media, culture and meaning > research > qualitative methodologies resources > constructivist methods

Constructivist Methods in the Symbolism, Media and the Lifecourse and the Symbolism, Meaning and the New Media @ Home Projects

by Lynn Schofield Clark, Ph.D.

It is our belief that there is no single methodology that is superior to any other methodology in every case. Different research questions lend themselves to different methodologies. We find that our current projects borrow most from constructivist and critical methods. In this document, we explain these methods and their relation to our research questions and to research directions in general.

Introduction: Quantitative research as the foundation for comparison

The first distinction that needs to be made is between qualitative and quantitative research. Constructivist approaches are a recent development under the umbrella of qualitative research, which itself has been defined by its contrasts from the quantitative research paradigm. Because the assumptions of quantitative research still define how we think of research in the contemporary U.S., it is worth describing this paradigm further.

Quantitative research is the term used to refer to research that can be rather easily translated into the examination of functional or cause/effect relationships. Physics, biology, and chemistry are models for this kind of research, known as logical positivist in approach. In the social sciences, we have borrowed from these models, employing first positivism (which focuses on the efforts to prove that something is true) and then, after the influence of Karl Popper, post-positivism (which attempts to falsify existing theories).

Positivism is the belief that all knowledge comes from experience with a knowable, unchangeable reality. Resting on this assumption of a "concrete" reality, positivism looks to empiricism - the science of gaining knowledge through observation -- for its method of inquiry. Because reality is believed to be knowable, the goal of the scientist is to achieve separation from that which is studied so as to be "objective," or to not introduce "bias," into one's research.

Surveys and experimental designs have been the primary methods used in positivist approaches, with statistical modeling employed so as to achieve generalization across populations. While social science has assumed that research into human beings is challenging because no two humans are the same, it has also assumed that humans are similar and therefore generalities can be made about the human experience. Research into human behavior has thus necessitated a great number of subjects (hence the designation quantitative) to verify findings that are consistent and generalizable, or "true," across the population.

Quantitative research has many strengths in studies of mass media, particularly when looking at questions of aggregate groups. For instance, quantitative research can compare the number of hours adults watch television to the number of hours children devote to television. These methods can also help to determine the demographic representation of the audience of particular programs or genres on television or radio, or of users of the Internet. These methods are also helpful in content analysis, which can demonstrate the over- or under-representation of certain groups or concerns in electronic and print media.

Weaknesses of quantitative research, strengths of qualitative research

We believe that there are several weaknesses of quantative research. First is the problem of meaning and operationalization. Even with well-constructed surveys, it is impossible to know whether or not a respondent understands a certain word, phrase, or value-laden statement in the same way the researcher does. Qualitative research, with its interview style, allows researchers to investigate meanings made by specific audiences, and thus is able to address this issue to some extent.

A related problem is the issue of social desirability. People know that it is desirable to do certain things and not others (using the Internet for research and not for downloading pornography, for instance) and they may therefore tend to over-report the desirable behaviors and under-report those deemed undesirable. This is a factor of research with people regardless of the methodology. In qualitative research approaches, however, this issue is minimized as researchers are able to explore seeming contradictions as well as reasons why persons might report the way they do.

While quantitative research predominated in studies of the mass media for most of the twentieth century, an interest in qualitative research began to emerge in the latter decades. Still, the concerns of quantitative research - in particular generalizability, validity, and reliability - continue to dominate discussions of all research and thus will also be discussed below.

Qualitative research

Although contemporary qualitative research in media studies has its predecessors in the cultural anthropology and sociology of the 19th century, this style of inquiry experienced a comeback in the 1960s and 1970s, as scholars began to realize that all knowledge-production is related to the assumptions investigators bring to their study. Reality, scholars began to argue, was knowable only imperfectly. Researchers in mass media began to recognize that survey research was stripped of context. Thus, qualitative researchers sought to get as close as possible to the meanings ascribed to various words and actions by members of the groups they wanted to study.

The purpose of inquiry in qualitative research, therefore, is in understanding the world from the point of view of those who live in it. While the natural sciences have as their goal scientific explanation, the goal of qualitative research is the grasping of understanding, or the "meaning" (Verstehen) of social phenomena. We seek not just to observe and describe, but to offer, in anthropologist Clifford Geertz's (1973) term, a "thick description" of how people as actors understand and ascribe meaning to their own actions.

