observer's paradox

tape recorder effect | Jerry Talandis Jr. | March 13th, 2002

Hello,

I'm doing my MET assignment now, and I'm looking for any kind of literature on the topic of "tape recorder effect," sometimes known as "observers paradox." I found this phenomenon taking place in my discussion classes when I recorded my students. I found that they practiced much better with a microphone in front of them than without, and am interested to find out more about this. Could anyone out there guide me to some relevant articles?

Thanks!

Jerry Talandis

Re: Observer's paradox | James Hobbs | March 14th, 2002

Hi Jerry and Jake,

I'd say that the observer's paradox is pretty relevant to Jerry's situation here. As I understand it, Observer's paradox refers to difficulties encountered in trying to observe natural/typical/normal behaviour. That is, the observed are liable to change their behaviour simply because they are being observed, such that their behaviour ceases to be natural/normal/typical. Hence the paradox: you can't observe 'normal' behaviour because people don't behave 'normally' when they are being observed.

So if Jerry is recording his students to determine changes in student performance resulting from an innovation in his teaching, then I see a clear problem; the introduction of the tape recorder is in itself an innovation, so which changes are the result of the teaching innovation, and which are due to the presence of the tape recorder?

Anyway, Jerry, I should stop rambling here as I'm afraid I don't have any articles for you! But a quick Yahoo search turned up 132 hits, so maybe you'll find something useful there.

James Hobbs

PS: ERIC turned up one useful looking article:

TITLE: The Sociolinguistic Paradox: Data as a Methodological Product
AUTHOR: Wilson, John
PUBLICATION DATE: 1987
JOURNAL CITATION: Language and Communication; v7 n2 p161-77 1987.

ABSTRACT: Observer's paradox, the basic problem facing any analyst interested in understanding how people talk to each other within naturally occurring contexts, is discussed. In observer's paradox, by recording or observing everyday interaction, one is contaminating interaction by process of observation. An extension to sociolinguistic method is suggested to facilitate reduction in negative context of observer's paradox (45 references) (JL).

Re: observer's paradox | Jake Kimball | March 14th, 2002

Thanks, James, for your input. That does help but I'm still unsure about one point. Refer to yesterday's ERIC abstract.

'Observer's paradox, the basic problem facing any analyst interested in understanding how people talk to each other within naturally occurring contexts, is discussed. In observer's paradox, by recording or observing everyday interaction, one is contaminating interaction by process of observation'.

If students are preparing conversation cards and analyzing examples of task examples beforehand, is that still considered a "naturally occurring context?"

Jake Kimball

Re: MET: Observer's paradox | James Hobbs | March 14th, 2002

If students are preparing conversation cards and analyzing examples of task examples beforehand, is that still considered a "naturally occurring context"?

Fair point, Jake. Yes, I suppose it would be pushing it to refer to such interaction as "naturally occurring". Allright and Bailey (Focus on the Language Classroom, p70-1) illustrate the observer's paradox with reference to Bailey's experience of observing a class in which the teacher announced to the students "Here comes the spy!", so they clearly think it can apply to classroom observation. But whether this includes a teacher observing his own students in action, you've got me wondering now. As you say, this is obviously not the sort of thing the article referred to is thinking of.

The issue of students changing their behaviour in response to being recorded is certainly one that needs to be considered if you're engaging in such observation, but whether the term observer's paradox includes this the sort of thing .....hmmmmm ....perhaps not. I really don't know. Keith?

James Hobbs

Re: Observer's Paradox | Maria Leedham | March 14th, 2002

This preparation for a task is "natural" for the classroom situation though, isn't it? This point was made by M Breen in a lecture I heard in Lancaster circa 1988!

Maria Leedham

Re: "observer's paradox" or FoF | Jerry Talandis Jr. | March 15th, 2002

Dear Jake and James,

Thanks a lot for your help and discussion about observer's paradox. I've taken James' advice and ran a Yahoo search. I came up with a definition:

'observer's paradox:'

The problem, faced by sociolinguists in particular, that, in observing or interviewing people to find out about their habits of speech, investigators will, by their own presence and participation, tend to influence the forms that are used. First described in these terms by Labov in the late 1960s.

