write yourself a concise interview guide  which uses plenty of simple open questions.   The psychology-style interview schedule (20 simple questions) tends to be much easier to work with than a detailed 6-page guide.   Make sure that you really understand the brief before you start – if someone is busy, your time can be cut short.

If you are ringing to make appointments, you’ll find that a proportion of people will give you an interview then and there – you need to be ready to go.

2. Set up and test your equipment in advance

If you’re going to record your interviews, you need to test the whole process.  I use a simple Olympus voice recorder with a telephone jack which plugs into my office line right at the wall.    You can buy them from Maplin in the UK.   Respondents need to know in advance that you will want to record, and you need to ask their permission again at the start of the interview.

Test and double-test the recording.

Also test the phone-to-PC transfer process.  Voice recorders typically have enough memory for a heavy day’s worth of interviewing; unlike old school tapes, it can be painful to label them properly.  Make sure you can transfer and name your files in good time.

A headset is also good for avoiding a cricked neck and sweaty palms.

Have some water nearby, for your throat.

3. Pace your schedule appropriately

Telephone interviews often start late as the respondent forgets the appointment/ takes one last call/ wanders off.  Don’t schedule your interviews too close together, or you will end up chasing your tail all day.   As a guide, an one-hour face-to-face interview will probably take 25-45 minutes on the phone.  People are a little bit briefer on the phone.

Allow a 15-minute gap between interviews to jot down some notes and make the mental break between one interview and the nest.

You need to allow a bigger break after 3 interviews.  Telephone interviews make you even more antsy and high than face-to-face interviews and by the end of the day you will be exhausted.  Build some time in to eat your lunch.  It will be worth it.

4. Introduce and explain yourself

The first few minutes of a telephone interview are critical to the success of the call.   Before calling, focus on the person that you are calling, what you know about them and what you want to learn from them.   Still yourself, if that makes sense.    When you get through, introduce yourself.   Check that your respondent is able to talk; remind them of the topic and how long you think it will take; explain privacy policies and if necessary gain permission to use the voice recorder.

In these initial moments, it’s very important to give the person on the other end of the line time to answer.   People are often very anxious on the phone at first. As soon as they realise that they are not talking to a double-glazing salesman, they begin to relax.

I usually explain that the interview will run like an informal conversation, and give them a quick outline. I then pitch into a very general, non-threatening  introductory question, such as ‘Tell me a little about your organisation and what you do.’   As they talk, listen and comment back, if you can, so that your interviewee knows that you are concentrating on them.

5. Match the respondent’s mood and pace

As you ask your first questions,  tune in to your respondent’s mood and way of speaking.  If they are brusque and quickfire, quicken your own pace to match while staying friendly and ultra-professional.   You are an ambassador for the research that you are doing and the organisation (s) that you are working for.  A no-nonsense individual will still give you a good interview provided that you don’t waffle.  This where knowing your brief becomes vital.

6.  Turn up the volume (project yourself strongly)

Because there is no body language, you will usually have to work harder to communicate a strong sense of yourself as a thoughtful, involved interviewer.

Before I begin a sequence of interviews, I think through my approach and I usually start by adopting a madly friendly, upbeat, intelligent persona which is ramped up at least 30% from my normal laid-back self.    I am probably more informal and straight-talking on the phone than in a face-to-face situation.    I also ‘give back more’ – they can’t see me nodding thoughtfully, for example, so I will make ‘uhuh’ noises and I will comment a lot more as we move from topic to topic:  ‘That’s very interesting’.

I’ll also self-disclose more, where it’s appropriate: the interviewee can probably tell my gender, my age and where I grew up, but that’s about it.

Prompting or giving examples of what you mean is also very helpful.

The embarrassing thing about telephone interviewing is that you quite quickly develop a whole new persona.  I like to think I sound like Kirsty Young; my accent usually drifts from Modified Weedy Scottish to Authentic Hard Scottish when talking to anyone north of Watford.

7. Listen intensely

If you have good questions and you’re genuinely having a conversation, it should flow nicely.  You need to listen hard, to comment back, to question, and to spot ambivalences and incoherency in the acccount that you’re hearing.

8. Sensitive questions are possible

As long as you have a good sense of connection with the person at the end of the line, you can ask anything.  People can be very forthcoming in a way that they might not be in person; plus as you are listening, you can often follow up on issues and concerns that are lurking in the conversational spaces.

