Q-methodology in the study of child phenomenology.
Undoubtedly, phenomenologists such as Merleau-Ponty (1942/1963), Sartre (1956), and Giorgi (1970) helped keep alive the flame of mind in psychology during years of domination by methodological behaviorism. Phenomenology has long been and remains a variegated philosophy and psychology (Kockelmans, 1971). Although those claiming to follow phenomenology differ on many issues, all phenomenologists who trace their thinking back to Edmund Husserl, nurturer of contemporary phenomenological psychology (Kockelmans, 1971), take human experience as their fundamental subject of study. The phenomenological perspective leads one to view as promising several attempts in cognitive psychology and in cognitive science to open up experience to observation, including thought listing (Cacioppo & Petty, 1981), protocol analysis (Ericsson & Simon, 1984), and content analysis (Kulkarni & Simon, 1988).
Still needed are research methods that faithfully reflect experience from the point of view of the person studied (Brown, 1980; Kuiken, Schopflocher, & Wild, 1989). The basic problem with cognitive methods to date is their use of researcher-derived categories and scales, the result being that the investigator has already determined what individuals' responses will mean before they have had an opportunity to respond (Brown, 1980). Regardless of how the individual responds to scales derived ahead of time, we have no assurance that the data are telling us more about the individual's experience than the researcher's. Even worse, when the investigator decides in advance that a response to a particular scale reflects a particular construct, the subject's response vivifies the construct and provides it with a spurious existence (Brown, 1980).
Kuiken et al. (1989) addressed the need for methods that yield quantitative data on experience as it is given to the experiencer. Their methodology centered on the application of cluster analysis (Everitt, 1974) to representative verbal statements that participants made in unrestricted descriptions of their experiences. When the investigators examined the derived clusters for their relatively characteristic properties, they found three distinct classes of experiences under the common conditions to which they exposed all participants. Our interests lie in child phenomenology, and the work of Kuiken et al. (1989) is important for us because it suggests that it is possible to rigorously describe aspects of individuals' personal experiences by beginning with participant-centered data and then incorporating multivariate techniques and phenomenological interpretation.
Researchers are giving increasing attention to the development of conscious experience in very young children. Studies have shown that children as young as 24 months are able to use words that express their own feelings and desires (Brown & Dunn, 1991). Although language-in-use is symbolic and self-referential (Brown, 1980; Wittgenstein, 1971), it can also be limiting, especially with very young children who may lack the requisite language skills to faithfully communicate their experiences. Verbo-vocal reports can only describe experience from the point of view of the person studied if that person has the language skills with which to communicate the experience. Fortunately, individuals need not rely on verbo-vocal reports to describe their experiences insofar as they can use objects, pictures, and gestures as symbolic and self-referent forms of communication (Stephenson, 1980).
The primary goal of phenomenological research is "to describe the essential properties of a particular kind of phenomenon" (Husserl, 1913/1967; Kuiken et al., 1989, p. 376). This amounts to taxonomic activity, but where the phenomena are psychological, hence requiring psychological categories. Basically, phenomenologists suggest that psychology has largely bypassed a necessary phase of any science, that of the descriptive classification of its fundamental subject matter--lived experience. This is because taxonomic efforts thus far have overly emphasized nonexperiential phenomena such as purely hypothetical mental structures and overt behavior or "geographical behavior" (Merleau-Ponty, 1942/1963).
