22.  Gathering Data by Interviewing.     

Introduction
Working in "cutting edge" technologies that develop week-by-week means that there are few standard written sources available to answer questions that may have never been thought of until last month.  The process of creating, recognizing, and promulgating so-called standard references or authorities requires a decade or more.  After the research is finished, the paper that reports it must go through the blind referee process and be published.  After it is published a process assessing its significance occurs, suggested by the Science Citation Index.  So-called paradigm studies or experiments are finally described in textbooks that eventually find their way to libraries, depending on the talents of the acquisitions librarian and diligence of faculty members.  Finally, professors and researchers (those who struggle to keep up with current changes in their fields) educate advanced students in the new theories, techniques, and mechanisms.  There is still a slower process at work.  When professors consider which texts to order for classes, they are often influenced by the fact that a text is in its 4th or 5th edition.  
You are most often concerned with the beginning of this process.  Consequently, your source of information is often technical experts who are developing a procedure, system, or product.  Recent estimates of how much time technical writers spend researching topics, including talking to the experts, varies from 40% to 75%.

Asking for the Interview: 
In many organizations, scientists and engineers consider technical writers to occupy a position somewhere between a secretary and a pest.  Consequently, the writer needs to develop a thick hide and the manners of a diplomat.  Be polite even when the person you need to interview isn't.  This "confession" by a technical writer is hardly unique:

Confessions of a Tech Writer       I learned the importance of maintaining good interpersonal relations.  Programmers often consider writers to be secretaries at best and nuisances at worst.        They generally think their job is only to develop programs, not to explain their programs in person to manual writers.  Also, their perspective is narrowly technical.  Thoroughly familiar with the product they invented or developed, they are unable to see it from a user's point of view; a user who is probably a high school graduate with an 8th grade reading level.        When the writer asks the programmer to explain something, the programmer often says, "Don't worry about that, the user will understand" (implying that you are too thick-headed to understand, but the guy buying the program at Walmart will know as much about computers as the programmer.        Of course the programmer doesn't really care about the end user.  That is your job.).  I learned to be persistent, tactful, and thick-skinned.

It is understandable that programmers under pressure to meet deadlines feel that they are paid to program and not to talk to a manual writer; to feel that the writer is an interruption and obstacle.  Perhaps you risk becoming something of a crank, but you may wish to practice the "we are in this together" attitude when requesting an interviews, at the beginning of the interview, and in the follow-up after the interview.  How valuable is a program that end users cannot use?  If the technical expert remains uncooperative and arrogant, you should feel confident that someone in the upper echelons of the organization understands how important your job is; understands that manuals or online tutorials are as important as the program itself.  
We all know about the gap and even the mutual antagonism between the arts and sciences.  Technical writers are often thought to be suspiciously on the artsy side of the divide by their engineering and science colleagues.  To make some of your interviews more pleasant, schmooze your interviewee by indicating that you respect his field enough to have an educated layperson's knowledge of recent events.  Of course, this requires you to actually have such an attitude and do the continuing education work.    

Interviews can be done in person, by e-mail or on the phone.  A personal interview is most often requested by e-mail or the phone.  Be professional, organized, and concise.  

• State your professional identity (job designation) and the purpose of the interview.
• Explain how you chose this person (because of a mutual professional acquaintance, journal reference or authorship, or convention presentation).
• Explain the scope or type of information you need and (perhaps) why you chose this person to provide it.
• If possible, offer to interview the person at his or her convenience.  Watch the calendar.  Holidays and vacations are easy to forget when they are weeks away.

Preparing for the Interview:
You are likely to get out of the interview what you put into preparing for it.  Over the years I have been embarrassed for public interviewers who demonstrate an unconscionable ignorance of the accomplishments of the person they are dutifully and inappreciably questioning.  You do not appear to be professionally competent when you show up and expect the interviewee to tell you what you need know.  You must do some homework.

• In-house and published sources provide essential information on or by the interviewee.  Use Internet search engines to find more material.
• If the topic or field is new to you, spend some time to understand the "big picture" of the profession in which your interviewee is expert.  Study paradigmatic concepts and professional terminology.  There seems to be a "for dummies" book about everything these days.
• Learn what you can about the interviewee and his professional affiliation.  The Internet is probably the place to begin, but there are gaps.  If the person is truly accomplished, there will be lots of secondary material.  If  the person is fairly young she may author Internet documents and have her own Website.  The problem comes with older experts who are not much interested in spending the time it takes to become conversant with the Web, but who are not themselves accomplished enough to have a following who will put material on the Web on behalf of the "star" or expert.
• Brainstorm potential questions.  Write them as they occur to you without regard to order.  After you have enough or feel that you have reached the end of the process, determine a logical order.  Cluster questions under topics.
• When you are arranging questions into an order, don't forget about the audience.  Remember that you are not primarily interviewing the person for your own education, but doing so on behalf of a readership whose interests and needs you should keep in mind.  You may wish to frame some questions by saying, "readers who are concerned with maintaining titanium toasters would  probably like to know what kind of wax to use."

