Journal for the Theory of Social Bchauioar 22:2

0021-8308 $2.50

Triangulation Revisited: Strategy of Validation

or Alternative?

UWE FLICK

I. INTRODUCTION

In the study of social behaviour psychologists ~- as before sociologists and

ethnographers - also begin to rediscover interpretive or qualitative research
methods, where the subject under study cannot be conceived with experimental
methods and requires a different approach. Recently, the question of “validity”
becomes more and more important in qualitative research. Generally, there are
two different strategies to find an answer. T h e first one is to apply and modify
traditional criteria, especially reliability and validity (Kirk & Miller I 986,
Kvale 1987).This strategy recently has been sceptically regarded by Liiders &
Reichertz (1986, p. 97), who doubt that “it is possible to apply the elaborate
criteria of a “hard” social research on qualitative research because the
“concepts of reality” of both differ too much”. Similar reservation can already
be found in Glaser & Strauss (1967, p. 224), who “have raised doubts about the
applicability of these canons of rigor as proper criteria for judging the credibility
of theory based on flexible research”. Instead, the authors “have suggested that
criteria of judgement be based instead on the detailed elements of the actual
strategies used for collecting, coding, analyzing, and presenting data when
generating theory, and on the way in which people read the theory”. Here we
find already arguments for the second strategy which starts in two ways from
the main idea of “subject-appropriateness”: As well as research methods should
be appropriate to the subject under study, criteria and examinations used to
their evaluation must be appropriate to those research methods. Thus the
consequence is to develop specific criteria that are appropriate to the methods
to be judged and to their results. In this discussion about “method-
appropriate” criteria (cf. Flick, I 987) of proving credibility and ofgeneralizing
qualitative data and results recently more attention is paid again to the idea of
176 U w e Flick
triangulation. After a short overview of the discussions on triangulation in
different disciplines a methodological framework for its application in quali-
tative research will be outlined and demonstrated on an example of
triangulating mainly the reconstruction of subjective theories - as a special
way of conceiving everyday knowledge - and conversation analyses of
situations in which these become relevant, completed by analyzing the
institutional context and historical backgrounds of such knowledge and
practices.

a. IDEA OF TRIANGULATION - SOURCES AND HISTORY

Nonreactive Measurement: Multitrait-multimetkod-matrix

The history of this idea can be followed back about thirty years and into other
contexts of research. Originally it was taken as a metaphor “from navigation
and military strategy that use multiple reference points to locate an object’s
exact position” (Smith, 1975, p. 273). As a methodical principle in social
research this idea was developed in the discussion about nonreactive measure-
ment (cf. Webb et al., 1966) to find out, “if a hypothesis can survive the
confrontation with a series of complementary methods of testing” (Campbell &
Fiske, 1959, p. 82). Already in this context the initiating questions were, if and
how subjects under study are more or less constituted by the methods employed
for their examination, too. But here we find rather the negative version of
conceiving this fact - that the subject under study might suffer a bias from the
employed methods, that the results might be artifacts. The methodological
strategy to handle this problem was to combine different methods of
measurement in the “multitrait-multimethod-matrix”. Here, strategies of
triangulation are conceived and employed as strategies of validation of
empirical procedures and/or results and not so much as an additional
epistemological source.

Multiple Triangulation -Basic Theoretical Concepts

Denzin‘ (1970/ 1978) introduced the idea of triangulation into the discussion of
qualitative research as “the combination of methodologies in the study of the
same phenomenon” (1978, p 291). By this definition Denzin also conceives
triangulation as a strategy of validation. He distinguishes different types of
triangulation. O n one hand in “data-triangulation” the combination of
different data sources that are examined at different times, places and persons is
proposed. Denzin thinks that this strategy is comparable with theoretical
sampling by Glaser & Strauss (1967).As a second type Denzin suggests
“investigator triangulation”, which means the employment of different
 Triangulation Revisiled ‘77
observers or interviewers to control or correct the subjective bias from the
individual. “Theory triangulation” means “approaching data with multiple
perspectives and hypotheses in mind (..). Various theoretical points of view
could be placed side by side to assess their utility and power” (1978, p. 297).
Central concept finally is “methodological triangulation”, either “within-
method” ~ for example using different subscales in a questionnaire - or
“between-methods”. Denzin further suggests three principles of method-
ological triangulation: “First, the nature of the research problem a n d its
relevance to a particular method should be assessed (..). Second, it must also be
remembered that each method has inherent strengths a n d weaknesses (..).
Third, methods must be selected with a n eye to their theoretical relevance”
(‘978, P. 303).
Finally, at that time, Denzin characterizes the aim of the latter strategy as
follows: “TO summarize, methodological triangulation involves a complex
process of playing each method off against the other so as to maximize the
validity of field efforts” (1978, p. 304).

Ethnographic Fieldwork: Rejlexice Triangulation

Recently, more attention is paid to triangulation in ethnographic field research.
In this context Hammersley & Atkinson (1983, p. 198) argue, that “data-
source triangulation involves the comparison of data relating to the same
phenomenon but deriving from different phases offieldwork, different points of
respondent validation, the accounts of different participants (including the
ethnographer) involved in the setting”. Besides of triangulating data-sources,
these authors suggest “technique triangulation. Here, data produced by
different techniques are compared. To the extent that these techniques involve
different kinds of validity threat, they provide a basis for triangulation”.
Compared to Denzin’s conception, one central point is mentioned further by
Hammersley & Atkinson: “What is involved in triangulation is not the
combination of different kinds of data per se, but rather a n attempt to relate
different sorts ofdata in such a way as to counteract various possible threats to
the validity of our analysis” (1983,p. 198).Because of mentioning this aspect,
the authors call their conception “reflexive triangulation”.

Combination of Qualitative and Quantitative Melhods

Finally, the combination or “mixing” of qualitative and quantitative research
is labelled with this term currently (Jick, 1983,Lamnek, 1988). I n this context,
a central question is, how to conceive the relation ofboth types of methods and
their results. Thus, triangulation gains further relevance, since “qualitative and
quantitative approaches are complementary rather than competitive methods
(..and the) use of a particular method (..) rather must be based on the nature of
 1 78 U w e Flick
the actual research problem at hand” (Wilson, 1981,p. 58). In this conception
quantitative research is no longer per se dominant and superior to qualitative
research and the first is not per se the instance of evaluation for the latter. This
will be further discussed later on.