Applying Q-Methodology to Phenomenological Descriptions of Children's Experiences
The above considerations dovetail well with the theory behind Stephenson's (1953) Q-methodology. The main thrust of Stephenson's development of Q-methodology was to address the need for tools to study subjectivity. Stephenson distinguished subjectivity from objectivity on the basis of self-reference in the former but not the latter. In the most common application of Q-methodology, the participant first rank orders (e.g., from "most disagree" to "most agree") a sample of Q-items (objects, statements, or other symbols) in terms of some condition of instruction. The items so arranged make up what is called a Q-sort. Q-sorts obtained from several participants or from a single participant under different conditions of instruction are then correlated and factor analyzed. Factors indicate statements, objects, or pictures that have been ranked with much the same pattern. When the Q-sorts of several persons are analyzed, factors indicate persons who have ranked the Q-items in much the same way. In the case of analyzing Q-sorts of a single individual under different conditions of instruction, factors indicate conditions of instruction that bear a family resemblance to one another on the basis of subjectively shared attributes. The researcher interprets factors in terms of their commonly shared perspectives or points of view. Q-methodology is therefore the body of theory and principles that guides the application of technique and interpretation. Thus, Q-methodology is fundamentally a taxonomic undertaking in which the researcher uses derived factors to identify psychological categories.
Both conventional phenomenological procedures and Q-methodology make use of everyday communication. As already mentioned, the only limitation of phenomenological research for study of child phenomenology as typically practiced is the reliance on verbo-vocal linguistic communication. Q-methodology makes it possible for researchers to use alternative means of communication by way of other symbolic material. Stephenson (1980) maintains that preschool children can communicate in a self-referent way with everyday objects and pictures. This ability, and given that children seem to have developed the concepts of "me" and "mine" by the age of three, should allow children as young as preschoolers to perform Q-sorts (Stephenson, 1980).
The purpose of the present study was to explore the benefits of using Q-methodology as a means to provide information about preschool children's phenomenal worlds. The children were not tested for anything but were only asked to represent their own point of view (Stephenson, 1980). The children's sorts of pictorial material were factor analyzed to show the sorts that shared variance between children and sorts that shared variance within a child. The study was exploratory. We had no hypothesis in the hypothetico-deductivistic sense. We merely asked if we could detect regularities in the children's experiences. In fact, because even very young children have different experiential bases, it could be anticipated that there would be considerable variation in the factors that emerged.
We obtained parental permission for 8 children between the ages of 3 yr, 10 mo to 4 yr, 10 mo to participate in the study. All were enrolled in a private preschool program. They performed Q-sorts individually during the first hour of class while seated with the adult female administrator (first author) at a table in the classroom.
Q-Items and Distribution Guide
The children sorted 3- by 5-in. cards on which were pasted pictures of children cut out of a children's fashion magazine. The children in the pictures ranged in age from 3 to 10 yr old, and a single child was pictured on each card. The picture cards depicted children of different ethnic groups engaged in various activities. The Q-items were unstructured in that we made no attempt to systematically include any particular variables (e.g., ethnicity, activity) in them. There were 18 picture cards of boys and 18 picture cards of girls.
We used a distribution guide to show the child how they were to place the Q-items. The guide consisted of seven columns or piles. Each pile was labeled with a different heading ranging from -3 to +3.
The 18 cards with pictures of a child of the same sex as the participant were placed in front of the child on the table. The child was given time to look over the cards and was free to discuss them with the administrator. The administrator then asked the child to "Show me the card that is most like you." This condition of instruction ("Most like you/not like you") was used first for all participants. Once the child had selected a card that was most like them, it was placed under the +3 heading of the distribution guide. The administrator then asked the child to "Show me the card that is most not like you." The card was placed on the distribution guide under the heading -3. The remaining cards were reshuffled by the child and administrator. Next the administrator asked the child to "Show me two cards that are most like you." These two cards were then placed in the column headed +2 on the distribution guide. After the child placed three cards in the +1 and -1 piles, six cards remained. These six cards were placed in the '0' column. The administrator reviewed the sort with the child. Pointing to the '+' side of the chart the administrator said, "These are the pictures that you feel are most like you." Pointing to the negative side of the chart, the administrator said, "And these are the pictures that you feel are not like you." The child was then asked if they wanted to change any of the pictures. If the child wanted to change any of the pictures, they were allowed to do so. Once the child was satisfied with the location of the picture cards on the distribution guide, the administrator recorded the pile number of each card.