Education scholar Thomas Schwandt (1994) has argued that there have been three different approaches, or methods, of studying the phenomena of meaning:

1. The scientific examination of meaning. Scholars in this approach attempted to study meaning in ways that borrow from quantitative research, often seeking causal explanations or striving to separate facts from values (e.g., Wilhel Dilthey, Max Weber, Alfred Schutz).

By early in the twentieth century, however, such approaches had largely been abandoned, and the next approach to the study of meaning had taken center stage.

2. Scholars attempt to synthesize realism and constructivism, focusing on error-elimination strategies. Scholars in this tradition advocate the judicious use of method to avoid the errors that may result from what they believe are overly subjective interpretations. Scholars operating in this approach attempt to conduct research that is free of bias, is able to suggest generalizable findings due to the typicalty of their respondents, and is reproducible by other investigators, among other things. Researchers assume that there is no single reality, and that reality is only partially apprehendable, an assumption that separates them from those who would scientifically investigate the (presumably single) meaning of a phenomena. To assert their study's credibility, scholars in this tradition rely upon scientific claims relating to the study's "rigor"(defined in terms of how well it approximates standards of logical positivist research, such as generalizability, reliability, unbiased exploration, etc.)

Following Guba and Lincoln (1994), we use the term post-positivism to refer to research in our field that grew out of these impulses. While employing interviewing rather than large scale surveys or experiments in its attempt to get closer to the subject's "reality," we argue that this type of qualitative research still tends to import most of its assumptions from the model of research found in the hard sciences, and thus encounters difficulties in its claims for credibility. Researchers are admonished to be "as objective as possible," for instance, and usually write in a language of scientific neutrality. As another example, reports in this tradition sometimes use statistics as a means of strengthening a study's validity. But given the fact that sample sizes are too small to be considered meaningful representations of some larger group (not to mention the fact that they are usually not systematically gathered representative samples), these statistics are often meaningless or misleading. Research in this tradition is often challenged for its lack of generalizability, its difficulty in establishing reliability, the difficulty of establishing a research participant's typicality, and sometimes even for the "descriptive" rather than theoretical tone of its conclusions.

We feel it is more appropriate to meet the questions emerging from quantitative research paradigms by beginning with a reminder, therefore, of how qualitative research draws upon humanistic and critical approaches to research in addition to social scientific traditions; interpretation is not so much a science as an art, as Geertz (1980) has argued. The qualitative approaches to research differ in their understandings of the status of knowledge, and thus not only take different paths in research design and conduct but also in data analysis, interpretation, and arguments made from the research. Thus, our research begins with what Schwandt identifies as the third approach to the study of meaning.

3. A full acceptance of the hermeneutical character of existence. To those in this tradition, including ourselves, interpretation is not merely a methodology, but a fundamental aspect of any and all research endeavors. As cultural anthropologists Paul Rabinow and William Sullivan (1987) write: "The interpretive turn is not simply a new methodology, but rather a challenge to the very idea that inquiry into the social world and the value of understanding that results is to be determined by methodology." ( p. 20). Both language and history condition and limit interpretations. Thus, researchers must engage in reflexivity, analyzing their own role in the specific knowledge that is created in any research situation. In this approach, meanings are intersubjectively shared. The world is understood through "social artifacts, products of historically situated interchanges among people" (Gergen, 1985).

Constructivism

We borrow the term constructivism from Guba and Lincoln (1994), who likewise identify their qualitative research work as constructivist, The term references the acknowledgement of the social construction of knowledge. As constructivists, we begin with a great deal of skepticism that bias could truly be eliminated from scientific inquiry. Constructivist researchers, with our roots in symbolic interactionism, are more interested in the co-construction of knowledge between researcher and researched and thus discuss bias in relation to the situatedness of all interviewer/interviewee situations.