3. CRITICAL DISCUSSIONS: FROM VALIDATION TO IN-DEPTH-UNDERSTANDING

Denzin’s earlier conception of triangulation is not only the most discussed and
quoted one. It also has been objective of most critique on triangulation in
general. For example ethnomethodologist David Silverman (1985, p. z I )
criticizes Denzin’s assumption of a “master reality in terms ofwhich all accounts
and actions are to be judged. This casts great doubt on the argument that
multiple research methods should be employed in a variety of settings in order
to gain a “total” picture ofsome phenomenon (..). Putting the picture together
is more problematic than such proponents of triangulation would imply. What
goes on in one setting is not a simple corrective to what happens elsewhere -
,each must be understood in its own terms”. Here is argued -ifwe think a little
further - that Denzin neglects the basic idea of triangulation, that was the
starting point of the discussion in Webb et al. (1966)- the reactiveness of
methods or in other words: That every method constitutes the subject under
study in a specific way. Not before users of triangulation make allowance for this
problem sufficiently, they can ignore the following critique by Fielding &
Fielding (1986,p. 33): “Multiple triangulation as Denzin expounded it, is the
equivalent for research methods of ‘correlation’ in data analysis. They both
represent extreme forms of eclecticism”.
Both critiques point at Denzin’s conception of triangulation as a strategy of
validation in the classical sense of the meaning, which assumes one reality and
one conception of the subject under study independent of the special methodical
approach - in Denzin’s words “the same phenomenon”. Thus Fielding and
Fielding (1986,p. 33) condense their critique of Denzin’s conception in the
following programmatic argumentation: “Theoretical triangulation does not
necessarily reduce bias, nor does methodological triangulation necessarily
increase validity. Theories are generally the product ofquite different traditions
so when they are combined, one might get a fuller picture, but not a more
‘objective’ one. Similarly different methods have emerged as a product of
different theoretical traditions, and therefore combining them can add range
and depth, but not accuracy”. Central point of these critiques is the technicistic
conception of triangulation in Denzin’s program. Although he suggests
triangulating theoretical perspectives, too, his concept of methodological
triangulation as a strategy aiming at validity neglects theoretical differences
between methods and the way each of them constitutes the research object. In
the idea of playing methods off against each other and in the idea of the same
 Triangulation Revisited ‘79
phenomenon, not only the complexity of triangulation in general but also the
potentials of this strategy are simplified and artificially reduced.

Using Triangulation as an Alternative to Validation

Through the attempt of gaining validity by way of triangulation on one hand
the subject under study is necessarily objectivated. O n the other hand the
question arises, which result should be preferred, if different methodical
approaches lead to contradictive outcomes. Corresponding to these critiques
several researchers try to find a way out of the outlined dilemma by discussing
triangulation no longer as a strategy of validation but as an alternative to
validation. So Fielding and Fielding (1986, p. 33) suggest as a rksumk of their
critique of Denzin: ‘ “In other words, there is a case for triangulation, but not
the one Denzin makes. We should combine theories and methods carefully and
purposefully with the intention of adding breadth or depth to our analysis but
not for the purpose of pursuing ‘objective’ truth”. Also Lamnek (1988, p. 236)
pleads for expecting rather complementary than congruent results by tri-
angulating different methods. Finally, these conceptual aims are summarized
by Kockeis-Stangl (1982, p. 363) as follows: “‘Instead of talking about
validations, perhaps it would be more adequat to see our control-processes as
more-perspective triangulation (..) and in advance to be prepared for receiving
as a result no uniform picture but rather one of a kaleidoscopic kind”.

Denzin’s Reaction to his Critics and his Current Position

As already mentioned, Denzin’s current position has changed. In the third
edition of his manual (1989a) he is taking a more consequent interpretive
position (see also Denzin I 989b, I 988) towards research and especially towards
the status and use of methods. This revision also affects his position towards
triangulation. Here (1989a, p. 234sq.), he reacts directly to the critical
comments of Silverman (1985) and Fielding & Fielding (1986) we already
discussed. He also abandons the idea ofvalidity as a main purpose for the use of
triangulation, which he now characterizes as follows (198ga, p. 236):
“Triangulation, or the use ofmultiple methods, is a plan ofaction that will raise
sociologistsabove the personal biases that stem from single methodologies” and:
“..triangulation of method, investigator, theory, and data remains the soundest
strategy of theory construction”. So, Denzin keeps the types of triangulation he
proposed earlier and we discussed above, even though he now stresses more or
less different points in their definitions and descriptions. For instance, he holds
that “between-method triangulation can take many forms, but its basic feature
will be the combination of two or more different research strategies in the study
of the same empirical units” ( I 989a, p. 244, my italics). The shift in the purpose of
using triangulation is shown by Denzin as follows: “. . .the concept ofhypothesis
I 80 Uwe Flick
testing must be abandoned. The interactionist seeks to build interpretations,
not test hypotheses” (1989a, p. 244). With regards to the critique by Fielding
and Fielding (1986),Denzin now sees multiple triangulation as follows: “The
goal of multiple triangulation is a fully grounded interpretive research
approach. Objective reality will never be captured. In-depth understanding,
not validity, is sought in any interpretive study. Multiple triangulation should
never be eclectic. It cannot, however, be meaningfully compared to correlation
analysis in statistical studies” (Denzin 1989a, p. 246).

New Question5

Authors as Fielding and Fielding or Kockeis-Stangl, and recently also Denzin,

see the relevance of triangulation not as a strategy of validating qualitative
results but as an alternative to this and as a “method-appropriate” strategy of
founding their credibility. Ifwe want to check the relevance of triangulation in
the way outlined last, two questions arise: First, we have to ask, how to proceed
in order to gain a really many-sided kaleidoscope and a picture of the subject
under study, that is really including different perspectives. Second, we have to
ask, if and how we can solve the problem of founding the credibility of
qualitative data and results this way without recourse to quantitative research.
This question is even more important as the opinion starts to succeed that
qualitative research should no longer be seen as an explorative first step (only to
develop hypotheses), whose results subsequently must be tested by “hard”
methods, but rather as a self-reliant branch ofresearch with its own legitimacy.
For both questions an answer shall be outlined in the following (see also Flick
1992 more generally for this conception). Because these questions are discussed
on the background of the state of the art of qualitative research in Germany,
some preliminary remarks on this state are necessary,