The children usually performed one Q-sort per class time although several chose to perform two sorts at a time. Over a period of 6 weeks the children performed eight Q-sorts with the same cards under the following conditions of instruction in the order given:
1. Most like you/not like you (Like you)
2. What Mommy thinks is most like you/what Mommy thinks is not like you (Mommy)
3. What Teacher thinks is most like you/what Teacher thinks is not like you (Teacher)
4. What Big Bird(1) thinks is most like you/what Big Bird thinks is not like you (Big Bird)
5. The very best boy/girl/not the very best boy/girl (Very best)
6. Most like you when you grow up/not like you when you grow up (Grown up)
7. Most like your friend/not like your friend (Like friend).
8. What your friend thinks is most like you/what your friend thinks is not like you (Like you (friend)).
Each participant performed eight sorts, one under each of the eight conditions of instruction. Because each sort was based on 18 pictures and each picture was given a pile number on each sort, the sorts could be correlated using these pile numbers. After each participant's eight sorts were correlated, the resulting correlation matrix was subjected to a principal components analysis. The resulting factors were rotated by means of Varimax and the number of factors to be rotated in each analysis was based on the root-one criterion. Tables 1 and 2 present the results of these separate Varimax rotations for individual boys and girls, respectively. Only the salient conditions in each factor are indicated and saliency was defined as a statistically significant (p |is less than~ .05) absolute value correlation of 0.50 or greater between the sort and the factor.
Table 1 Factor Structures for Boys Boy 1 Boy 2 Boy 3 Boy 4 Condition A B C A B C D A B C A B C Like you X X X X Mommy X X X X Teacher X X -X Big Bird X X -X X X Very best X X -X X -X X Grown up X -X X X Like friend -X X X X Like you (friend) X X X X Note. Uppercase letters indicate factors; salient conditions for each factor indicated by X; -X indicates condition loaded negatively on that factor. Table 2 Factor Structures for Girls Girl 1 Girl 2 Girl 3 Girl 4 Condition A B C A B C D A B C A B C Like you -X X X X X Mommy X X X X Teacher X -X X X Big Bird X X X X -X Very best X X X X Grown up X X X X Like friend X X X X Like you (friend) X X X X X Note. Uppercase letters indicate factors; salient conditions for each factor indicated by X; -X indicates condition loaded negatively on that factor.
It should be noted that each analysis resulted in at least three factors. This certainly suggests that the children sorted the pictures in different ways under the conditions of instruction. Inability to make differential sorts would more than likely have led to single factors. But the analyses also indicate that the sorts shared common variance. Otherwise, more factors in each analysis would in all likelihood have exceeded the root-one criterion.
It is of special interest that the factor structures for each participant are not the same. We interpret this as reflecting that the experiential bases for the sorts performed by each child were not the same. In other words, the technique is capable of revealing differences that may exist between the individuals. But the technique can do more than this. It is also capable of revealing where there may be experiences that are in common between individuals. For example, Boy 1 and Boy 4 made consistent sorts under the instructions of sorting the pictures as to which are most like (unlike) themselves and what their mother thinks are most like (unlike) them. But this in no way dictates that these two children will make other sorts in a manner consistent with one another. For example, Boy 1 made consistent sorts under "Big Bird" and "Very best" instructions, and Boy 4 made consistent sorts under instructions of "Very best" and "Grown up."
As we discuss below, we do not view the actual content of the various factor structures to be especially noteworthy. In the absence of any other systematic information collected on these children, we are hesitant to even speculate as to why Boy 1 and Boy 4, for example, made some sorts that defined the same factors and other sorts that defined other factors (but see discussion section). Furthermore, these analyses do not obviate the possibility that sorts defining the same factors for one child might be performed in a markedly different manner by another child whose same sorts define a factor. The major point of these findings is that they illustrate the value of a technique that can be used to identify similarities and differences both within and between children. With additional information, these similarities and differences may be more readily understood, and thereby provide more access to, and information about, the domain of subjective experience.