Conducting the Interview: 

• Arrive on time.  My problem is always arriving too early.  A few minutes isn't a problem, but 15 minutes or more becomes imposing.
• Use a tape recorder.  You do not want to ask the interviewee to "slow down" or "wait a minute" while you are scribbling notes.  If your handwriting is as bad as mine, you won't be able to read the notes anyway.  Freed of the burden of trying to taking notes, you will be a more attentive listener and develop better rapport with the interviewee.  
  Examining the tape provides the added benefit of recognizing flaws in your method, such as interrupting the speaker or failing to follow up on important points.
  Try to be as unobtrusive or nonthreatening as possible with the recorder.  Ask for permission to use it.  Explain why you wish to use it, if the interviewee hesitates to give permission.  Then forget it, until it threatens to run out of tape.
• After introductions and small talk, ask for permission to record the session.  Record the name of the interviewee, the date and perhaps the place of the interview.  Then follow your script.  Be pleasant and conversational, but purposeful.  You are there to get information.  It wastes everyone's time and you fail your job by being timid.  Be flexible.  You do not want to go on to the next question when the person has just told you something exciting, problematic, provocative, or otherwise requiring follow-up questions.  
• Your questions should be friendly and objective but not timidly social.  You are not there to make a new friend, nor to debate.  Avoid framing lengthy questions that ask for "yes" or "no" answers.  Have some concern for the fact that your interviewee may feel constrained by security, legal, ethical, or proprietary concerns.  If you are alert to "reading between the lines" about such concerns, you can revise questions hoping to get the information in ways acceptable to the interviewee.  But you are not Mike Wallace.  The last thing you want to do is "poison the well" so that the person resents not only you but the agency you represent.
• The final question is usually, "is there anything else you would like to add."  If the interview has been long, you may wish to mention 4 or 5 major topics and give the person enough time to think.
• When you return to your office, send an e-mail to thank the person for her time and remind them that you are waiting to receive whatever documents they may have told you would help you.

Managing the Interview: 
Be tactful.  

Poor:   "That contradicts what Dr. Smith told me about this."
Better: "I may have misunderstood what you said.  I recently heard that . . . "

Poor:   "Why didn't you design the program to perform this function? Didn't you think of the user's needs?" 
Better: "Would it have been possible to have incorporated this function when the program was developed?" 
             [I know about the passive construction, but in this case we have a reason to be muddy.]
Best:    "Don't be concerned.  The design for the initial release is finished.  The change will be made in 
             the next revision of the program."  

There is a point of balance you wish to achieve between appearing to know nothing about the interviewee's profession and appearing to compete with him by suggesting you know as much as he does.  Over-emphasizing what you know can turn the interview into a competition that will cause you to fail.  Imagine how happy your boss will be when you return saying that you spent your time lecturing the person you were sent to interview.

Some interviewees may be either too terse or too rambling.  The tight-lipped ones apparently think that everyone knows about arcane technical processes.  The rule here is the same as when writing: never repeat or merely record technical terms or phrases you do not understand.  You cannot afford to worry about whether the interviewee thinks you are stupid.  At least half of nearly every class I have taught for 30 years can be characterized as students who are more concerned about appearing to ask stupid questions before their peers, rather than being dedicated to learning the material.  They are either unconcerned about their "C" grades or blame me.  If you cannot follow the explanation the expert is giving you in the interview, how are you going to explain it to your readers?  You can't.  
For the verbose, you may wish to at least appear to be taking notes.  Jot down major "headings" or steps of a process.  Interrupt if you have to in order to clarify the sequence and make each point discrete.  A quickly done visual outline may help the person focus on the discrete steps of the process.  Remember that if you leave without getting the information you need, you have to embarrassingly ask for another interview, rely on e-mail or the phone to clarify points, or simply fail your assignment.

Maybe camcorders are so pervasive that you don't mind being on camera.  I remember a class on secondary teaching methods in which the scariest requirement was being video-taped teaching some lesson for ten minutes or so.  The point was to get a sense of how you appear to an audience.  Your demeanor, as much as your carefully prepared questions, communicates your interest in both the topic and the speaker.  If the speaker thinks you are bored, he probably shortens his answers and hopes for a quick end.  Maintain eye contact without becoming intimidating or giving a robotic stare.  You don't want to appear nervous or timid by shyly looking at your notepad and asking questions in a whisper.  