4. QUALITATIVE RESEARCH I N GERMANY - STATE OF T H E ART

Periods of re-discovery
Ifwe want to have a look at qualitative research in Germany, first ofall we have
to distinguish sociology and other social sciences on one hand and psychology
on the other hand. While all those sciences can look back on a long tradition in
using understanding as a n epistemological principle (see Bergold & Flick I 987
for this topic) and all of them at some time have started to ignore those
traditions, there are great differences in the renaissance of qualitative research
in each case. In the social sciences we consider an import of the discussion in the
North-American sociology in the sixties concerning ethnomethodology, sym-
bolic interactionism, ethnography and so on to Germany. O n the detour of the
 Triangulation Revisited 181
import of this discussion we see also a rediscovery of their theoretical sources as
phenomenology and the sociology of understanding by Alfred Schiitz, Max
Weber and Georg Simmel. First hints on the new methods in the USA were
given in the late sixties by Jiirgen Habermas ( 1967).More attention was paid to
them in the beginning of the seventies, when a group of sociologists from
Bielefeld (among them for example Fritz Schiitze) translated, commented and
published basic papers of Blumer, Garfinkel, Sacks, Cicourel and Wilson
(Arbeitsgruppe Bielefelder Soziologen 1973). Then a discussion on the
relevance of the “interpretive paradigm” started and some more collections of
translated basic papers on qualitative research in the USA followed
(Weingarten et al. I 976, Hopf & Weingarten I 979, Gerdes I 979). In this period,
the conception of the research process proposed by Glaser and Strauss (1967)
with the ideas ofdeveloping theories in empirical research instead ofonly testing
them and of abandoning theoretical preconceptions of the subject under study
as far as possible played an important role in the discussions (Hoffmann-Riem
I 980, Kleining I 982).
In German psychology, on the other hand, this renaissance of qualitative
research was long time ignored. Not before the middle of the eighties at least
some psychological books on these topics (seeJiittemann 1985,Bergold & Flick
1987)were published in Germany.
The next step was in the late seventies the beginning of an independent
methodological discussion in German sociology which led to developing new
methodologies which also set standards for the methodological discussion in
general. O n one hand the “narrative interview” was proposed by Fritz Schiitze
(1983) as a specific method of collecting data aiming for the stimulation of
autobiographical narrations of biographical processes. This method is mainly
used in the context of biographical research. The development of this method
and the discussion it stimulated are one reason for that most of the discussion on
qualitative methods in Germany and most of the practical research is centred
around interviews, while participant observation here never played a role as
important as in the United States and in the history ofqualitative research. For
the application of interviews, the narrative interview was discussed as the most
consequent way ofgiving way for the respondent’s point of view, because the
main demand for the interviewer is not to interrupt or disturb the respondent’s
narration.
A second method developed in the German discussion is that of “objective
hermeneutic” proposed by Ulrich Oevermann and colleagues (1979). This
method of inlerpreling data is aiming for latent structures ofsense by interpreting
verbal protocols ofinteraction and other sorts of text. It is based on theoretical
conceptions of psychoanalysis? Levi-Strauss, Piaget, Bourdieu, Chomsky and
others and mainly makes a difference between the subjective meaning ofactions
and interactions on one hand, which could be reported by the respondent, and
their objective or latent sense on the other hand, which only can be seized by
I 82 Uwe Flick
extensive interpretations. With its exhaustiveness of interpretive effort and
rigor, this approach was discussed as the most consequent way of interpreting
any sort of data, mostly everyday interactions, but also written texts, pictures,
interviews and so on (see also Denzin 1989b for discussions of these methods).
A third prominent approach developed in the German discussion on
qualitative methods, this time in psychology, is reconstructing subjective
theories (see Groeben 1990). Here, after a half-standardized interview, a
graphique representation technique is learned and applied by the respondent,
who is studied, to visualize the contents and structures of his/her subjective
theory in the way he or she sees it. This procedure aims for grounding data on
the consensus with the respondent and is therefore also discussed by using labels
as “communicative validity” or “dialogue consensus’’ (see Scheele I 991).
Finally, many studies in German qualitative research are influenced by
ethnomethodology and especially by conversation analysis (Bergmann I gg I ).
But this discussion, which had its strongest influence on the general discussion
on qualitative research in the early eighties, kept very close to Anglo-american
discussions.
Now, after a period of importing discussions from United States sociology,
and a period of developing genuine methods and, in some aspects rather
radical, methodological discussions concerning the “right” way of qualitative
research, in the late eighties, a period of wide-spread application of qualitative
methods in empirical studies in sociology, pedagogy and -with some delay -
also in psychology followed. These developments have recently been docu-
mented in great detail (see Flick el al. 1991) and some manuals have been
published by German authors (Spohring I 989, Lamnek 1988, 1989). Finally,
the discussion on criteria for the evaluation of research and results obtained by
qualitative methods has been intensified (see Gerhardt 1985, Flick 1987, Kvale
1991).

Current Research Perspectives

Also it can be stated, that there is no longer one single qualitative research, but
that different theoretical and methodological perspectives of research with
different methodical approaches and theoretical conceptions of the phenomena
under study can be identified within the field ofqualitative research. Recently,
several attempts to structure this field with its variety of methods and their
theoretical and methodological backgrounds have been undertaken.
Luders and Reichertz (1986, p gzsq.) bundle up the current multifold of
qualitative research (at least in Germany) in research-perspectives “aiming at
( I ) the understanding of the subjective sense of meaning, ( 2 ) at the description
of social action and social milieus and (3) at the reconstruction of in-depth-
structure generating meanings and actions”. For the first perspective, the
concentration on the respondent’s viewpoint and experiences and the “maxime
 Triangulation Revisited I 83

to d o justice to the respondent in all phases of the research process as far as

possible” are characteristic features. These goals are mostly pursued by using
interview strategies. I n the second perspective methodical principles are
“documenting and describing different life-worlds, milieus and sometimes
finding out their inherent rules and symbols”, which are realized for instance
through conversation analyses. I n the third perspective, subjective sense,
intentions and meanings as surface phenomena are differentiated from
objective in-depth-structures as an own level of reality, that generates actions.
This differentiation is methodologically realized mostly by using Oeverman’s
objective hermeneu tic (Liiders & Reichertz I 986).
Bergmann ( I 985) on the other hand discriminates “reconstructive methods”
(for example interviews or participant observation) and (in a strict sense of the
term) “interpretive methods” (conversation analysis, objective hermeneutic) as
fundamentally different approaches. While the first group of methods is
employed to produce data (by questions or interventions in the field) to reconstruct
events and participants’ viewpoints for the purpose of research, in the second
group research activities are restricted to merely record and interprete social
activities in their ‘natural form’, using a “natural design” (Nothdurft 1987).
Each of these approaches discloses or obstructs different points of view on the
phenomena under study.
Bergold and Flick finally ( 1987, p. 5f.) structure the spectrum ofcontexts, in
which statements and actions of respondents studied with qualitative research
can be categorized, by the poles “categorization in structures localized in the
individual subject” (for example by reconstructing subjective theories) and
“categorization in socially shaped patterns of interaction” (for example by
conversation analyses of mundane talk). Other authors suggest comparable
taxinomies of qualitative research.

