integ6

AN INTEGRATED APPROACH TO
SCIENTIFIC RESEARCH METHODS:
Psychological Research on the Human Spirit

PART FIVE

John Davis, Ph.D

Metropolitan State College of Denver
Department of Psychology

Click here to go to send email to John Davis.

Click here to go to CONTENTS of this article.

The Question of Scientific Adequacy

The question of adequacy of scientific research has been a key issue for human scientists. Natural scientists argue that research methods must be rigorous and that the way to achieve rigor is to quantify, operationalize, manipulate, and control the phenomenon under study. Since much human science research does not quantify, manipulate, or control, it has been deemed inadequate by these experimentally-oriented psychologists. However, both natural science, using a positivist model, and human science, with a phenomenological orientation, share a concern for adequacy.

It can be seen that positivism's broad evaluative criteria are reasonable standards for existential-phenomenological [i.e., human science] research. That is, research conclusions should be empirically based; research should strive to be free of personal biases, prejudices, and dogma; other individuals should be able to agree that conclusions are justified by the data; and criteria should be provided for evaluating competing knowledge claims. . . . Although the ontological and methodological assumptions of existential phenomenology and logical positivism differ, both share a common commitment to conducting rigorous, empirical research that is open to careful scrutiny. (Thompson, Locander, & Pollio, 1989, p. 142)

While human science researchers often use the term rigor to describe their research methods, adequacy may be a better term. The notion of adequacy suggests research methods that are equal to the task of understanding, are responsive to the nature of the subject matter, and are open to critical evaluation. Although the underlying standards of good research are the same, the particular criteria for scientific adequacy differ in natural science and human science (Kirk & Miller, 1986). Lincoln and Guba (1985, 1994) identify the primary aspects of rigor or adequacy in scientific research and detail how these aspects are expressed differently in the two approaches. Three of these aspects of adequate scientific research (accuracy, consistency, and neutrality), along with their parallel versions in natural science and human science, will be examined.

Accuracy

The foundation of the adequacy of scientific research is its accuracy or, to use Lincoln and Guba's term, its truth-value. In this sense, truth-value can be understood as the extent to which the research findings are true to the phenomenon under study.

In natural science, where the goal is to determine cause-and-effect relationships, accuracy depends on internal validity. Internal validity is a means for establishing that the independent variable is the major cause of changes in the dependent variable. This is accomplished by controlling extraneous variables in order to eliminate alternative hypotheses such as subject selection, history, and demand characteristics. Control and systematic manipulation of conditions are central requirements for establishing internal validity. However, achieving such control and manipulation cannot be done with many kinds of psychological phenomena, including experiences of the human spirit.

An alternative in human science aims for constructing credible accounts of a phenomenon in its context, rather than control. This is based on describing the deep structure or pattern of an experience or phenomenon in a way that is faithful to the experience. According to this standard, credibility is achieved when someone who has had the experience can say, "This captures my experience better than I could have." By this standard, the accuracy of accounts of the human spirit can be evaluated and established.

Consistency

Consistency is another critical aspect of scientific adequacy. A particular scientific method must be consistent from one study to another and from one setting to another. Inconsistent methods are akin to using a rubber ruler to measure. In natural science, reliability of both tests and experimental settings is one of the first requirements for rigorous research. However, in the view of human science, many phenomena, by their very nature, cannot be replicated. This makes it difficult or impossible to establish reliability for most experiences of the human spirit. An alternative to reliability is auditability. To achieve auditability, the research provides enough details of the research setting, the informants, the data collection methods, and the data analysis to allow other researchers to come to their own conclusions, i.e., to audit the original research process. In good human science, auditability allows other researchers to follow and evaluate a research study. These auditors may reach different conclusions which leads to dialogue and, ideally, a deeper understanding. In any case, other researchers can see how and why the original researchers arrived at their conclusions.

Neutrality

Neutrality is one element which distinguishes scientific research from dogma or propaganda and may be conceived as objectivity or confirmability. Mechanistic natural science seeks to establish objectivity by maintaining distance between researcher and subject. Their relationship is mediated by protocol, theory, and instrumentation. The more the interaction can be mechanized and the person-to-person intimacy reduced, the more objective the research will be. Human science, on the other hand, argues that this distance is not possible or even desirable. Attempts to achieve this kind of objectivity run the risk of removing the meaning of the experience, which Mishler (1986) called "context-stripping." In the attempt to remove the researcher from the research and to quantify experiences, the experience is removed from its context and meaning is altered. Yet, the need to keep scientific conclusions neutral and free from bias remains. Human science suggests that this is possible in qualitative research by a combination of credibility, fittingness, and auditability. Confirmability allows researchers' conclusions to be evaluated in terms of their depth, richness, usefulness, and faithfulness to informants' experiences. It allows evaluations of the degree of fit in other situations. Finally, it allows for thorough examination of the research process. By virtue of allowing confirmation of human science research, researchers can be involved with informants and, at the same time, neutral; they can be scientifically honest while retaining their focus on the subjective realities and meanings given to and derived from informants' life experiences.

Human science recognizes that a phenomenon is affected by being researched; the act of observing a phenomenon changes it. For example, asking informants to report their experiences (either on a quantifiable survey instrument or in an open-ended interview) may induce more careful self-examination which leads to a different, often fuller, experience. While natural scientists view such reactivity as an extraneous variable to be eliminated in the interest of objectivity, human scientists recognize it as a central feature of human experience and interactions. The concepts of credibility, auditability, and confirmability are reminders that attempts to be objective limit the vitality and meaning of the phenomenon being studied and offer alternatives for research that meets the requirements for adequate, "rigorous" research. These concepts can serve as guidelines for good scientific research methods in the realm of the human spirit.