Confidentiality: 
I have only looked through your Security Comprehensive Handbook (if that is the correct title of MNL-SEC0006, issue C: 1998  Despite its title, I found it to be a fairly well written document).  You know far more than I about confidentiality.  My comment is more in the vein of advice to new employees about office politics.  And since I have almost invariably been on the losing side of those battles, I guess my textbook advice is hardly worth mentioning.  Anyway, the point is to remember that you are comporting yourself as a representative of some organization.  Otherwise your interviewee would probably not have consented to grant an interview to an interested fan off the street or some individual who called him out of the blue.  Perhaps not in a legal sense, there is nonetheless a tacit expectation of professional confidentiality.  This focuses less on the content of the interview (after all it is destined for publication) than on gossip about the appearance of the interviewee: what he looks like (mismatching socks), what his office or home looks like (cobwebs in the corners of the room), his spouse, his dog, the Yugo he drives, the fact that he buys his ties at the Goodwill, etc.  Perhaps worse, that he "obviously" hates his boss, his colleagues, or his dog; that he "obviously" implied that . . . ."      

  Observation / Field Study: 
Explanation by experts is often most useful to understand theory, application, significance, likely effects, cautions, and the like.  In studying fiction, critics recognize "the intentional fallacy," which credits the expressed motive or intention of the writer as more important or better revealing of what the book is about than critical assessment of the work.  The recognition that the critic is the best judge of what a literary work means is somewhat comparable to audience analysis, shifting the concern to the end user.  
Aristotle criticized his teacher, Plato, for not recognizing that a great deal of knowledge is performative.  We know how to walk and use chop sticks and ride bicycles and type and recognize our kids even though these are not theory driven.  They are embodied performances.  There is a pragmatist vein in the philosophy of science (see Michael Polanyi's Personal Knowledge and Thomas Kuhn's Structures of Scientific Revolutions) that argues for the recognition that actual scientific and engineering work is largely performative.  Do you want a surgeon who can explain the theory or one who is less articulate but recognized by his peers as the "cutter" they would go to?  
Unlike anthropologists, natural biologists, or psychologists, technical writers are rarely trained in formal observation technique.  It helps to recognize that some such techniques can help technical writers, especially in the area of writing manuals.

• Try not to simply show up.  Plan observations as carefully as you do interviews.
  • Request permission ahead of time.  Respect always pays.
  • Observe the chain of command to inform, if not request the permission, of intervening authorities.
  • Flowchart or sketch "theoretical" steps of the process in preparation for observation.  If this is a complex process, this stage may be longer than the observation process.
• Do not be so theoretically oriented that you only see what you are looking for; evidence that matches your expectations.  Look for the unexpected or unintended; the accidents, the side-effects.  These may be the very things that your manual recognizes and fixes.  You have more experience than most professionals in relying on performative knowledge rather than mechanically following written guidelines.  You also know the associated problem: the difficulty of rendering a performative process into a number of discrete, formal steps.  Even something as simple as driving a stick shift car prompts the person trying to explain how to pull away from a stop sign at the top of a steep hill to say, "you have to do it all at once -- release the brake, disengage the clutch, advance the accelerator."    
• Observation of so-called "low level" techniques often presents the same problems as interviews: the expert resents having to explain what he is doing to outsiders or lay people.  The further you go down the organization chart -- generally speaking -- the expectation is that sophistication, global concern, and communication skills lessen.  Consequently, you may have to be more concerned with the politics of arranging and conducting the observation than with the data.  To make this easier the next time around, be sure to share the end product with those involved.  When fairly low level workers find themselves "starring" in a video-tape or acknowledge by name in a report, they are likely to be more appreciative than jaded experts who are used to being in the limelight.

Questionnaires: 
Why not stay out of everyone else's business by sending them questionnaires?  Because:

• No one answers them.  Who is going to take time to answer anonymous questionnaires when you can't threaten sanctions for not doing so?  Questionnaires are junk mail that nobody wants.
• The people who do respond often have axes to grind.  They hold strong opinions that they hope to express through the questionnaire, thereby providing something other than "the general opinion" that most questionnaires hope to find.
• You often discover what it is you are looking for midway through a project or an interview.  You cannot afford to rely on this method with questionnaires.  Consequently, the results may be unusable.  Or the requirements for producing a professional questionnaire that goes through pretesting, to discover such problems, may exceed the budget, time, or scope of the project.
• Questionnaires are too casual and/or statistical.  Questions are posed for "yes" and "no" answers.  They cannot produce the kind of information you typically need to write manuals or reports that require expert knowledge.  You do not, for example, have an opportunity to ask follow-up questions or to clarify points that remain unclear or ambiguous.  You must have fully thought out such points when constructing the questionnaire.

Occasionally questionnaires are useful.

• When you wish to get responses to specific questions from a large number of people.
• You can reach a very large potential number of respondents over huge geographical areas with less of the "preselection" that occurs when relying on such techniques such as e-mail interviewing.  
• Respondents to a questionnaire are not influenced by concerns that arise in personal interviews: they like the person, they don't like the person, they feel the questions are too personal or intrusive, etc.  They should have no such extraneous influences. On the other hand, what are you likely to do when a survey questionnaire pops up on an Internet screen as an obstacle to be overcome before reaching information you want?  I suspect most people type in anything or make random selections simply to make the obstacle disappear as quickly as possible.
• Questionnaires generally cost less than personal interviews.

Back to navigation