5. SYSTEMATIC TRIANGULATION OF PERSPECTIVES

T h e potential of triangulating different qualitative approaches in the way

outlined earlier lies in combining such different perspectives and in focussing on
aspects ofthe subject under study which are as different as possible. This way the
subject respectively will present itself in the way it is constituted by each
method. Triangulation will be most fruitful if the selection of triangulated
perspectives and methods is substantiated: “What is important is to choose a t
least one method which is specifically suited to exploring the structural aspects of
the problem and at least one which can capture the essential elements of its
meaning to those involved” (Fielding & Fielding 1986, p. 34, my italics).
Correspondently methodic approaches should be combined that realize several
of the perspectives of qualitative research mentioned by Luders and Reichertz
and others - for example on one hand interviews reconstructively allowing to
‘84 Uwe Flick
understand the subjective meaning of the problem for the respondents and on
the other hand conversation analyses allowing to describe social action and
disclosing an interpretive approach to structural aspects of the examined
phenomenon. So we get different “sorts of data” that can be related to each
other (cf. Hammersley & Atkinson, I 983) and present complementary aspects
of the subject under study. As the following figure shows, the triangulation of
conversation analyses with interview strategies for example to reconstruct
subjective theories can serve for this purpose by realizing two different
perspectives as proposed by the authors mentioned above:

Systematic Triangulation of Perspectives

Authors Perspective I Method I Perspective II Method I1

for example for example

Liiders & Description of Conversation Understanding Interviews

Reichertz social actions Analysis subjective sense
(1986) and milieus of meaning

Bergmann interpretive Conversation reconstructive Interviews

(‘985) approaches in Analysis approaches
the strict sense

Bergold Socially shaped Conversation Structures in Interviews

& Flick patterns of Analysis the individual
(1987) interaction subject

Fielding & structural Conversation Meaning of the Interviews

Fielding aspects of the Analysis problem for
(‘986) problem those involved

Fig. I .

Systematic triangulation of research perspectives in qualitative research was

applied in a more general context of multiperspective triangulation that will be
discussed in what follows:
 Triangulation Revisited ‘85
6. AN EXAMPLE FOR THE APPI.IC.4TION OF SYSTEMATIC TRIANGULATION OF
PERSPECTIVES

Multi-perspective Triangulation of Micro and Macro Perspectives

In a study on trust in counseling in socio-psychiatric services (Flick 1989),
multiperspective triangulation was used on different levels:
- First of all, relevance and subjective meaning of the phenomenon for the

investigated counselors was examined by reconstructing their subjective

theories, (cf. Scheele & Groeben 1988) on trust in half-standardized
interviews and the use of a graphic representation technique. So the
contents of the interviews were reconstructed in dialogue consensus with the
counselors who could develop the structure of their subjective theory
themselves.
- These reconstructions were triangulated on one hand with conversation

analyses ofcounseling talk, by which relevant social action descriptively was

interpreted and the structural aspects of trust were examined.
- O n the other hand, these data further discussed below, which were collected

and interpreted in a micro-analytic perspective, were triangulated on a

macro-analytic level with studies concerning the history of the institution
since the beginning of the century, and the official versions of the aims and
tasks of the institution as well as with analyses of the official statistics on the
institution’s activities with clients over a period of twenty years. These
macro studies showed shifts in the institution’s tasks as well as in its main
activities from putting people under legal control and forcibly sending them
to hospitals to helping people to stay in their everyday context by offering
help and counseling according to principles of “community psychology”.
These studies showed the increasing relevance of the phenomenon of trust
for professional work in this institution and for winning and maintaining a
professional identity as a helping professional or a counselor instead of
conceiving oneself as an agent of control as typical for professional work in
this institution in earlier periods of its history.
But, in what follows, the focus will be laid on the triangulation of the two
qualitative methods and their theoretical backgrounds.

Two Levels of Triangulating Qualitative Research Perspectives

This systematic triangulation ofperspectives again was employed on two levels:
First, on the level of the single-case to answer the question ifrelations between a
counselor’s subjective theory and a consultation he or she had done with a client
can be found. In the sense of credibility-founding this shows functionality and
action-relevance of the single subjective theory for counseling in the examined
context. Second, on the level of comparative analyses: Comparative system-
I 86 Uwe Flick
atization of the course of counseling shows regularities. If subjective theories
shall be functional for those forms of talk and counseling, they must contain
representations of those regularities found i n the different course of talk. This
way a set of categories can be developed out of one source of d a t a
(consultations), which can be used for interpreting the other source of d a t a
(subjective theories). Based on these findings the entity of examples can be
interpretatively evaluated in the last step. Now, these relations shall be
demonstrated in two concrete examples of triangulating a counselor’s sub-
jective theory with the interpretation ofan excerpt from a consultation he or she
had done with a client.

7. TRIANGULATION OF SUBJECTIVE THEORIES AND CONVERSATION ANALYSIS - T W O

EXAMPLES

In the following two examples of triangulation subjective theories a n d

conversation analyses of consultations shall be demonstrated on the level of
single-case analyses. This may demonstrate the different perspectives that each
of the two methodical approaches discloses.

7 .I Convergeant Perspectives: Completing the Kaleidoscope

Thus, in the first step a short excerpt of the beginning of the counseling
interaction between a female social-worker (B,) and a male client (K, )will be
presented (see Flick, 1989for further examples of complete interpretations):

B,: Hmrn, well, your grandfather came to us (Kt: yes), ne, he seemed to be very
worried about you?
K , : Yeah, I was feeling quite bad
B,: Yes, what was the matter at that time?
K , : In May, (.) you know, I drank too much a couple ofdays in a series and then I was
feeling so0 bad, because of the circulation (B,: hmm), well everything, what you-,
break out of sweat (Bi: hmm) raving of the heart, uuuh, burning eyes and
everything, anyhow, and I didn’t feel like laughing at all
B,: And then your grandfather also said, uh, well (.) your family doctor had said,
meanwhile you are in a very urgent danger of death. Do you have an urgent
organic-
K,: Wellwell, danger of death
B,: complaints?
K,: not really, ne? (B):hmm) There’s just my fear, if I carry on that way, that still
might come, (B,: hmrn) and that must not really happen, you know. I don’t lay
any stress on this (B,: hmm) and therefore it’s kind of a thing about drinking in
my case.
B,: How did it begin?
 Triangulation Revisited ‘87
If we have a look at this opening sequence of a consultation, we see that the
counselor does not start with the general exploration (“How did it begin?”)
immediately. This would be a typical action at this point, as the other examples
analyzed make clear. Here she doesn’t go the usual way before clearing up step
by step, whether there is a n urgent situation given in the client’s case. This
would imply an acute demand of action -for example to present the client to a
physician in the team (B, is a social worker). Ifwe have a look at her subjective
theory, it -as Schiitz says “at one look - monothetically” ( I 972) makes clear,
why the counselor at this point first communicatively - and that means in the
sense ofSchiitz again (1972p. I 50) “polythetical1y”- runs through the process
of clearing documented in the course of the interaction. Step by step she
discusses all the hints for an urgent situation with the client - the fear of the
grandfather and the reported statement of the family doctor. Not before
clearing up these points successively, she can turn to the ‘normal’ making of
relationship and trust. In her subjective theory this process of experience is
condensed in the conceptional meaning-coherence “client comes in a situation
of urgence (which hinders) development of trust”. So we can find in our
example the following structure of sense resulting of comparable experiences in
the counselor’s subjective theory (see Flick 1989, p. 377 for the complete
subjective theory):

contact comes about in an urgent

situation

trust between
counselor always has in mind to D client and counselor
observe if any strange, suspicious
facts appear because ofwhich she has

Fig. 2 .