As with a face-to-face situation, asking a simple question and then staying silent usually works well; but it only works, I think, if you trust each other. Again, you can telegraph your intentions: ‘The next question is a bit sensitive.’ ‘You might think this is silly, but…’

9. Finally

Close your interview gently, in general conversation.  Thank them lots.   When done, switch off the tape and make some brief notes about your impressions.   After a few interviews, you’ll need to decompress: but hopefully, all that work will have resulted in a good experience for both sides, and a thoughtful, productive interview with plenty of insights.

Any more tips?

Early days in online communities: access and social presence

The community staircase

The community staircase

This is a model of research community socialisation that I developed in a white paper for Virtual Surveys a couple of years ago.   I was inspired by two sources: first, the ‘forming, storming, norming, performing’ model of focus group dynamics that all qualitative researchers have drilled into them; and a similar five-step model developed by Gilly Salmon to account for online socialisation in online learning environments.

Most of the chat about community moderation skills focuses on the higher level issues of discussion and debate.   What I wanted to stress in this model was the importance of the two bottom steps, access (getting in) and social presence  (establishing your voice).

Access is probably the most-neglected element of all.  We might invite people.  We might screen them to find certain characteristics.  Whatever we do, participants do not arrive at an online discussion relaxed, chipper and ready to go.   Like the focus group attendee who’s late for a group, they’ve just been on a journey.  It probably involved an invitation and a link; then it may have involved some registration and some screening; then it may have required that they set up a profile.

If we are lucky, then the journey to community entry will have been smooth and enjoyable, like arriving on a clean train at a nice, well organised conference.  If we are unlucky, the whole experience will have been the satanic lovechild of Facebook, MySpace, and the worst online survey you’ve ever taken; and you arrive, bedraggled, twitchy and suspicious, in the online community space.

So, my first memo to community developers: please please put the same amount of effort in designing the entry journey (invitation and screening) as you do into the rest of the community. It will pay off in happy, soothed participants who are reasonably confident that they know who you are and what they’re doing.

Memo number 2 to developers is to think about your conversation feedback loops.   Assuming that your participants won’t be camped on the site 24/7, how are you going to tell them about new content, and how are they going to find out about answers to their own comments?   Emailed comment notification is usually a good idea; if you don’t use this, you need ways of being very sure that participants will visit and revisit regularly.

The last part of access is welcome.   Once you’ve made it in, it’s nice to get a friendly message with a bit of orientation thrown in, maybe a first task.  It makes you feel wanted and valued.

NB Apparently minor things in this journey can be quite important.   If you give no clues at all about the choice of a screen name (and there is another conversation to be had about that in the first place), do not come crying to me later about Bigbottom29’s sense of being bullied.  Think about your audience and their likely online experience.  The under 20s may crop and upload user images at the drop of a hat;  the inexperienced participant in your over-50s life insurance community may panic and flail. Give them some pre-prepared options they can choose from.

The second stage in community development is establishing social presence. For me, a true sense of social presence iss essential for proper discussion to take place.  This rather fluffy phrase means that participants can easily get a sense of what the community itself is like, what other participants are like, and equally are able to communicate themselves reasonably fully within the online setting.

In a community of passion – let’s say a Dr Who community – this will be done at a personal level through username, avatar/signature and point of view.

In a research community, the initial site content that a new recruit finds will be extremely important in helping them develop an understanding of what the community is all about.    How do new recruits create their own presence?  Usernames, avatars and profiles can all be helpful,  but I firmly believe you need to structure the initial online discussions carefully so that you and your participants can get a full sense of each other right from the start.

A good, simple way of doing this is to have a nicely-designed  Introduce Yourself thread.   Model the introduction carefully (model model model I would say) and (1) you’ll get some lovely data right there  (2) the participants will feel a little bit loved and valued and (3) the participants will start to come to life.  (Is this just me?  There is often a golden moment in an online discussion where you truly begin to understand who the other person is.  It usually comes out of authentic exchange, and it’s really what I’m trying to spark in those initial conversations).

What do I mean by model?  Usually there is some important story or introductory background that your participants want to get off their chests.  Let’s say you’re an online retailer.  You’re mostly interested in response to a new design concept; you’re tempted to rush on to that and make early introductions minimal.   If you leave intros entirely up to participants (perhaps in the interests of saving pixels) they will default to name, age, job that’s all 4 now luv u byeee!!!

You will have saved pixels and a bit of effort from the moderator, but your community will have a rather limited sense of itself;  having started out with brief comments and no strokes/feedback, they may never wind up to giving you more heartfelt or difficult comments on the design.   You didn’t appear to care about them, so why bother?   Good introductory experiences pay back tenfold, just as they do in a focus group.