Back to TOP of this page

Go to REFERENCES

AN INTEGRATED APPROACH TO SCIENTIFIC RESEARCH METHODS: Psychological Research on the Human Spirit PART FIVE
John Davis, Ph.D
Metropolitan State College of Denver Department of Psychology
Click here to go to send email to John Davis.
Click here to go to CONTENTS of this article.

The Question of Scientific Adequacy	The question of adequacy of scientific research has been a key issue for human scientists. Natural scientists argue that research methods must be rigorous and that the way to achieve rigor is to quantify, operationalize, manipulate, and control the phenomenon under study. Since much human science research does not quantify, manipulate, or control, it has been deemed inadequate by these experimentally-oriented psychologists. However, both natural science, using a positivist model, and human science, with a phenomenological orientation, share a concern for adequacy. It can be seen that positivism's broad evaluative criteria are reasonable standards for existential-phenomenological [i.e., human science] research. That is, research conclusions should be empirically based; research should strive to be free of personal biases, prejudices, and dogma; other individuals should be able to agree that conclusions are justified by the data; and criteria should be provided for evaluating competing knowledge claims. . . . Although the ontological and methodological assumptions of existential phenomenology and logical positivism differ, both share a common commitment to conducting rigorous, empirical research that is open to careful scrutiny. (Thompson, Locander, & Pollio, 1989, p. 142) While human science researchers often use the term rigor to describe their research methods, adequacy may be a better term. The notion of adequacy suggests research methods that are equal to the task of understanding, are responsive to the nature of the subject matter, and are open to critical evaluation. Although the underlying standards of good research are the same, the particular criteria for scientific adequacy differ in natural science and human science (Kirk & Miller, 1986). Lincoln and Guba (1985, 1994) identify the primary aspects of rigor or adequacy in scientific research and detail how these aspects are expressed differently in the two approaches. Three of these aspects of adequate scientific research (accuracy, consistency, and neutrality), along with their parallel versions in natural science and human science, will be examined.
	Accuracy The foundation of the adequacy of scientific research is its accuracy or, to use Lincoln and Guba's term, its truth-value. In this sense, truth-value can be understood as the extent to which the research findings are true to the phenomenon under study. In natural science, where the goal is to determine cause-and-effect relationships, accuracy depends on internal validity. Internal validity is a means for establishing that the independent variable is the major cause of changes in the dependent variable. This is accomplished by controlling extraneous variables in order to eliminate alternative hypotheses such as subject selection, history, and demand characteristics. Control and systematic manipulation of conditions are central requirements for establishing internal validity. However, achieving such control and manipulation cannot be done with many kinds of psychological phenomena, including experiences of the human spirit. An alternative in human science aims for constructing credible accounts of a phenomenon in its context, rather than control. This is based on describing the deep structure or pattern of an experience or phenomenon in a way that is faithful to the experience. According to this standard, credibility is achieved when someone who has had the experience can say, "This captures my experience better than I could have." By this standard, the accuracy of accounts of the human spirit can be evaluated and established.
	Consistency Consistency is another critical aspect of scientific adequacy. A particular scientific method must be consistent from one study to another and from one setting to another. Inconsistent methods are akin to using a rubber ruler to measure. In natural science, reliability of both tests and experimental settings is one of the first requirements for rigorous research. However, in the view of human science, many phenomena, by their very nature, cannot be replicated. This makes it difficult or impossible to establish reliability for most experiences of the human spirit. An alternative to reliability is auditability. To achieve auditability, the research provides enough details of the research setting, the informants, the data collection methods, and the data analysis to allow other researchers to come to their own conclusions, i.e., to audit the original research process. In good human science, auditability allows other researchers to follow and evaluate a research study. These auditors may reach different conclusions which leads to dialogue and, ideally, a deeper understanding. In any case, other researchers can see how and why the original researchers arrived at their conclusions.
	Neutrality Neutrality is one element which distinguishes scientific research from dogma or propaganda and may be conceived as objectivity or confirmability. Mechanistic natural science seeks to establish objectivity by maintaining distance between researcher and subject. Their relationship is mediated by protocol, theory, and instrumentation. The more the interaction can be mechanized and the person-to-person intimacy reduced, the more objective the research will be. Human science, on the other hand, argues that this distance is not possible or even desirable. Attempts to achieve this kind of objectivity run the risk of removing the meaning of the experience, which Mishler (1986) called "context-stripping." In the attempt to remove the researcher from the research and to quantify experiences, the experience is removed from its context and meaning is altered. Yet, the need to keep scientific conclusions neutral and free from bias remains. Human science suggests that this is possible in qualitative research by a combination of credibility, fittingness, and auditability. Confirmability allows researchers' conclusions to be evaluated in terms of their depth, richness, usefulness, and faithfulness to informants' experiences. It allows evaluations of the degree of fit in other situations. Finally, it allows for thorough examination of the research process. By virtue of allowing confirmation of human science research, researchers can be involved with informants and, at the same time, neutral; they can be scientifically honest while retaining their focus on the subjective realities and meanings given to and derived from informants' life experiences. Human science recognizes that a phenomenon is affected by being researched; the act of observing a phenomenon changes it. For example, asking informants to report their experiences (either on a quantifiable survey instrument or in an open-ended interview) may induce more careful self-examination which leads to a different, often fuller, experience. While natural scientists view such reactivity as an extraneous variable to be eliminated in the interest of objectivity, human scientists recognize it as a central feature of human experience and interactions. The concepts of credibility, auditability, and confirmability are reminders that attempts to be objective limit the vitality and meaning of the phenomenon being studied and offer alternatives for research that meets the requirements for adequate, "rigorous" research. These concepts can serve as guidelines for good scientific research methods in the realm of the human spirit.