Structural Aspects, Subjective Meanings

In triangulating these two perspectives systematically we received both aspects
that Fielding and Fielding require: O n one hand we can “explore structural
aspects of the problem” under study - how does the counselor organize the
social situation of counseling and what is the special demand in this specific
situation? How does she try to win the client’s trust this way? If we now had
more space to continue analyzing this example ofcounseling, we could see, that
this process ofclearing is part ofa bigger processing ofclearing with the client the
informations the counselor had received about him in advance, so as to install a
direct and dyadic relation this way. In this relation, informations, maybe
I 88 Uwe Flick
prejudices about the client are transformed into shared knowledge and can
become the object of mutual negotiation between counselor and client. T h e
client gets the status of a participant in the interaction instead of being only
object of professional actions and the counselor’s interactions with third
persons. Such a process of explicating the presuppositions of the interaction
proved to be one central step to installing and developing trust in counseling
(Flick 1989).So the analysis of the conversation shows, how the counselor tries
to cope with structural demands of the situation: First, a demand of explicating
the preliminaries and second, in difference from typical course of comparable
sequences not only explicating, but also a demand of clearing the relevance of
informations she had received from others about the client.
On the other hand we can “capture the essential elements of its meaning to
those involved” - at least for the counselor by reconstructing her subjective
theory. This shows, which meaning and relevance the given situation has for her
in accomplishing the client’s trust and what kind of demands she has to cope
with -to make clear, ifshe can handle the case alone or ifshe has to involve the
physician. Also the subjective theory shows, that the counselor attributes a
hindering effect on the client’s trust to both - the given situation and the
potential need to cooperate with a colleague. Thus we can see, what stock of
knowledge the counselor uses in defining and handling this situation.

Theoretical Background: Knowledge and the DeJinition ofthe Situation

Schiitz and Luckmann (1975,p. 154)point out the implications ofsuch a stock
of knowledge for the constitution of subjective meaning in the current situation
of action as follows: “Acquisition of knowledge is sedimentation of current
experiences in structures of sense according to relevance and typicality, that on
their part influence the definition ofcurrent situations and the interpretation of
current experiences”.
The part of the counselor’s subjective theory mentioned here (shown in Fig.
2 ) goes into the definition of the situation insofar, as the need of answering the
question - whether there really is an urgent situation given in this case and
whether the corresponding “imperative of action” (cf. Wagner et al., 1 9 8 1 )
should be realized here (to present the client to the physician) - becomes
directive for the actions of the counselor in the first phase of the conversation as
opposed to the ‘normal’ run of such sequences. Application of this stock of
knowledge in the situation of action should not be understood as a deliberate
and explicit application of those theories. In the situation of action they rather
become activatable and activated tacit knowledge (cf Polanyi, I 966).
So, in this example, triangulating these two approaches to trust in counseling
offers two perspectives that complement one another: The subjective theory (or
the part ofit mentioned here) furnishes an explication for the deviation from the
usual course of counseling interactions we found in this example. These
 Triangulation Revisited ‘89
concepts give an explanation, why the counselor first checks out this special
problem before starting a general exploration of the client’s situation and
problems, which is -as the other cases in our study show as well as those we find
in the literature (cf. Wolff I 986) -usual for this phase ofcounseling. Ifwe have
a reverse look on the two data sources, the conversation analysis of this
counseling shows how the subjective theory influences the counselor’s actions in
the interaction with a client. There are other pieces ofevidence in this case that
could be presented. So, this shows that the subjective theory is relevant for the
subject’s actions. We get a fuller picture by combining the two approaches: we
get the subjective perspective on ‘trust in counseling’ and on the professional
work with clients in this institution by one approach, while the other one
furnishes insights on the problems in realizing such concepts in concrete
interactions and under concrete institutional circumstances. So triangulation of
these two perspectives makes the picture of ‘trust in counseling’ more
kaleidoscopic and adds range and depth to our understanding of the complexity
of the phenomenon.

7.2 Divergent Perspectives: Ideals and Failure

Ifwe now have a look on our second example, we find the following concepts in
the subjective theory amongst others (see Flick 1989, p. 387 for the complete
subjective theory):

Trust between
client and counselor

The institution as bureaucracy Counselor meets most clients only

\
Counselor is clicked into Counselor’s oFfers to
in bad situations the client’s life somewhere the client must be
and clicked out again carried through
Counselor sits between the
and the
bureaucracy’s interests 10
Prejudices against a certain client, -
somewhereelse
m s m i t t e d by the colleagues at the Most clients come to the
against the
institution’swills
exercix control welfare office institution because they Contact with the
are estimated by the client mostly is very
Attempts ofother institutions (welbreoffice) to accomplice the as short in this
institution against the client’s interests non-trust-worthy institution

Fig. 3.

Together with some other concepts, these concepts show that three topics
play an important role in this counselor’s subjective theory on trust in relations
with clients:
‘90 Uwe Flick
- first, the conflicts between taking the viewpoint of the client and the
bureaucratic tasks and interests of the institution and
- second, the image of the ‘‘client in a bad situation” and

- third, the attempts ofother institutions to functionalize the institution (and

the counselor) for their interests and against the client.

These topics and the space they occupy in the subjective theory show the
influence of concepts from community psychology on the latter and its
orientation towards strengthening the viewpoint of the client against the
interests of the institutions he has passed during his career as a patient and
currently is confronted with. So analyzing the counselor’s subjective theory
discloses a certain professional attitude, a “version of the world” (Goodman
1978) of counseling and the role of trust in it.