A better self-introduction task would be to ask the participant to introduce themself and say a bit about the last item of clothing they bought, and their personal fashion philosophy.    If the moderator models this by talking about him/herself, or giving a full example, the participant immediately sees what you mean and has a go.    The participant has now contributed fully; add in a couple of replies from the moderator or another participant with similar tastes, and you are on the way to creating a community that will eventually talk without you. The trick is to pick something that is relevant to you and interesting for participants to do; and in the analysis you will probably come back to this thread more than any other.

Teaching my grandmother to suck eggs, I’m sure.

One final comment on the model: each step has participant views and moderator views.  The moderator just as much as the participant has to create a sense of genuine personality and social presence in the content that they write, the questions that they ask and the replies that they make.   Personally, I like moderators to be able to establish authentic contact, and that may mean disclosing more about themselves verbally than they would do in a face-to-face setting.

Still relevant? Do I need to tweak this now?  What else do you think needs to be in place during the first stages of community formation?

The basic qualitative interviewing kit, part 1

The basic interviewing kit

The basic interviewing kit

I’m on the road this week, for the first time in a while.  Just packing up and sorting through my supplies.  My current interview kit is so tiny that I’ve had to invest in a bright stripy pouch from Paperchase in order to stand any change of finding it in the bottom of my bag. 

What do I have?

Olympus digital voice recorder.  These are about £60 at the moment, and this dinky little recorder connects to the computer via USB.  It takes a single AAA battery (essential spare also pictured).  The micrphone on this machine is quite excellent, producing really good sound quality for interviews and group discussions.   When empty there are about 12 hours of recording time available – good for long interviewing days.  Battery life is a bit skittish – it’s easy to leave on by accident – so spares are still important.

I still slightly mourn the demise of my Sony Professional Walkman, where at least you got a physical cassette to keep at the end of every interview, but these are great.

Sony mini noise-cancelling headphones, around £99.  I love these.  They’re comfortable, they’re pretty, and they do exactly what it says on the tine.  Noise-cancelling headphones make it possible to listen back to voice recordings even when you’re in a noisy environment, like a train.  They are also perfect for listening to music and watching DVDs on your laptop.  White means they’re quite hard to lose, too. 

All this fits in one little pouch – tuck in a spare pack of batteries and even a couple of pens, and you’re good to go.

Are we allowed to talk about downloading?

A few months ago I ran some groups with the usual warm-up of discussing mobile and internet use.  The one difference between this and normal practice was that for this project, we rang up the attendees a few days before the groups and had a short conversation with them.

The intention was merely to check that the attendees were using the specific services we were researching, but it had some interesting effects in the subsequent sessions.

In the focus group, we started off with a nice conversation about Internet habits.   I gradually began to notice that people I’d interviewed earlier weren’t sharing parts of their internet use.    That woman wasn’t talking about her Ebay addiction.  That young man wasn’t talking about his use of dating sites.  As we got on to talking about music and film, one man leaned forward, threw a quick glance at the client who was sitting in, and said, ‘Are we allowed to talk about downloading?’

It was an interesting one.   The observers, I think, thought he was referring to iTunes or the BBC iplayer or Channel 4 on demand.   From the conversation we’d already had on the phone, I knew he was talking about torrenting and Limewire.   Our focus in the research was rather different, so we had a brief  and somewhat coded chat about music downloading and then moved on.

Talking about internet habits in a focus group poses some interesting challenges.  Most people use the internet – YouTube apart – when they’re by themselves.  Ask about internet habits these days and you may be prying into some very private territory indeed.   And that’s well before we get onto s*x.

What are the reasons for not sharing habits in a group?

Websurfing is solitary and private Sharing one’s favourite sites may be like sharing favourite books or TV programmes.   The amount of time spent checking celebrity gossip sites may not be something that the respondent wants to share.

Respondents fear being judged for their interests Both the Ebayer and the internet dater didn’t want to talk about these specifics.     Things might have been very different if the group were composed of like-minded people, but it wasn’t.   These people stayed quiet.  They joined in the discussion of Facebook, because Facebook was something that everyone could share, but they didn’t want to discuss some of the sites that actually meant a great deal to them.   The internet dater was a big user of gay dating sites like Gaydar: talking about that site to a mostly straight group would be a step too far.