We advocate that knowledge and truth are the result of perspective. Research participants may be more or less reflexive regarding their own perspectives, and more or less articulate in expressing them. But we affirm that the knowledge that emerges from interviews with research participants is at least in part created, not discovered, by the researcher. Knowledge and interpretation in a constructivist research paradigm is thus the result of a collective, not an individual, process. Therefore, we must attend to three things when writing up our research:

1. the assumptions we as researchers bring to our subject of inquiry, and to the research situation;

2. the socially constructed meanings that occur in the context of a particular interview;

3.the socially constructed meanings that existed prior to, and shape or limit, the meanings that may emerge in a specific interview context.

Constructivists assume that there are many possible interpretations of the same data, all of which are potentially meaningful. Constructions are therefore not separate from those who make the constructions; they "are not part of some 'objective' world that exists apart from their constructors," as Guba and Lincoln (1989) write (p. 143). Guba and Lincoln further argue that a "malconstruction" would be an analysis that is "incomplete, simplistic, uninformed, internally inconsistent, or derived by an inadequate methodology." (p. 143).

There are two readily identifiable weaknesses of the constructivist paradigm. First, in our emphasis on meaning -- particularly at the individual and local level -- there is a tendency to downplay power relations that privilege certain constructions over others. Constructivists are therefore often rightly accused of being idealists with little to say about the material world. Our own position is that while we are abandoning the positivist's notion of a single reality, we do not abandon the position that certain actions are quite real in their consequences. Thus, while our analyses focus on the meanings ascribed to actions by people, we aim to place these meanings in a wider framework that analyzes these meanings in light of learnings from critical social theory (for more on this, see Critical theory and constructivist methods, Clark).

Second, constructivists frequently encounter the charge of relativism: if all meanings are co-created, how can one meaning -- the researcher's -- be any more important than any other? We recognize that constructivism, along with most of the writings of qualitative research, confront this problem as a result of the challenges to objectivity that are inherent in these approaches. We argue that we can still create worthwhile interpretations of social life without needing to claim that our understanding is either complete or final. As our theories are tested in the context of other experts, as well as within the lived situations of those we purport to study, these theories can gain some authority -- until, of course, they are challenged by further interpretations. This relates, then, to the goal of the research, which is to offer an interpretation that challenges, provokes, or encourages further questions rather than one that provides definitive explanations.

A central research question: Accounting for the media

Our interest in constructivist methods is rather directly related to the emergence of what we now identify as a central research question for the project that has been written into the manuscript titled, Accounts of the Media: Reflexive Parenting in Late Modern Life. The central question of that work is: what is the role of the media in the construction of a family's identity?

When we began our research, we expected the investigation of this question to be rather straightforward (as described elsewhere, we began with a research agenda shaped by post-positivism; see Clark, Learning from the field: The journey from post-positivist to constructivist methods). We thought we would ask people about their family media practices and beliefs and then would categorize their responses, perhaps on a continuum from heavy to light users or what we called "media-suffused" to "media-distinctive" families. Rather than a clear pattern emerging along this continuum, however, an unanticipated theme became prominent and seemed to be a part of all of the interviews. Almost all of the families related their own practices and beliefs to what they thought families should be doing with the media. Along the lines of previous research that has suggested that certain media practices are more socially desirable than others, we found families complaining of television violence, families worrying about overt sexuality in cable television and films, families concerned about the amount of time other peoples' children spent watching television or playing video games, etc. Sometimes, these complaints were blatantly inconsistent with family media practices. Nevertheless, we gradually came to see that it was impossible to separate a family's identity statements from "public scripts" of the media, or what media theorist Ellen Seiter (1999) has called the "lay theories of media effects" of audiences. While families seemed to be eminently familiar with these "scripts", we came to suspect that they were playing different roles in relation to the meaning-and identity-making strategies of various families. When families claimed these scripts or theories as their own, therefore, we call these the family's account of the media. We use the term account in the sense that it is meant by symbolic interactionists, such as the description offered by Gergen: "Accounts of the world...take place within shared systems of intelligibility - usually a spoken or written language. These accounts are not viewed as the external expression of the speaker's internal processes (Such as cognition, intention), but as an expression of relationships among persons" (p. 78).