The information presented in Table 3 supports this argument. This table presents the Varimax-rotated factor structures based on selected analyses in which conditions of instruction were held constant and the sorts performed by each child of the same sex (4 males and 4 females) were correlated and analyzed. Saliency was defined as a statistically significant (p |is less than~ .05) absolute value correlation of 0.50 or greater between the participant and the factor. Under the instruction "Most like you," Boy 1 and Boy 4 sorted the pictures in a manner consistent with one another, and Boy 2 sorted the pictures in a manner rather contrary to Boys 1 and 4. This suggests that although Boys 1 and 4 have somewhat similar perceptions of themselves, this perception is markedly different from that held by Boy 2. But Boys 1 and 4 made no other sorts in a manner that was directly consistent. In fact, Boy 2's sort under the "Teacher" condition is consistent with Boy 4's sort but unrelated to Boy 1's sort. In brief, our analysis seems capable of revealing the specificity or complexity of psychological commonalities among individual children. For example, commonalities between Boys 1 and 4, on the one hand, and between Boys 2 and 4, on the other, are conditional upon specific conditions of instruction.
Table 3 Selected Factor Structures with Conditions of instruction Constant Condition of Instruction Like you Teacher Boy A B A B 1 X X 2 -X X 3 X -X 4 X X X Condition of instruction Like you Very best Grown up Girl A B A B A B 1 X X X 2 X X X 3 X -X X 4 X X -X Note. Uppercase letters indicate factors; -X indicates child loaded negatively on that factor.
Other sorts that reveal the complexity of how experience is organized in even relatively young children are found in the analyses of the girls' data. Although Girl 1 and Girl 3 made consistent sorts under the conditions of "Most like you," they yielded sorts that are negatively related under the condition "Very best." Thus, although these girls may have similar perceptions of themselves, clearly, what constitutes "Very best" for one girl is at odds with what is "Very best" for the other. They do, however, make consistent sorts under the instructions of "Grown up," revealing that individuals may have similar experiences in several domains while markedly disagreeing in other domains. We cannot determine from this information if either Boy 2 or Boy 3 have "good" or "bad" relationships with their teacher, but the data strongly suggest they have different relationships. Nor can we determine from these data if Girl 1 or Girl 3 is a "good" or "bad" little girl, but it is rather clear that the meaning of "Very Best" is different for the two girls.
Q-methodology provided insight into the organization of young children's phenomenal worlds. The analyses of the participants' sorts of the pictures exhibited clear patterns whereby (a) each individual showed an organized experience (i.e., factor structure), (b) different factor structures over children revealed unique experiences, and (c) when children shared experiences, this was restricted to particular conditions. On the basis of these findings, we argue for the descriptive utility of Q-methodology for phenomenologically oriented investigations of young children's mental activity.
We are especially impressed with the apparent capability of Q-methodology to contribute descriptively rich phenomenal data under conditions of minimal researcher-imposed constraints. We took the initiative only for selecting the pictures the children sorted, for identifying the conditions of instruction, and for determining the distribution of the sorts.(2) Each child was free to distribute the Q-items in any way he or she wished, and, in the manner of phenomenological methods, we made no attempt to determine in advance what each response (placement of picture into a pile) meant. In fact, psychological meaning is one of the issues upon which the kindred areas of Q-methodology (Brown, 1980; Stephenson, 1953) and phenomenology (Giorgi, 1970; Merleau-Ponty, 1942/1963) agree. Their common perspective enables us to avoid the strategy whereby the researcher decides in advance what each participant's response means. Under the aegis of Bridgman's operationalism as misconstrued by psychologists (Kantor, 1938; Koch, 1992) and hypothetico-deductive reasoning, researchers, as well as clinicians, typically predetermine that a particular response means anxiety or depression, low or moderate amounts of anxiety, and so on. What results from this is a concern for a measure's validity and reliability. As Brown (1980) puts it, "only if we can convince ourselves of the validity of a measure, i.e., that a scale measures some platonic thing-in-itself, can we rest content that we are on firm inferential ground; otherwise, we are thrown back into the deep abyss of chance and uncertainty". Researchers so influenced take various precautions lest the participant wrest the meaning of their scales from them. These "controls" include lie scales, measures of acquiescence, multiple items, buffer items, and unobtrusive measures. But these conventions have arisen because researchers are intent on dovetailing the participant's experience of the world with the researcher's experiences. Q-methodology and phenomenology allow us to assess subjective experience, minimally encumbered by researchers' biases.