Some habits are grey in terms of their legality An in-depth discussion of someone’s torrenting habits may be possible one-to-one, but in a viewing studio with cameras, microphone and three people behind a mirror taking notes, it’s easy to decide not to mention it.

Researchers aren’t aware of what’s out there The researcher who has only ever used Facebook or perhaps read the occasional technology blog does not have a good feel for the myriad of ways in which people connect online.    While some researcher naivety can be helpful, lack of awareness can mean that certain questions never get asked.

Clients may be even less aware (and in any case, are tightly focused on their own organisation’s interests) The typical research client is heavily overworked and either has little time for personal exploration of say, social media, or is in the wrong demographic for it to be second nature.    There are some very web-savvy exceptions, of course.

Research (especially market research) is heavily normative Market research tends to be commissioned by rich white business people who want to sell things to a relatively quiescent audience.   People working in large companies would probably agree that internet privacy is really only a concern for people who have something to hide.

I would argue that the effect of all these forces is to downplay the discussion of messy or problematic habits, especially in group discussions.   The problem then is that the research user ends up with, at times, a heavily edited and skewed version of reality, which may leave out some important yet uncomfortable truths.

The practical implications of the shadow web is that researchers should be aware of group pressures when talking about internet habits, and should, where possible, be digital natives themselves.

Researchers also need to use mixed methods.  Telephone interviews and web-enabled interviews can be far more revealing than a one-and-a-half hour focus group, for some subjects.  Message boards may encourage quiet people to speak their minds.

Researcher openness also helps.  Although it may go against the grain, it can be very helpful to share some of one’s own messy habits.    It sets the right kind of non-judgemental atmosphere.  Once the group knows about your addiction to websites about Jennifer Aniston, they may relax and become more open.

The research client may secretly pity you, but that’s how it goes.

What I did while away, in 7 points

One of the curses of doing work that is usually confidential is that – for me at least – it becomes somewhat difficult to blog about it.   Mix up ‘confidential’ with ‘busy’ and you have a bit of an issue, that I don’t think I’ve really solved yet.   The conventional wisdom is that you write about stuff in the news in your area, and pile in lots of links.    That does work well for many people, but I work across quite a range of areas, and my core area of expertise, market research, is particularly deathly as a news topic. 

I’ve had the pleasure of working on a range of projects for quite varied clients in the last few months, few of which I can talk about.   Some observations, then, from four months of rushing around:

1. Telephone depth interviews are a marvellous method, much underused. They require quite a different approach to face-to-face, but when they go well, there is a great moment of connection with the user at the other end of the phone.  People tell you stuff, in your ear, that they’d probably never confess in a group.

2. Running discussions in viewing studios is like trying to direct a stage play without an actual script. 

3. The most difficult resource on any project is time spent with the client talking about the problem.  These days, many clients spend their entire working lives running from one meeting to another, managing lots of  different things but not able to spend much time on the core work itself. 

4. It is now much easier to get wi-fi connections working in British hotels and offices. Not foolproof, though.  Sigh.

5. The first 10 minutes of any interview or discussion are the most valuable – all the background detail of who, what and when.  We always want to rush on in, but time spent understanding the interviewee and their world is repaid 10 times over when it comes to writing reports.

6. Despite all the moaning about Powerpoint, most of us struggle to write pithy, compelling presentations. What starts out as a jewel of communication turns into an 80-page deck in the twinkling of an eye.

7. Milton Keynes is a so-and-so of a place to drive around in fog.

10 tips for productive online conversations

I’ve been trying to have online conversations for a while now, in different spheres of my life, and here are my top 10 tips for running online forums and groups.  There may be more along later, but these are the ones that occur to me first.

1. Aim for an intimate public conversation

Conversational style is a hard thing to get right.  Personally, I feel that online discussion should be fairly close to natural conversation.  In other words, as an online facilitator (just as offline), you’ll get the best out of other people by being relaxed, genuine and curious.    My best examples of interviewers who do this brilliantly come from Radio 1: just listen to daytime DJs like Edith Bowman or Jo Whiley talk to their audience.  

Who does it badly? News interviewers, especially from the Today programme on Radio 4, manage to be narrow, aggressive and hectoring.  Double glazing salesmen and telephone interviewers, they all want to lead you in a direction that you’re not keen to go…

2. Get the relevant background from people

I often see people acting as if the internet was a scarce resource.  The whole point of having an online forum, say, as opposed to a questionnaire, is so that you can have a broadband conversation, not a narrowband, bounded conversation of the sort that resembles a questionnaire.  Anyway: find out what you need to know about your audience’s lives, work, habits, attitudes…it doesn’t need to be general, it can be very, very specific, but get that background.  It will serve you very well later.  You don’t ask, you won’t get.