The Concise Oxford Dictionary of Linguistics, © Oxford University Press 1997.

This definition is along the lines of what James wrote, and what I had in mind about the subject.

Jake, you wondered if the improvements I found when taping my students should be attributed to the microphone or the focus on form. James, you mentioned that it would be difficult to determine the cause of the improvement, since the tape recorder was in itself an innovation. These are very good points! I had not really considered them. I can mention this in my assignment.

But, for my assignment, I am interested in simply improving pair work. I want my students to use more English, practice with greater energy, focus, and enthusiasm, to increase their language awareness, and to come away with a means of gauging progress. The question of why my students improved, while interesting, is "beyond the scope of this paper". Bottom line is, they got better when I stuck this big honking microphone in front of them! That's what I wanted. I accepted that I was "contaminating" the observation by recording it, so I played it up. I wanted to increase "facilitative anxiety" (Murphy 1996). I found that creating a "performance event" increased the "on-stage" feeling for students, bringing the "absent but salient audience, the referee" (Hancock 1997) ever closer. I found that they were naturally pushed to put more energy and attention into their talking as a result. Was this because of the microphone or the focus on form? I don't know.

My impression is that when students talk in class, it's like they are performing on a stage. Imagine you are an actor, and all you do is practice all the time. That would get boring after a while, wouldn't it? I may have stage fright, but I know I'd end up feeling unsatisfied if I put all this effort to learn a craft and then never had the chance to use it for real.

Once in a while students need to go out on to that stage and perform. Sure, it can be stressful, but if they constantly shy away from the stage, they can never feel the rush of accomplishment that comes with having gone through the experience and survived. Pressure, creatively and sensitively applied, can really help. This is what I'm really exploring in my assignment, so the actual reason why they improved is not as important than the fact that they did.

I got lots of positive student comments about being recorded. Many went like this: "I felt embarrassed speaking into a microphone, but it was a useful experience." Even the ones who hated being recorded found that it helped them notice their strong and weak points. I was happy about that.

Thanks for your article leads. I'll try to find them. Also, there is a ton of stuff on the net; I should have looked there right away...

Jerry Talandis

Re: Hawthorne beating a dead horse | Jake Kimball | March 15th, 2002

How to determine if the improved speaking is the result of a specific method or the presence of the tape recorder?

If you make a control group, the control group must also be observed, thereby contaminating those results as well. I Spy?

This issue was talked about on another board. Someone there brought up the something known in psychology circles as the 'Hawthorne effect'.

Also, how long would one keep using a tape recorder in class before the effects naturally wear off.

Here is a website that defines the Hawthorne effect:

Hawthorne Effect (Psychology & Sociology): A distortion of research results caused by the response of subjects to the special attention they receive from researchers.

Some other sites on this phenomenon:

Drawbacks of Responsive Evaluation | Donald Clark

Jake Kimball

Re: Observer's paradox | John Bartrick | March 15th, 2002

Hi fellow observers!

I've come across the observer's paradox from a different angle, that of judging the reliability of interview/questionnaire data. Now that I come to think about it, maybe this has a different name, anyway to continue... I conducted an opinion survey among MSc participants in Greece at a weekend workshop. To tell you the truth I hadn't really thought of people answering with a particular audience in mind, until I saw one of the responses that included the observation, "especially Fiona." Fiona Copland was the Aston tutor present at the workshop and the response suggested Fiona was the most supportive, generally OK tutor at Aston. While I'm sure Fiona is absolutely supportive and generally OK, I also realized how easily responses for questionnaires are "corrupted" by a desire to impress one's audience.

Something else. The IIC module, in its glossary of terms gives, "Observer's paradox: This is an expression first used by Labov to describe the paradox that when the observer is present to observe, his or her presence will mean that those being observed will not behave as they normally would. This paradox applies equally well to recording devices, and it is the researcher's task to minimize its effect. The simplest way of doing this is to rely on the fact that when something becomes "part of the furniture" it is ignored".

Conclusion (mine): Get classes used to the sight of the microphone / being recorded before you record lessons that will form data for analysis. As for questionnaires, get classes used to the idea of giving their opinion in questionnaire form, which will hopefully encourage frankness.