Although, as in the present study, Q-methodology routinely yields organized factor structures that emerge out of the joint action of participant and researcher, the determination of what the factors mean requires activity subsequent to their extraction. In some cases, Q-methodologists use the phenomenological method of reflection (Giorgi, 1970). They examine the factors, Q-items, conditions of instruction, and participant characteristics; reflect on these and other considerations; and eventually derive an interpretation, not of particular responses (where the participant placed a given item in the Q-sort distribution), but of factors and of the person's selfhood.
Our study nicely illustrates how Q-methodology handles meaning and how meaning is inextricably tied to the researcher's experiences but not imposed on the participant's responses a priori. If the Q-items (pictures of children) already entered into meaningful dialogues between us as researchers and our worlds in a manner we felt at least approximated how the items participated in the childrens' lived-worlds, then we would have reflected upon the pictures, factor structures, conditions of instruction, and so on, perhaps along with further dialogues with the children, to make a preliminary interpretion of the factors in terms of what they meant to each child. In the absence of this information, we can only speculate as to the influences that may have operated to lead the children to sort so differently. Again, referring to the factor structures in Tables 1 and 2, because Boys 1 and 4, for instance, both have a factor (Factor A and Factor C, respectively) defined by how they see themselves and how they perceive their mother sees them, it may be that they have strong senses of attachment to their mothers. Boy 2, in contrast, may have developed to a stage where he is so closely identified with a special friend that he sees himself, sees his friend, and thinks his friend sees him in much the same way, because each of these three sorts define a factor (Factor B) for this boy. It may be that Girl 3 feels no one really knows her because her sort "Like you" is the only salient variable on a factor, or it may be that she was not developmentally able to make reliable differentiations between the other conditions of instruction because five sorts define one factor and four sorts define another. Girl 4 may recognize that, at her "Very best," her mother and teacher would be happy, because these three sorts define a factor, but how she sees herself is really not that way because her "Like you" sort does not enter into the factor.
Speculations such as we have just made address the utility of Q-methodology in helping us to better understand the dynamics of development. If we did know more about the differences among these children, hypotheses about factor structures could have been generated and tested. Furthermore, we could hypothesize which children would sort in consistent ways under the same conditions of instruction. Our admission that we have no idea how the pictures enter into the lives of the children does not diminish the utility of Q-methodology. The children did provide more than satisfactory factor structures and, phenomenally, our perspective leads to some amazement that each child produced other than random sorts of the pictures. A major point is that the factors represent inductively derived categories or codes (e.g., Ray & Delprato, 1989) in the phenomenal domain. Their meaning is relational, not independently measurable, and can only be inferred by identifying what relationships they enter into in the child's lived-world (e.g., Giorgi, 1970). Necessary to establish phenomenal meaning are data from domains in addition to the phenomenal, such as the social, geographical, and organic.
