Previous Section: Blocks 1 & 2

Critical Interpretive Research: An Introduction

Blocks 3 and 4: Explanation Part I

The transition from the first pair of Blocks to the second and third pairs involved a shift in focus from description to explanation.

In both the empirical-analytic and interpretive paradigms explanation is addressed in two distinct ways:

In empirical-analytic work, explanation is a matter of establishing a relationship of cause and effect, by determining whether a significant difference in an effect or outcome-variable can be attributed to a causal variable. To do this, two different strategies are employed to explore such causal relationships between measured variables: the first is experimental design (Block 3), while the second is quasi-experimental design (Block 5). In both cases the goal is one of control, but these two kinds of design differ in the kind of control by means of which the existence of a causal relationship is established: by direct manipulation of the causal variable, or by statistical control of the differences between groups.

In interpretive research, explanation is also sought in two different ways: first, in terms of people's practical accomplishments in their conversation and interaction (Block 4), and second, in terms of the constitutive causality of a social field (Block 6). The first approach to explanation emphasizes that social interaction is a skilled production, with a complex organization that typically goes unnoticed. Explanation of an educational phenomenon, for example, can take the form of an articulation of the work that has produced it. The second approach to explanation in interpretive inquiry emphasizes that people are located in intersubjective social fields that both facilitate and constrain their activity. It follows that explanation can take the form of explicating the constitutive causality of these fields. These two approaches to interpretive explanation complement one another, rather than being alternatives.

Block 3: Empirical-Analytic Explanation -- Direct Manipulation (4 weeks).

In Block 3, the first of these two Blocks to deal with explanation, experimental design was introduced in the context of the questions about causation considered appropriate in the empirical analytic paradigm, and the kinds of answer considered acceptable: establishing cause and effect relationships between treatment and outcome variables. The use of random assignment to control initial (pre-treatment) differences between groups provided an introduction to the use of inferential statistical techniques to determine whether the differences observed between groups are larger than expected by chance alone, and whether it is appropriate to generalize from a sample of data to the larger population. The curriculum evaluation case materials examined in Block 1 were re-examined, investigating the effect of the group discussion program on students' performance on the interpretive writing task evaluated in Block 1. The Block Project was to evaluate a component of the curriculum reform -- a group discussion program -- considered as a treatment.

Block 4: Interpretive Explanation -- Practical Accomplishments (5 weeks).

It is sometimes asserted that qualitative research in general, and interpretive research in particular, can provide descriptions but not explanations. But this is not so. Interpretive research can provide explanation, in fact there are two different ways in which explanation can be provided: in terms of the work people do to produce a social phenomenon, and in terms of the social context in which this work is done. Block 4 explored both these approaches to explanation.

Detailed Syllabus for Block 4

Handout 4: General Introduction to Block 4

From discourse to interaction;

From what's said, to what's done.

Block 4 began an examination of people's practical accomplishments in their conversation and interaction. This required a two-fold shift in attention: first, whereas in Block 2 we were considering the structure of one person's discourse, now we must consider the organization of interaction among speakers. At the same time, whereas in Block 2 we were considering primarily how people talk (the narrative and argument structure of discourse), now we must consider what people do by talking.

Philosophers, linguists and social scientists now recognize that talk doesn't just convey information (in its "propositional content"), talk also accomplishes something socially. It does this by virtue of being composed of "speech acts" (Austin, 1975; Searle, 1969). Block 4 introduced the notion that social phenomena (for example, a therapy session, a classroom mathematics lesson) are the practical accomplishments of people in social interaction. Everyday conversation is a skilled production, and a means by which people collaborate to accomplish various social outcomes. Consequently, an explication of the social interaction by means of which a social event is accomplished can provide an explanation of that event.

Study of the organization of social interaction requires observing both the routine character and the unfolding organization of the events of everyday social interaction. Skills of observation and note-taking are important. But there are limits to analysis of "live" action; the interpretive researcher must work with action that has been recorded, in audio- and video-recordings, and transcripts. I've already noted that a transcript is never a perfect verbatim records. This is one consequence of the way that "fixing" action transforms its character in a variety of important ways (cf. Ricoeur, ).

The interpretive researcher must also make some important decisions about the format and style of written reports. These were also introduced in this Block.

Readings for this Block: Mehan, 1979; Cazden, 1986; Evertson & Green, 1986.)

Week 1 introduced the primary analytic technique: conversation analysis (Nofzinger, 1991; Levinson, 1983). Students watched a short video-recording of one of the group discussions that had been considered in Block 3. They worked in groups to write an account of this interaction. When doing this they were asked to follow a number of injunctions:

  1. to ground their interpretations in evidence, and to follow the ethnomethodologists' aim to use chiefly "internal criteria," making judgments that participants themselves gave evidence of making. For example, they considered evidence that the participants understood the end of the taped segment as a transition (utterances like "we've exhausted that topic;" the teacher providing a "wrap up;" a long silence or gap; the utterance "shall I go on?;" and finally a change in topic).
  2. Hugh Mehan's recommendation "to limit the use of dispositional properties and background factors [such as socioeconomic status or gender] as explanatory devices when their influence cannot be located in the interaction" (Mehan, 1979, p. 121) and instead to seek explanation in pragmatic features of the discourse such as the organization of turn taking;
  3. Habermas' methodological injunction, to presume that the participants were acting rationally even when their interaction may at first glance appear senseless.

The pedagogical aim here was to sharpen the students' perception of everyday interaction, and to demonstrate that with repeated viewing their understanding could and would change, often significantly. (Readings: Ball, 1990; Jacob, 1987; Wilson, 1977; Handout 5.)

