The Five Dimensions of Participant Observations

Thus, in thinking about these various dimensions of participant observation, we come to recognize that the prolonged, highly immersive Margaret-Mead types of ethnographic time series studies would constitute:

1. The left-hand side of Dimension I; and
2. The right-hand side of Dimensions IV and V.

Please note, too, how the right-hand side of Dimension V is ideally suited to purely exploratory, grounded theory work. We may not even have any idea as to what variables or factors we are going in to observe. That is, in fact, our desired outcome: we're hoping that they will 'shake out' as a result of our immersion in the setting.

Also, with regard to the right-hand side of Dimension I, one example of this might be a focus group interview being conducted but also being observed by a one-way mirror on the other side of the room.

Which leads us into a thicket of potential controversy!

It won't surprise you to know that there has been considerable controversy surrounding Dimensions II and III. How much "candor" is desirable? Is too much openness on the part of the researcher likely to pre-bias or 'sensitize' subjects' responses? e.g., halo effect, Hawthorne effect, 'we're being watched so we had better put on a happy face!' On the other hand, is too much secrecy harmful to these subjects? Issues of trust and credibility are natural concerns. And again, these are especially likely to 'nag at the researcher' if he/she has spent considerable, prolonged immersive time with these subjects, as oppposed to, say, an impersonally mailed open-ended survey instrument. You may have come to the point where you know the names and the faces. You've socialized with them, met their families and friends, and so forth. Are you harmfully betraying their trust by either withholding (covert) or blatantly disguising (overt) information germane to the purpose of your study?

Thorny issues, indeed -- and also not surprisingly, there are no clear-cut guidelines to follow. Gray areas can and do occur, along with the potential of erring in one direction or the other.

However, a couple of procedural safeguards do exist. While, again, these hold no guarantees, they do help to focus your awareness of the key issues:

1. Institutional Review Board/Human Subjects Approval Process. Our dissertation-stage candidates are already aware of this one! IRB's exist at major public universities. Their mandate is to review all proposals of research that is to be conducted under the auspices of the university setting - which means any faculty member, student, administrator, staff member, etc., affiliated with that university must 'clear the IRB' before proceeding with his/her data collection. The IRB examines and deliberates on the research proposal from the vantage point of human subjects. Are their rights adequately protected? Will there be any risk or harm accruing to them as a result of any phase of this research? In cases where there might be psychological harm as a result of deception, secrecy, etc., the IRB could require the researcher(s) to attend a hearing in which they are asked additional questions regarding the research. They may also be given some guidance from the IRB as to how to reduce the risks to the subjects. I recall one case when I served on the IRB not too long ago as one of the 2 CEE representatives. The concern over having a 'secret revealed' and associated distress to the subjects was addressed in this way. The researcher, a doctoral student, conducted a followup "extensive debriefing/support" focus group interview with the subjects. This was done in order to probe for any psychological distress they might have suffered as a result of the 'deception.' In addition, in the event that such stress did surface, a university professor who is also a licensed clinical psychologist was also present at these debriefing interviews. His role was to step in as needed at that point, as per his expert assessment, and assist the subject(s) in resolving any such psychological stress.
   
   
2. The IRB is a "must" if you are doing your research as part of the university setting. There also exists a second, non-binding (except of course in the moral sense of a professional code of ethics) set of guidelines. These were established in 1981 by the Joint Committee on Standards for Educational Evaluation. Among other things, these professional evaluation standards advise explicit awareness of the rights of human subjects throughout the research process; sincere respect for the "dignity and worth" of subjects; and sharing the findings and results with such subjects after they are compiled (e.g., the "member check" briefly mentioned in the preceding discussion). This one is akin to "coming clean with them after the fact," with the opportunity for you to observe any such signs of distress and do similar debriefings to those mentioned above if need be. Again, while these are not mandated in a legal sense, they are similar to "official codes of ethics" developed and revised by professionals in other fields and treated with commensurate moral obligation.
   
   
3. Institutional Review Board for the Protection of Human Subjects at NAU

1. The setting - this one follows the caveats given earlier regarding detail and descriptive, rather than interpretive, narrative. Michael Quinn Patton illustrates this very nicely with the following:
   
   

   The room was large enough for a three-person couch across one side, six chairs on the walls next to the couch, and three chairs by the wall facing the couch, which included the door. With twenty people in the room, each person had space to fit, but when everyone was standing there was very little space between people. Several participants were overheard to say, "This room is too crowded."

   Now: compare the preceding to the following entry: "A crowded room!"

2. The human, social environment - could include, but not necessarily be limited to, the following:
   * Characteristics of the subjects (e.g., gender,ethnicity, approximate age grouping, style of dress)
   * Patterns, frequency, direction of interaction and communication
   * Decision making behaviors - who initiates it, who ultimately makes the decision and type/manner of communication regarding the decision
   
   
3. Activities and behaviors - could include, but not necessarily be limited to, the following:
   * Who initiaties the activity?
   * In what way?
   * What were the participants' verbal and non-verbal reactions?
   * What happens at each step of the activity?
   * Who is involved?
   * What is communicated, both verbally and non-verbally?
   * What are the 'closure points' or signals that this activity is about to end?
   * Who is present?
   * What is communicated regarding such intended closure, both verbally and non-verbally?
   * How is the completion of this particular activity related to other activities and behaviors which are observed?
   