The psychologist whose approach is guided by psychological operationalism, hypothetico-deductive reasoning, and attendant sharp lines of demarcation between description and explanation will tend to find our data "merely descriptive," thus incomplete. We agree that our research is descriptive. However, from our perspective, psychology has largely bypassed a necessary component of any science--the descriptive and taxonomic (cf., Giorgi, 1970; Verplanck, 1970; 1983). This is strikingly evident in the case of human experience. With the exception of research exploring intuitive (e.g., Collier & Kuiken, 1977) and statistically aided (Kuiken et al., 1989) taxonomic strategies for classifying adults' phenomenal descriptions, psychology has very little to offer for identifying the fundamental categories of experience as it is given to the individual. That the young children in our study sorted the visual stimulus objects to yield independent factors (i.e., categories) suggests that Q-methodology is a promising tool for taxonomic activity in studies of the phenomenology of young children. Given that Q-methodology has always been above all else concerned with subjectivity (Brown, 1980; Stephenson, 1953, 1980), it is a suitable candidate to contribute to Giorgi's (1970) call and prognostication that "a real breakthrough in psychology will be achieved when we learn to study the phenomenal domain in its own terms and in a rigorous and systematic manner".
Finally, of considerable import is Q-methodology's firm foundation in the philosophy and psychology of Kantor's (1959) interbehaviorism (Stephenson, 1953). This ensures that descriptive constructions derived from proper applications of Q-methodology will not misconstrue what has traditionally been regarded as mentality to be other than completely naturalistic. Our endorsement of a marriage between Q-methodology and phenomenology does not apply to all of what passes under the banner of phenomenology today. Lichtenstein (1971) clearly addressed aspects of phenomenology that cannot, and aspects than can, be properly incorporated into the science of behavior. Stephenson (1953) earlier distinguished between phenomenological assumptions that impede understanding and those that stand more of a chance to advance our understanding of subjectivity. Subsequently, Stephenson (1988) discussed at length various aspects of phenomenology and their relationship to Q-methodology. Briefly put, Stephenson (1988) finds Husserl's (e.g., 1913/1967) "mode of thought, and postulates representing his phenomenology, ..., congruent with the principles |of Q-methodology~". We believe the following reply to a hypothetical, but representative, advocate of one popular version of the traditional dualistic way of handling subjectivity captures the essence of how Q-methodology might redirect thinking about "inner experience" in the direction of Husserl's phenomenology.
You refer to our experience, the world of direct experience being encoded, and what we experience as but a human (personal) version of reality. You reach the conclusion that objectivity in the sense of unfettered reports of the external world is impossible. I agree on the latter, but suggest that the notion of experience is mischievous. I take the more radical position that the inner world of experience is more of a telling than a place or happening (e.g., Schafer, 1978). We do not experience cognitive processes or mental contents that introspection reveals in pure form. There are no inner (or private), as distinct from outer (or public), experiences. There is no inner world as distinct from outer world. Experience is always a construction related to the personal and subjective--but not to the mental as opposed to the physical. Therefore, I do not know any more about my experience than I do about any other aspect of the world. Turning this around, I have the potential of knowing as much about my or your experience as I have knowing about any other aspect of the world. In brief, my and others' experience is as much a part of the world, and subject to understanding as any other aspect of the world (as by way of Q-methodology). It is only cultural happenings that have contributed to the idea that personal, subjective experiences are special and virtually, if not completely, other-worldly (see Kantor, 1963, 1969).
1 At the time of this study, Big Bird was a character in a widely distributed children's television program on a government-supported network in the United States.
2 Indeed, the children could have actively participated in selecting the pictures as well as the conditions of instruction, in which case we would have imposed only the shape of the Q-sort distribution on the children, and previous research shows that the shape of the distribution is without effect (Brown, 1980).
BAARS, B. J., & BANKS, W. P. (1992). Editorial. Consciousness and Cognition. 1, 1-2.
BROWN, J. R., & DUNN, J. (1991). 'You can cry, mum': The social and developmental implications of talk about internal states. British Journal of Developmental Psychology, 9, 237-256.
BROWN, S. R. (1980). Political subjectivity. New Haven, CT: Yale University Press.