When we have a look at what the conversation analysis of this (male)

counselor’s interaction with a (male) client discloses, we can take the following
excerpts as exemplary data-base:

B,: But, what’s the matter now in this moment? Uhh, Mrs X (official in the welfare
office) now has sent you to us (Ku:yes!) with what, with which consideration? Well,
what does Mrs X think, that you should d o here?
K,: Well, she knows that I am an epileptic, they know that and I a m an alcoholic, they
know that, only, that it is dangerous, when I drank nothing, yeaah? then I lie, uh,
then I am laid upon the street (B*:humm) and that’s the fear, I, I, I, that is the fear,
simply, that I pass out, (B2:Yeah, and-), some days ago again once more. (..)
B,: Back to your epileptic attacks. Are you interested in clearing that really up once?
And that you are correctly adjusted with medicine then? (K*:Yes) There is at the
University Hospital AAA, there a specialised service-
K,: Could you write that down for me?
B,: I can do that
K,: Yeah, that’s nice
B,: I’ll write it down for you, uhh, could you tell me once more, where you live?
K,: 45, E.-Street that’s in Berlin 67 (B2: 65?) 67, Berlin 67
B,: Yeah, but E-Street?
K,: Berlin 45, eheh, 45, E-Street, 45 E-Street, that is Berlin 67, well, in the district
B,: That’s really weird, then Mrs. U. would be responsible for you, well (..). I have to
check that out again before (..) hababab, E,- wait a moment, that was which
number, 45, or what?
K,: 45, E-Street, that is the district Berlin 67 (..)
B,: 45, E-Street, yeah, yeah, there Mrs U. responsible, I wouldn’t, I am not at all
responsible (..)
B,: You know, for today, I could have made it, but that will last longer, that will take a
while, I think, before it gets going. So, would you please come with me.

Ifwe have a brieflook at this excerpt ofcounseling, we can see the counselor’s
efforts to realize a helping and trustful relationship with the client according to
 Triangulation Revisited ‘9’
the concepts in his subjective theory. Thus, he offers the client to organize a
specialized diagnostic of his problems (alcohol, epilepsy) after clearing the
client’s own viewpoint, who was sent by some other institution (the welfare
office). I n his attempt to offer a helping relationship with some permanence and
stability, the counselor does not succeed. But this failure is not a personal failure
of the counselor, nor is it a result of the client’s specific participation in the
interaction. The counselor fails with his attempt because of the way
responsibilities in the institution are organized: responsibilities for cases are
distributed according to the clients’ addresses, and so the counselor finally has
to pass this client to his colleague, who is responsible for that part of the street
and of the district where the client lives. This shows exemplarily the difficulties of
counselors in this institution to put their good ideas into practice with real
clients. These difficulties lead to recurrent professional dilemmas in attempts to
realize -as in this example -subjective maximes or ideals and -as in the first
example -to apply professional routines as counseling in their usual way in this
specific context. So, to find the reasons for the divergence between the
counselors’ subjective viewpoints on trust in this context and the ways of
realizing them in concrete interactions, we have to focus on the institutional
context.

7.3 Institutional Backgrounds of Professional Dilemmas

The reasons for these dilemmas can be found in the double rationality of the
institution’s activities with clients: On one hand it shall try to help people in
trouble as far as possible, on the other hand it has to guarantee and maintain
public order by putting people in trouble under legal control and sending them
to psychiatric wards, in many cases forcibly. This institution is not only a
helping institution, but is also responsible for the administration of mental
disorder as a social problem and of the persons that are (often chronically)
mentally ill. During the long history ofthis type ofinstitution in Berlin since the
beginning of the century and the history of many of the single institutions still
existing today, the relation of these two types of professional activities has
changed: As activity reports and statistics show, the shift was from maintaining
order to helping people, but in the current official descriptions ofthe institution’s
tasks and aims these two types of activities still co-exist as well as in the
professionals’ everyday practice. While in the official version of the institution’s
public presentation helping people is described as its central task, in its
everyday practice it has to be cleared in each case in the direct contact with the
single client, if an urgent case is given and if this client still is able to lead an
independent life of his own or has to be put under control. This conflict is
intensified by the fact that most clients come to this institution because of
initiatives of a third person (like in our first example the grandfather) or an
other service (like in our second example the welfare office), who expect a certain
192 Uwe Flick
type of intervention by the institution for or against the client. These other
people or social services mostly are more interested in the institution’s
administrative activities than in its capability to help. In this situation the
counselor needs the client’s trust in different respects: to enter a relationship
with him that allows to help him but also to receive the information that allows
to decide whether to leave the client in his everyday life or to send him to the
hospital for a while.

8. MULTIPERSPECTIVE TRIANGULATION IN THIS STUDY

We find these dilemmas on all of the levels on which we studied the problem of
trust in counselor-client-relationships:
- On the level of the historical changes in the shift from force and control to

the current mixture and co-existence of help and control;

- on the level of the official and political descriptions of tasks and aims of the

institution in the demand to prevent forcing interventions where possible

and to exercise them where necessary;
- on the level of counseling interaction in negotiations with clients on their

case against the interests of other persons and institutions and in the
bureaucratic necessities and limits helping interventions are confronted
with;
- and on the level ofsubjective theories in which these conflicts and problems

are reflected in concepts of factors that hinder the development of trust in

relations with clients.
Figure 4 shows the different perspectives and levels we have taken to study the
phenomenon of trust between counselor and client in this institutional context:
The various methodical approaches applied on these levels and their different
theoretical backgrounds stress different aspects of the problem under study:
- reconstructing the subjective theories of counselors stresses the subjective
construction of the phenomenon trust and of the specific institutional
context. This stress on the subject’s viewpoint leads to the consequence of
applying the interview as a method, that is concentrated on subjective
experience and knowledge, and of applying the graphique representation
technique not only to visualize the structure of the subjective knowledge
but also to achieve a dialogue consense with the respondent on the contents
and structure ofhis or her version or construction ofthe phenomenon. This
version consists as well of former experience of own actions, participants
and contexts as of imaginations of ideal versions of these. But, what we get
are meanings of counseling and trust in this context for the subject;
- analyzing counseling conversation takes seriously into account, that
reality is constructed interactively and that it is a result of a process of
negotiation among the participants and with a specific context. It shows
 Triangulation Revisited ‘93

History and Changes of the Institution

f \
Counseling Interactions

subjective theory

of Counselor and Client

Fig. 4 .