3. Ask for stories

Where you can, ask for a whole story, not just the simple answer to a simple question.  Ask your audience to tell you something: their image of Brand X, their first experience of getting a bank account, their favourite night out…whatever.  You may need to give your own examples to get people going, but stories are rich.

4. If you have lots of questions, bundle them into conversations

I see people trying to string out the separate parts of their conversations like beads on a wire.  So, in discussing whether people would like to keep an elephant as a pet, the discussion gets split into attitudes to elephants, ease of housetraining elephants, cost of elephants, propensity to buy an elephant, when quite frankly it would save quite a lot of trouble to simply ask what people think about having an elephant as a pet.   And then follow up with some good questions, to make doubly sure that it is the elephant volume and cost of food that is really turning everyone off, and not the elephant poo.

However, you can quite cheerfully talk about elephants as pets, and then, say, about the religion symbolism of elephants.  Separate conversations.

5. Build trust

Trust is built in various ways.  It’s important to be open, to use everyday language rather than marketing language. If you can be very open about your agenda, then do so.  Share your own attitudes and opinions, if you can.

Trust also builds over time.  In planning conversations, make sure you move from simple and unthreatening questions about habits and experience, to more personal or more demanding ones.  And make sure you listen and respond.   You can’t just go for the jugular – people have to feel comfortable about the conversation or they won’t play.

6. Ask about feelings

One of the oddest difficulties I see is that people who are trained in very rational ways of thinking can struggle with the personal and emotional dimensions of an issue.   So, for example, in discussing the car I drive, a rationally-minded questioner would find out that it has many positive attributes, such as its size, acceleration and build quality.  A better researcher might also dig out the fact that I find it very dull.  (A therapist would uncover the complex reasons why I drive a Honda rather than an Audi, but we’ll leave it there).  

Asking about feelings can be as simple as saying, ‘So how do you feel about X?’   People will tell you.

7. Use projective approaches to ask tricky questions

Unless you want to be a salesman, asking ‘Would you buy this? How much would you pay?’ sounds over-personal and pushy.  In the online environment, it can feel a bit like spam.  You still can ask it, just in a more roundabout way: ‘Do you think other teachers/accountants/forklift truck drivers will be interested in this?  How much do you think they’ll be willing to pay?’

But, I hear you cry, you haven’t asked about them.  *pats you on the head*  Really, 9 times out of 10, they will assume that Other People feel exactly like them.  They project, in other words.  You can always check.  ‘How do you yourself feel?’  It will mostly be the same thing.

8. Use natural language

I used to share an office with a woman who worked in advertising, and on Mondays I’d ask her how the weekend went, and she’d typically say, ‘Well, I went out on Friday and had 5 premium lagers.’  

This is not normal language.  People buy Twixes, not ‘in-hand countlines’.  So, examine your language.  Is it normal? Would your mum know what you’re on about? (assuming she doesn’t work in marketing).  If not, drop the jargon.

9. Allow negative as well as positive opinions

People can be terribly polite online, up until the point where they become incredibly, amazingly, breathtakingly rude.  If your group is sponsored in some way and you want honesty, you have to strive for that all the way through, in authentic questions and everyday language.   Otherwise, you’ll get very bland views.

10.  Arrange time for closure and feedback

And finally: as with real-world discussions, people don’t like to be thrown into the street the minute the conversation is over, they like time to chat, swap opinions and maybe business cards, and generally talk about their experience.  Make sure you use this time, either to follow up and get some feedback, or simply to let people  comment about things that are not yet covered.

 Phew.  Those are my first 10.  There are lots more.  In the meantime, listen to great radio and TV interviewers, and work out what they do.  The kinds of questions they ask ask their subjects or their audiences will serve you well.

Where did all the quallies go?

In getting ready to launch this  journal site I’ve been surfing around to find similar sites.  Although I have found some neat blogs created by web expert types, I’ve found very little on qualitative research.  I was hoping to find the equivalent of one of those anonymous-policemen type blogs which expose the raw underbelly of the job but there is surprisingly little.  There may be a little more that I simply can’t unearth, but at first glance it feels that there are genuinely few people writing about this.  The people I have found, I mostly know already. 

This may be down to fear of compromising commercial confidentiality or it might simply be that commercial qual researchers are too damned busy to write about their lives.