   
4. Informal interactions and unplanned activities - ah, the richest and most revealing things -- in life, too, never mind research! - can be the 'surprises!' For example, this could refer to "free/unstructured time" if what you are observing is a planned activity, such as a conference. What do participants choose to do during this time? With whom? This is a wonderful opportunity to, as Patton puts it, "stay open to the data." The researcher might simply mingle with them, as at a coffee break or refreshment table during a conference break time. Furthermore, the researcher's involvement could consist of:
   
   
   1. Explicit impromptu questioning/probing: "What did you think of Speaker X's presentation this morning? How do you think this presentation fits into the theme of the conference?" etc.
   2. Simply listening and observing "on the sidelines," without such direct intervention as in (a), above.

   
   As Patton astutely points out: "The fact that none of the participants talk about a session when it is over is data. [More on this 'non-' type of observation shortly!] The fact that people immediately split in different directions when a session is over is data. The fact that people talk about personal interests and share gossip that has nothing to do with the program is data."

5. The language of program participants - Each field, as we know, has its own 'jargon,' both formal and informal. It is important for you, as the participant observer, to be as familiar with such 'jargon' beforehand as possible. This is so that "when you hear it, you'll know what they mean by it."
   
   
6. Non-verbal communication - we've already discussed this one in terms of proxemics and kinesics. Body language, facial expressions, how people choose to arrange themselves in a room for a group session, customary and accepted ways of greeting one another - all contribute to your accurately and completely documenting the "true lived experience."
   
   
7. Unobtrusive measures - these may be part of thoroughly documenting the setting, as explained in Point # 1, earlier. We've also talked about these in terms of the "traces," or physical clues that are left behind at the scene but which may tell us much about other variables we are trying to understand. We used the example of a group of toys with lots of dust gathered on them at a preschool play center. We also talked about an art exhibit and noticing that the rug is particularly worn in front of certain paintings or exhibits. Such "archival traces" may reveal valuable clues as to underlying patterns, behaviors, choices, etc. even if the subjects themselves are not there when you gather these observations.
   
   
8. Documents - we've briefly discussed these in earlier lessons, as well. Examples include policy manuals, training materials, minutes of meetings, memoranda, computerized data files, etc. These are particularly valuable in comparing whether programs and policies in the field -- as per your on-site observations -- are indeed being carried out and working as intended -- as per the documents.
   
   
9. Observing what does not happen & other surprise findings - Ah, what a goldmine of information! As pointed out with regard to # 4, above, it's the 'surprises' that can be most rich and revealing!

   I like to say the following about this point. In a classic, rigidly controlled experimental design, an unexpected occurrence usually means, "stop the experiment and start again."

   But I ask you, reality-based friends: doesn't life throw us surprises as a matter of course? And isn't there much to be learned by watching and observing how we and others react to those "curve balls of life?" regarding human behavior?

   Thus, in qualitative research, we "joyfully keep going" in the face of the unexpected! Those can provide particularly revealing data regarding how we deal with unexpected events!

   An example of the first, or "what does not happen:" Say that a counseling session for teenage girls with eating disorders is intended to be "highly participative in nature," at least according to the training manuals. Aha - but you observe some of these sessions in action and you consistently notice that the girls do not speak up, volunteer their own experiences, share their own management of the disease, etc. Rather, you notice that the counselor seems to be sort of "top-down" controlling the conversation - doing most of the talking in the form of directives and suggestions. Noticing and recording that an intended event (clients freely initiate conversation) did not happen may be a goldmine discovery to program planners. They now have information that the program, for whatever reason(s), is not being implemented as planned. This discovery and mismatch, in turn, can spur further investigations as to why such intended outcomes are not occurring.

   As an example of the second, surprise occurrences, I like to share the following hypothetical scenario from my own real-life experience as the evaluator of the annual Arizona Leadership Academy. Suppose that an outdoor volleyball game is planned as one of the events. It is intended to foster team building, cooperation and similar positive outcomes. Well, lo and behold, it's monsoon season in Flagstaff. Quite unexpectedly, the skies open up, the torrential rains start pouring down, and there goes the volleyball game. Again, in classic experimental design, this would be a 'failure' and would imply 'start again.' Not so in qualitative! I as a participant observer watch with intense interest as to how everyone reacts to this. Do they get angry? disappointed? frustrated? and then what do they decide to do? I may observe that after an initial 'ain't it awful' gripe session, someone pipes up with, "Say, I know where I can get my hands on a table tennis set. How about if we set up the tables indoors in our large-conference meeting room and have a game?" And then another person chimes in, "OK, that's a great idea!" Before you know it, they've divided up who will go get the equipment,who will set it up and where, how they will choose up teams, etc.! Lost opportunity? Hardly! Did we not learn as much, if not more about our intended target variables of "team building, decision making, and cooperation" by continuing to observe and record the reactions in the face of this surprise development?! Such is the beauty of qualitative research: many such golden opportunities to observe and learn!

VI. Field Notes

You've already got the bottom line on this one: rich, thick, descriptive (as opposed to oversummarized interpretive) detail!

Let me close with yet another vivid example from Patton:

Here, then, is Patton's summary of the key guidelines for doing successful field work:

Go back to Topic 2: The Five Dimensions of Participant Observation

E-mail M. Dereshiwsky at statcatmd@aol.com
Call M. Dereshiwsky at (520) 523-1892

Copyright © 1999 Northern Arizona University
ALL RIGHTS RESERVED