everyday routines in constructing a specific practice and as well its specific

context. This approach transcends the subjective perspective towards a
social perspective. What we get here, are social meanings ofcounseling and
trust in this context constructed and negotiated in interactions. T h e
combination of these two perspectives allows to disclose in one direction
the subject’s reasons and motives for his or her way of participating or
intervening in a certain interaction o r context. I n the other direction it
allows to disclose how far the subjective viewpoint is to be seen as
idealization of given possibilities and circumstances. Divergent results
demand either theoretical explications or integrating other perspectives as
the following ones:
- analyzing official versions of institutional tasks, practices and results takes
into account, that in every institutional context two sorts of ‘realities’ are
constructed: an official one presented to the public, political authorities and
when ever needed in internal conflicts and an informal one which results
from routines, possibilities and everyday practices of the participants. This
shows the institutional ideologies, myths and expectations that are the
implicit background and burden of professional actions. It also shows the
space and limits for negotiating these two ‘realities’ in subjective
conceptions or ideologies and in everyday practices; on this level we get
institutional meanings of counseling and trust in this context;
- analyzing the history and changes ofthe institution takes into account that
‘94 Uwe Flick
every current situation and conflict participants are confronted with are
results of historical changes and developments. Here, we get the historical
meanings of counseling and trust in this context. These changes have an
influence on all the other layers and may offer reasons and explanations for
discrepancies between them.
So, to return to our starting points, the forms of triangulation proposed by
Denzin and discussed by others were applied as follows: We triangulated
different sorts of data (subjective theories, counseling interactions, current
official and historical documents), different methods (interviews, conversation
analysis, interpretation of documents and records) and different theoretical
backgrounds of these methods (subject-oriented or psychological theories,
ethnomethodology as a sociological theory, organisational theories and
historical approaches to institutional history). Investigator triangulation was
modified into the combination of approaches based on the intervention of
investigators in the field (by reconstructing subjective theories with interviews
and graphique representation techniques) with those that take activities
produced for. everyday purposes (like counseling interactions or institutional
presentation papers and records) as natural data and limit research activities to
the interpretation of natural data without interventions by the investigators in
the field.

9. PERSPECTIVES OF EVALUATION: NEW CRITERIA AND REFORMULATING OLD CRITERIA

Triangulation was discussed here as a “method-appropriate” strategy of

founding the credibility of qualitative analyses. Our research examples should
demonstrate exemplarily, how this form of mutual interpretive evaluation of
qualitative analyses works and how triangulation of these approaches opens a
way for mutual evaluation of our analyses, which adds breadth and depth to
them, without artificial objectivation of the subject under study.
Triangulating different perspectives in the way outlined here gives access to
different versions of the phenomenon that is studied: subjective, interactive,
institutional and historical versions can be gathered, interpreted, compared,
and complemented. So, triangulation as it is conceived and discussed here,
takes seriously into account, that there is no longer one reality against which
results can be verified or falsified, but that research is dealing with different
“versions of the world” (Goodman 1978). Triangulation takes into account,
that subjective knowledge and social interactions should be understood as parts
of (social, local, institutional) contexts and on the historical backgrounds of
those contexts.
Finally, it should be considered, that this article dealt with triangulation
exemplarily: Neither should triangulation be used as the only way to evaluate
qualitative analyses, nor is it by itself a sufficient way to answer this question en
 Triangulation Revisited '95
toto. But as we saw, it c a n be taken a n d developed as method-appropriate
criterium and as a n alternative to traditional criteria like reliability a n d
validity. O t h e r criteria should be taken, reformulated a n d used in similar ways.

D r U w e Flick
Technische Universitat Berlin
Institut fur Psychology
Dovestr. 1-5
1000 Berlin 10
Germany

NOTES

I Although Denzin himself (Denzin I 989a) now takes a somehow different position in
many points, his earlier propositions still are the basis for most of the recent discussions
on triangulation especially in Germany (Spohring I 989, Lamnek 1988). So, we will first
deal with this earlier position of Denzin and the critics related to it and later on have a
look at his reactions to these critics.

REFERENCES

ARBEITSGRUPPE BIELEFELDER SOZIOLOGEN (eds.) ( 1973). Alltagswissen, Interaktion und