Bye for now. I'm off to read "Validity and Reliability in Interpreting Qualitative Data," D. Silverman, 1993 Sage Publications.
[Editor's note: Link is to the 2nd edition of this book, printed in 2001]

John Bartrick

Re: IIC: Observer's Paradox | Janine McNair | March 15th, 2002

This is s subject that can't be avoided if you're doing any A/R which involves the presence of the microphone or of an observer.

I am just completing my IIC assignment that involves analyzing pieces of class interaction where conflict is present. For this I observed and taped a couple of my colleagues' classes, and taped a quite a few of my own. There were a few comments from the 9 year-old learners when they first noticed the presence of the mike, but apart from that I'd say the classes went as normal. If anything I felt more conscious of my language knowing that it was all being recorded. But I think that this is because children adapt more easily to change and have less inhibitions etc, than older students.

As for the two classes I observed, in one case the teacher said the students were "much better" than normal, but these were in fact 12 or 13 years old. While in the other group of 9 year olds, my colleague assured me that the class had gone as normal.

In the IIC Module, Keith Richards points out more than once that as researchers we have to be aware of our data being affected by the phenomenon of the "Observer's paradox", and demonstrates how this problem is certainly reduced when the research is long-term, and the "observees" start to accept the mike/or observer as a permanent fixture, as happened in his own investigation in a language school staffroom.

On more political note, the existence of the observer's paradox is just one of the arguments that we are using in the teaching centre where I work, in our current campaign against enforced "one-off" (not part of any developmental programme) observations by managers (most of whom have furthermore completed no training in observation techniques).

Janine Mcnair

Madrid

Re: Observer's Paradox | Keith Richards | March 15th, 2002

Hi

I've thoroughly enjoyed the exchanges on this and really have nothing to add to the contributions, which taken together provide a much richer and more extensive discussion of this than you're likely to find in any of the usual sources. But James asked me about my views on the conversation cards as a naturally occurring context, so here are my thoughts:

The first thing to remember is that the "Paradox" is just that: two things that sit together in a contradictory relationship. So as an observer you want to observe things as they ordinarily are, but your presence means that they cannot be as they ordinarily are. For physically present observers, that's just the way things have to be - it's a paradox and you're stuck with it.

However, in practice, and especially where tape recorders are involved, it is possible to reduce the effects of this paradox (as many of you have noted) by allowing the recorder to become so much part of the context that it's not noticed and things happen "as they ordinarily would". Occasionally people will be aware that they're being taped and the situation is by definition different (because ordinarily they don't have a recorder there), but for the most part for all practical purposes you're getting data that is "ordinary".

I don't think the conversation cards example changes this at all. The implication of the question is that the preparation of conversation cards and the analysis of examples of task examples beforehand leads to something that is not a "naturally occurring context", and I would agree that in one sense it's not "naturally occurring" (i.e. it's not like spontaneous conversation). However, it seems to me that the answer to the question "Is this therefore exempt from the paradox?" must be no. Even if we were talking just about the student talk, the situation would be different from normal because in addition to the normal audience (the teacher) there would be someone else (the observer). But the real point is that we are not just talking about this particular event. The "naturally occurring context" is the classroom and the lesson that is taking place, and the students' contributions are part of a much bigger picture that definitely is affected by the presence of an observer.

Phew! Seems I took an age to get there again. I'll finish off with a couple of additional but unconnected points.

First, it's worth noting that there are some contexts where the recorder is actually (and always) part of the event, e.g. police interrogations. In these cases, there is no paradox, provided that those involved don't know in advance that the tapes might be used for purposes other than the normal ones. In that case, you could argue that it still applies. (Complicated world, innit?)

Finally, the Hawthorne Effect is well known to researchers using human subjects in experiments, but I don't know whether the Dr Fox Effect is quite as well known. This was an experiment in which an actor gave a plenary address at a conference of doctors and medical researchers. He was introduced by a senior academic (in on the experiment), who gave him a hell of a build up, then he gave a speech that was most eloquent but without any significant content and the audience was asked to rate the value of his talk. It won't surprise you to know that it was highly rated. The Effect refers to the extent to which presentation can override content. Those of you familiar with political developments here in the UK will recognize the extent to which this has been taken on board by the political parties here.

Cheers

Keith

 

Archive Categories