CACIOPPO, J. T., & PETTY, R. E. (1981). Social psychological procedures for cognitive response assessment: The thought-listing technique. In C. Glass & M. Genest (Eds.), Cognitive assessment (pp. 309-342). New York: Guilford Press.
COLLIER, G., & KUIKEN, D. (1977). A phenomenological study of the experience of poetry. Journal of Phenomenological Psychology, 7, 209-225.
ERICSSON, K. A., & SIMON, H. A. (1984). Protocol analysis: Verbal reports as data. Cambridge, MA: MIT Press.
EVERITT, B. (1974). Cluster analysis. London: Heinemann Educational Books.
GIORGI, A. (1970). Psychology as a human science. New York: Harper & Row.
HUSSERL, E. (1967). Ideas: General introduction to phenomenology. (W. R. B. Gibson, Trans.). London: Allen and Unwin. (Original work published 1913).
KANTOR, J. R. (1938). The operational principle in the physical and psychological sciences. The Psychological Record, 2, 1-32.
KANTOR, J. R. (1959). Interbehavioral psychology (2nd ed.). Bloomington, IN: Principia Press.
KANTOR, J. R. (1963). The scientific evolution of psychology (Vol. 1). Chicago, IL: Principia Press.
KANTOR, J. R. (1969). The scientific evolution of psychology (Vol. 2). Chicago, IL: Principia Press.
KOCH, S. (1992). Psychology's Bridgman vs Bridgman's Bridgman. Theory & Psychology, 2, 261-290.
KOCKELMANS, J. J. (1971). Phenomenological psychology in the United States: A critical analysis of the actual situation. Journal of Phenomenological Psychology, 1, 139-172.
KUIKEN, D., SCHOPFLOCHER, D., & WILD, T. C. (1989). Numerically aided methods in phenomenology: A demonstration. Journal of Mind and Behavior, 10, 373-392.
KULKARNI, D., & SIMON, H. A. (1988). The process of scientific discovery: The strategy of experimentation. Cognitive Science, 12, 139-175.
LICHTENSTEIN, P. E., (1971). A behavioral approach to "phenomenological data." The Psychological Record, 21, 1-16.
MERLEAU-PONTY, M. (1963). The structure of behavior (A. L. Fisher, Trans.) Boston, MA: Beacon Press. (Original work published 1942).
RAY, R. D., & DELPRATO, D. J. (1989). Behavioral systems analysis: Methodological strategies and tactics. Behavioral Science, 34, 81-127.
SARTRE, J.-P. (1956). Being and nothingness. New York: Philosophical Library.
SCHAFER, R. (1978). Language and insight. New Haven, CT: Yale University Press.
STEPHENSON, W. (1953). The study of behavior: Q-technique and its methodology. Chicago, IL: University of Chicago Press.
STEPHENSON, W. (1980). Newton's fifth rule and Q methodology; Application to educational psychology. American Psychologist, 35, 882-889.
STEPHENSON, W. (1988). William James, Niels Bohr, and complementarity: V - Phenomenology of subjectivity. The Psychological Record, 38, 203-219.
VERPLANCK, W. S. (1970). An "overstatement" on psychological research: What is a dissertation? The Psychological Record, 20, 119-122.
VERPLANCK, W. S. (1983). Preface. In N. W. Smith, P. T. Mountjoy, & D. H. Ruben (Eds.), Reassessment in psychology: The interbehavioral alternative (pp. xi-xxv). Washington, DC: University Press of America.
WITTGENSTEIN, L. (1971). Philosophical investigations (3rd ed.). New York: Macmillan.
|Printer friendly Cite/link Email Feedback|
|Author:||Taylor, Priscilla; Delprato, Dennis J.; Knapp, John R.|
|Publication:||The Psychological Record|
|Date:||Mar 22, 1994|
|Previous Article:||Intermittent consequences and problem solving: the experimental control of "superstitious" beliefs.|
|Next Article:||The effects of schedules of reinforcement on operant sequential stereotypy in humans.|