We also find the term accounts useful for its reference, as Charles Taylor (1989) has noted, to accountability. These accounts are not merely narratives that draw upon intersubjective meanings and co-constructed in the presence of a researcher, although this is an important element of the term. We believe families (parents in particular, of course) offer us these narratives with some reference to their own sense of accountability as parents. In other words, they see themselves as the persons primarily responsible for understanding and counteracting the effects that are at the root of the "lay theories of media effects" (television causes violence, video game-playing is a waste of time, etc.). In most cases, parents want to present themselves to an outsider, or even to see themselves, as responsible and reasoned. Thus, the narratives they offer of their family's media behavior is inflected with the sense of what they feel they should be doing, or what they wish we and others assume that they are doing, to counteract the negative effects of the media. In the very process of conducting an interview, we are inadvertently perceived as holding them accountable for the oversight of their childrens' media practices and beliefs.

The previous paragraphs demonstrate the interrelation between knowledge-construction, research design, and research findings in a constructivist approach to research. We did not set out to explore the "accounts of the media" as a category, much less a hypothesis. Given our relatively small sample of 269 persons, we are not able to generalize that all or most people tend to give a certain kind of account of the media. However, our research offers an interpretation of how people describe their family's relation to media and why they might construct a narrative as they do, and why it is important to understand any kind of statement of identity or media use within a cultural context that suggests, first, that some media practices are more desirable than others, and second, that parents are ultimately responsible for mitigating the perceived negative influences of the media.

We find support for our analysis of the accounts of the media in family identity in similar studies, such as that of Seiter's (1999). Borrowing from the post-positivist tradition, as our research continues we are always on the lookout for cases that will challenge our emerging interpretations. Yet we believe that our research moves beyond the desire to analyze and categorize, challenging the very notion of how we as researchers participate in knowledge-construction about media practices.

Constructivist responses to commonly-asked questions:

Sample representativeness:

In our study, we embrace what Lindlof (1994) terms "maximum variation sampling." Thus, our sample roughly mirrors the demographics of U.S. society. It is important to note that we encorage this not to approach individuals in the sample as "represenrtative" of some particular demographic variable, however. In our understanding of co-created knowledge, we see a benefit in a diverse sample that may enable us to become aware of our own blind spots and limitations. In this way, we hope to confront our own latent assumptions that may have otherwise guided our interpretations. Again, we are seeking to say something about the possibilities and processes of a culture, not about individual (or group-specific) motivations and behaviors. We feel that research that inquires into race, gender, class, place, or other identifiers is important, and we rely upon existing research and expertise in this area to inform our own understandings of our particular research questions.

Data collection and analysis

Constructivist research recognizes that data-collection is a discovery process. While positivist and post-positivist research tends to focus on verification or falsification of hypotheses, there is little attention to what latent theories guided one's hypothesis-construction to begin with. Constructivists recognize that their hypotheses may change as their study evolves. Through their interactions with people, they may come to learn that their original latent hypothesis was too narrow, too broad, or simply inconsistent with the ways in which people actually experience themselves and their practices. Rather than relying upon a one-time survey or experiment with a representative sample, constructivist researchers have the benefit of returning to their site of study several times, adjusting their interview instrument as they learn and develop their own theories. In simple terms, while quantitative researchers begin with a hypothesis, qualitative researchers (including constructionists) are more likely to end their study with a working hypothesis. This is perhaps why some consider qualitative methodologies to be ideal "pilot studies" that may be further fleshed out in quantitative research. Many qualitative researchers would argue, however, that the depth of context and contradiction that can be discovered in a qualitative study simply cannot be duplicated in a larger, survey-based design.

It is our position that there may be some elements of our studies that will lend themselves to broader, more generalizable study in the future. We are still developing a theory that might be tested in such a design, however, and we believe that this theory development is an important and provocative effort of its own merit.

Reliability:

In survey research, reliability refers to the notion that one question should have the same meaning to different respondents. With our emphasis on the dialogical, constructivism assumes that no two interview encounters will ever be the same. Even if the same interviewer were to ask the same questions of the same respondent, on another day the answers might be slightly different depending on contextual factors such as contact with other people or experiences that shape an interviewee's views. This is not to say that there is no baseline for examining whether or not a statement is a reliable representation of a person's views or a trustworthy account of their experiences; it is simply to recognize that all such self-reports must be understood as constructed within a specific context and for a particular audience Such an approach makes it more difficult to establish reliability and contributes to the challenges of interpretation that are central to the qualitative approach. It also places much more emphasis on a study's validity.