Handout 5: Conversation Analysis

Conversation analysis Conversation analysis ("CA") had its origins in ethnomethodology, a social phenomenology itself influenced by Alfred Schutz (Nofzinger, 1991; Levinson, 1983).

Its central assumption is that conversation is interactionally and locally managed (that is to say, worked out on the spot, and moment by moment, with simple practical procedures and devices rather than being a product of complex generative rules).

Its goal is that of "trying to model... the procedures and expectations actually employed by participants in producing and understanding conversation" (Levinson, 1983, p. 319).

Conversation analysis presumes that the participants in a conversation are playing a "language game," in which they take turns and make moves. We can quickly see some of the "devices" people use to work out the turns of the game, allocating turns and selecting speaker with devices like the adjacency pair: linked speech acts such as question-answer, greeting-farewell, and invitation-acceptance that a speaker uses to establish expectations about who will speak next and what they will do. Other recognizable kinds of organization to everyday conversation include presequences and insertion sequences; different kinds of silence (gap, lapse and pause); and preferred response.

Various analytic strategies can be used, including the way that each conversational turn can provide an interpretation of the preceding turn, and attention to hitches and repairs in the conversation.

The Block Project was to observe and record a conversation and employ conversation analysis, moving beyond the identification of "turns" and "moves" to an interpretation of the character of the "game" the interactants were engaged in. (Reading: Nofsinger, 1991a, b; Handout 5.)

The labs for Block 4 included workshops on equipment for audio- and video-recording, and a survey of computer software for qualitative analysis: Atlas-ti; HyperQual, Ethnograph, and others.

Week 3. The notion that conversation consists of "turns," and that when participants in a conversation take a turn they make a "move," rests on the metaphor that conversation involves "playing a game." This metaphor was explored further in Week 3.

The game metaphor Wittgenstein (1953) suggested that different "forms of life" amount to diverse, polymorphous "language-games." Peter Winch (1958) built on this idea to outline a logic for social science. This marked the "linguistic turn" to the social sciences (cf. Bernstein, 1983; Giddens, 1977; McCarthy, 1978, pp. 162ff).

A game involves rules that assign turns and define valid moves, define states of initiation and termination, allocate resources and apply sanctions. A language-game also involves rule-following, in social conventions of exchange of expressions and actions, although typically the rules are not explicitly codified.

A language-game can also be said to articulate a world, and to embody a form of life. The exchanges people engage in when they interact together (the sequence of "turns"; the "moves" they make) show the "game" they are playing, and the way they are living.

The video was viewed again with an eye to moves as well as turns. The "QRE" (question, response, evaluation) structure of much classroom discourse was introduced. (Readings: Mehan, 1979a; Hammersley, 1990; Wittgenstein, 1953; MacIntyre, 1984; Bennis, 1989, Goetz & LeCompte, 1984; Everhart, 1977.)

Week 4 examined on the relationship between description and explanation.

What does it mean to say a researcher "understands" what people are doing? Clifford Geertz (1976/1979) has argued that understanding doesn't call for empathy, some special sensitivity and attunement to other people (a favorite myth of anthropologists). Rather, it calls for close attention to the symbolic forms that people use.

This involves "a continuous dialectical tacking between... local detail and... global structure," and a movement back and forth "between experience-near and experience-distant concepts."

Hermeneutics enters the picture As we've seen, a central assumption of the interpretive paradigm is that people actively organize their world, as they act and interact with others. It follows that, if we are to be consistent, we must acknowledge that the researcher, too, is actively working to make sense of phenomena, not simply recording objective facts.

In order to understand the researcher's role we turn to hermeneutics.

Hermeneutics -- the interpretation of texts and text-analogs -- insists that there is no description free from interpretation, shaped by presuppositions.

: an active interpretation of phenomena guided by tacit assumptions and interests.

The movements Gerrtz describes form a hermeneutic circle.

in the interpretive paradigm explanation begins with a "hermeneutics of everydayness": an articulation of people's talk and action, and the way they grasp and understand these symbolic forms. (Readings: Willis, 1977; Geertz, 1979; Farrell et al., 1988.)

Week 5 continued with theoretical matters while students were busy completing their Block Project. Interpretation can be seen in ontological terms: to be human is to understand and interpret. Empirical-analytic research, and some qualitative approaches to inquiry, treat knowing as the product of detached observation. The interpretive paradigm in contrast sees knowledge as grounded in practical activity in social settings. Artifacts (including utterances, which are short-lived artifacts, crafted in the moment) mediate our contact with, and understanding of, the world around us. A piece of chalk, for example, is grasped as a tool to write with in order to make a point when teaching. The chalk is grasped in the context of the "world of the classroom," a whole network of people and equipment, and this grasp reveals that world to us.

An "ontological blueprint" can be sketched: people are embodied agents located in a shared social field that provides a context for activity, and a background upon which artifacts are grasped and "projected" (Heidegger, 1927/1962): the background defines the kind of being that artifacts and people have. A person's position in the field provides a particular perspective and grasp. This ontological blueprint summarizes key assumptions of the interpretive paradigm: the active agency of those studied; the concerned involvement of the researcher; the constitutive and constituted character of social reality. An interpretive research report should include relevant aspects of the blueprint: the field; the positions of the participants; their tasks and projects; their perspectives or points of view; the kinds of exchange taking place; the position of the researcher, etc.. (Reading: Giddens, 1983;  Handout 6.)

We said at the very beginning of this class that the empirical-analytic and interpretive paradigms don't simply differ in the techniques they employ -- rating scales versus interviews, for example -- they involve different ways of understanding the world: different ontologies. (Ontology - study of the nature and kinds of being: what exists.)

Next Section: Blocks 5 & 6: Explanation Part II

© Martin Packer, 1999

Home Page

Developmental Psychology

Cultural Psychology