gesellschaftliche Wirklichkeit. Reinbek: Rowohlt.
BERGMANN, J. (1985).Fliichtigkeit und methodische Fixierung sozialer Wirklichkeit.
Aufzeichnungen als Daten der interpretativen Soziologie. In BonR, W. and
Hartmann, H. (eds.), Entzauberte Wissenschaft, Gottingen: Schwartz, pp. 2gg-320.
BERGMANN, J. (1991).Konversationsanalyse. In Flick, U., Kardorff, E.V., Keupp, H.,
Rosenstief, L.v and Wolff, S. (eds.) Handbuch Qualitative Sozialforschung. Miinchen:
Psychologie Verlags Union, pp. 2 I 3-2 I 8.
BERGOLD,J. B. and FLICK,U. (1987). Die Sicht des Subjekts verstehen - eine
Standortbestimmung und Einleitung. In Bergold,J. B. and Flick, U. (eds.) Einsichten
-Zugange zur Sicht des Subjekts mittels qualitativer Forschung. Tiibingen: DGVT-Verlag,
pp. 1-17.
CAMPBELL, D. and FISKE,D. (1969).Convergant and discriminant Validation by the
Mutlitrait-Multimethod Matrix, Psychological Bulletin 56, 8 1-1 05.
DENZIN, N. (1978). The Research Act. Chicago: Aldine (originally 1970).
DENZIN, N. ( I 988). Interpretive Biography. London: Sage.
DENZIN, N. (198ga). The Research Act. (3rd edition) Englewood Cliffs: Prentice Hall.
DENZIN, N. ( I 989b). Interpretive interactionism. London: Sage.
FIELDING, N. G. and FIELDING, J. L. (1986). Linking Data. Beverley Hills: Sage.
FLICK, U. ( I 987). Methodenangemessene Giitekriterien in der qualitativ-
interpretativen Forschung. In Bergold, J. B. and Flick, U., (eds.), Einsichtm -
Zugange Zur Sicht des Subjekts mittels qualitativer Forschung. Tiibingen: DGVT-Verlag,
pp. 246-263.
FLICK,U. ( I 989). Vertrauen, Verwalten, Einweisen - Subjektive Vertrauensfheorien in
sozialpsychiatrischer Beratung. Wiesbaden: Deutscher Universitatsverlag.
FLICK,U. ( 1992). Entzauberung der Intuition - Systematische Perspektiven-
196 U w e Flick
Triangulation als Strategie der Geltungsbegrundung qualitativer Daten und
Interpretationen. In Hoffmeyer-Zlotnik,J. (ed.),Analyse qualitativer Daten, Opladen:
Westdeutscher Verlag (in press).
FLICK,U., KARDORFF E. V., KEUPP,H., ROSENSTIEL, L.v. and WOLFF,S. (Eds.)
( I gg I ). Handbuch Qualitative Sozialforschung. Miinchen. Psychologie Verlags Union.
GERDES,K. (ed.), Explorative Sozialforschung. Stuttgart: Enke, 1979.
GERHARDT, U. (1985).Erzahldaten und Hypothesenkonstruktion. Oberlegungen zum
Giiltigkeitsproblem in der biographischen Sozialforschung. Kblner Zeitschrifi fur
Soziologie und Sozialpsychologie, 37, 23e256.
GLASER, B. and STRAUSS, A. L. (1967).The discovery ofgrounded theory. Chicago: Aldine.
GOODMAN, N. (1978).Ways of Worldmaking. Sussex: Harvester Press.
GROEBEN, N. ( ~ g g o )Subjective
, Theories and the Explanation of Human Action. In
Semin, G. and Gergen, K. J. (Eds.),Everyday Understanding. London: Sage, pp. 19-44.
HABERMAS, J. ( I 967). Logik der Sozialwissenchaften, Frankfurt: Suhrkamp.
HAMMERSLEY, M. and ATKINSON,P. (1983). Ethnography - Principles in Practice.
London: Tavistock.
HOFFMANN-RIEM, C. ( 1980). Die Sozialforschungeiner interpretativen Soziologie.Der
Datengewinn. Kiilner Zeitschnzt fur Soziologie und Sozialpychologie, 32, 339-372.
HOPF, C. and WEINGARTEN, E. (eds.) (1979). Qulitative Sozialforschung. Stuttgart:
Klett.
JICK, T. ( 1983). Mixing Qualitative and Quantitative Methods: Triangulation in
Action: In Maanen, J. v. (ed.), Qualitative Methodologv. London: Sage: pp. 135-148.
JUTTEMANN, G. (ed.) (1985).Qualitative Forschung in der Pychologie. Weinheim: Beltz.
KIRK,J. L. and MILLER,M. (1986). Reliability and Validity in Qualilative Research.
Berverly Hills: Sage.
KLEINING, G. ( I 982) UmriR zu einer Methodologie der Qualitativen Sozialforschung.
Kolner Zeitschrift fur Soziologie und So.?jalpychologie, 34,224-253.
KOCKEIS-STANGL, E. ( I 982). Methoden der Sozialisationsforschung. In Hurrelmann,
K. and Ulich, D. (eds.) Handbuch der Sozialisationsforschung. Weinheim. Beltz. pp. 32 I -
37 0
KVALE, S. (1987).Validity in the qualitative research interview. Methods., I ( z ) , 37-72.
KVALE,S. ( 1 9 9 1 ) .Validierung - Von der Beobachtung zu Kommunikation und
Handeln. In Flick, U., Kardorff, E. v., Keupp, H., Rosenstiel, L. v. and Wolff, S.
(eds.) Handbuch Qualitative Sozialforschung. Munchen. Psychologie Verlags Union. pp.
427-432.
LAMNEK, S. ( 1988). Qualitative Sozialforschung, Vol. I. Miinchen: Psychologie Verlags
Union.
LAMNEK, S. ( I 989). Qialilative Sozialforschung, Vol. 11. Munchen: Psychologie Verlags
Union.
LUDERS,C. and REICHEKTZ, J. (1986). Wissenschaftliche Praxis ist, wenn alles
funktioniert und keiner weiR warum -Bemerkungen zur Entwicklung qualitativer
Sozialforschung. Sozialwissenschaftliche Lileraturrundschau, 12,go- I 02.
NOTHDURFT, W. (1987). Gesprachsanalyse subjektiver Konfliktorganisationen - Ein
naturliches Design zur Rekonstruktion individuellen Konfliktverstandnisses. In
Bergold, J. B. and Flick, U. (eds.) Einsichten - zugdnge zur Sicht des Subjekts mittels
qualitativer Forschung. Tubingen: DGVT-Verlag, pp. 98- I I 4.
OEVERMANN, U., ALLERT,T., KONAU,E. and KRAMBECK, J . (1979). Die Method-
ologie einer ‘objektiven Hermeneutik’ und ihre allgemeine forschungslogische
Bedeutung in den Sozialwissenschaften. I n Soeffner, H. G. (ed.), Interpretative
Verfahren in den Sorial-und Textwissenschaften. Stuttgart: Meltzer, 1979, pp. 352-434.
POLANYI, M. (1966). The tacit dimension. Garden City, New York: Doubleday and Co.
 T r i a n g u l a t i o n Revisited '97
SCHEELE, B. ( 1 9 9 1 )Dialogische
. Hermeneutik. In Flick, U., Kardorff, E. v., Keupp, H.,
Rosenstiel, L. v. and Wolff, S. (eds.) Handbuch Qualitative Sozialforschung. Miinchen.
Psychologie Verlags Union. pp. 274-278.
SCHEELE,B. and GROEBEN,N. ( I 988). Dialog-Konsens-Methoden zur Rekonstruktion
subjektiver Theorien. Tubingen: Francke.
SCHOTZ,A. ( 1 9 7 2 ) . Collected Papers Vol. 11. Den Haag: Nijhoff.
SCHOTZ,A. and LUCKMANN, T . (1979).Strukturen der Lebenswelt, Vol. I. Frankfurt:
Suhrkamp (originally 1975).
SCHUTZE, F. ( I 983).Biographieforschung und Narratives Interview. Neue Praxis, 283-
293.
SILVERMAN, D. ( I 985). Qualitative Methodolou and Sociology. Aldershot: Cower.
SMITH, H. W., (1975).StrategieJ for Social Research. Englewood Cliffs: Prentice Hall.
G , ( 1989). Qualitative Sozialforschung. Stuttgart: Teubner.
S P ~ H R I NW.
WAGNER,A. C., MAIER,S., UTTENDORFER-MAREK, I. and WEIDLE,R. (1981).
Unterrichtspsycho.~ramme.Hamburg: Rowohlt.
WEBB, E. CAMPBELI.,D., SCHWARTZ,R., SECHREST,L. (1966). Unobtrusive
Measures: Nonreactive Research in the Social Sciences. Chicago: Rand McNally.
WEINGARTEN, E. SACK,F. and SCHENKEIN J. (eds.) (1976).Ethnomethodologie. Bei1raget.u
einer Soziologie des Alltagshandelns. Frankfurt: Suhrkamp.
WILSON.T. (1981).Qualitative 'versus' Quantitative Methods in Social Research. I n
Kuchler, M. et al. (eds.), Integration uon qualitativen und quantitativen Forschungsansatzen.
Mannheim: ZUMA.
WOLFF,S. ( I 986). Das Gesprach als Handlungsinstrument. Konversationsanalytische
Aspekte sozialer Arbeit. Kolner ZeitschriJ fur Soziologie und Sozialpsychologie, 38, 55-84.