Validity:

In constructivist approaches, the validity of a study is not determined with reference to scientific methods or a study's replicability, but on how a given interpretation may be judged. Is it thorough, coherent, comprehensive? Does it make sense, or ring true? Is it useful? In particular, is the interpretation provocative and generative of further inquiry? If a study meets these criteria, it may be said to be valid. However, as is true in post-positivist research, theories are always posited as partial and open to further elaboration or even discounting. A valid qualitative study is one that takes into consideration the context of those who are the subject of inquiry and offers a promising analysis of why an event occurs or how events, symbols, and narratives are made meaningful for people.

Generalizability:

Due to the small populations studied, constructivists do not generate findings that are generalizable to behaviors, psychological motivations, or even meanings made by other people. What is generalizable is the fact that people rely upon common symbols to communicate. As a people, we share intersubjective, common meanings. The goal of this type of research, therefore, is to seek out and interpret these common meanings people hold. We see these intersubjective meanings as "constitutive of social life." (Schwandt, p. 226). Thus, by offering an interpretation of how people make things mean, we offer a generalization about the possibilities and processes within a culture. We do not advocate that ours are the only interpretations, but they are among the possible ones, and they are given credibility by their resonance among others who research such matters.

Objectivity/lack of bias:

Constructionists see method differently than logical empirical scientists. While scientists attempt to limit or eliminate personal, subjective judgment, constructionists see it as an important aid in good judgment and understanding. The researcher is the research instrument, and thus the goal is not to remove the researcher's perspective, but to hone it so that the researcher is as equipped as possible to make a sophisticated analysis and argument about the phenomena observed. As cultural anthropologists Rabinow and Sullivan argue of this approach, "both the object of investigation - the web of language, symbol, and institutions that constitutes signification - and the tools by which investigation is carried out share inescapably the same pervasive context that is the human world." (p. 6).

Conclusion:

Clearly, constructivist approaches are influenced by postmodern theories of the self/other relationship. Our methodology is also shaped by critiques of ethnographic writing and interpretation as well as critiques of more "scientifically" oriented research methods that have been proffered by feminists, critical theorists, and scholars from parts of the world that are outside the U.S. and Europe (where many of the scientific paradigms originated). As such, the methodological approach attempts to respond to current theories of knowledge-gathering and writing, but is open to further criticism and refinement. Nevertheless, we believe that it is an important starting-point, and we invite you to read our initial analyses of our research efforts so that you might add your own critique to the postmodern chorus that contributes to our understandings.

Sources:

Clark, L.S. (1999). Learning from the field: The journey from post-positivist to constructivist methods. Paper presented to the International Communication Association, San Francisco.

Geertz, C. (1973). The interpretation of cultures. New York: Basic books.

Geertz, C. (1980). Blurred genres: The refiguration of social thought. American Scholar, 49, 165-179.

Gergen, K. (1991). The saturated self: Dilemmas of identity in contemporary life. New York: Basic books.

Gergen, K. (1985). The social constructionist movement in modern psychology. American Psychologist 40, 266-275.

Guba, E.G. & Lincoln, Y.S. (1994). Competing paradigms in qualitative research. In N.K. Denzin &Y.S. Lincoln (Eds.), The landscape of qualitative research: Theories and issues. Thousand Oaks, CA: Sage, 195-220.

Guba, E.G. & Lincoln, Y.S. (1989). Fourth generation evaluation. Newbury Park, CA: Sage.

Rabinow, P. & Sullivan, W. (1987). The interpretive turn: A second look. In P. Rabinow & W. Sullivan (Eds.), Interpretive social science: A second look. Berkeley: University of California Press, 1-30.

Schwandt, T. (1994). Constructivist, interpretivist approaches to human inquiry. In N.K. Denzin &Y.S. Lincoln (Eds.), The landscape of qualitative research: Theories and issues. Thousand Oaks, CA: Sage, 221-259.

Seiter, E. (1999). Television and new media audiences. Oxford: Clarendon Press.

Top of this page

Back to main qualitative methodologies resources page