This article, published under Winell's previous name, Wadsworth, with Donald Ford, appeared in the Journal of Counseling Psychology, Vol. 30, Pp. 514-526, 1983. It includes findings from her doctoral research.

Marlene Winell (Wadsworth) and Donald H. Ford
Pennsylvania State University

Journal of Counseling Psychology, 1983, Vol. 30, No. 4, 514-526.
Personal goals and the broader topic of self-direction are presented as important areas of study for both theoretical and applied reasons. The assessment of personal goal hierarchies is considered a useful step in counseling procedures. A new method for measuring goal hierarchies was developed using an in-depth interview procedure with a subject group of 53 adult men. A coding system was then used for constructing an AIMS (Adult Intentional and Motivational Systems) Chart for each subject, using dimensions including life domain (e.g., work, family, social life), level of generality, and means-end connections. Evidence of validity and reliability suggest the continued use of the chart for research on self-direction and adult development as well as intervention efforts. Case study material is presented that indicates promise for counseling applications.

Personal goals have always been important for counseling psychology. Clients participate in individual or group treatments because they want to move from their present situation toward some desirable alternative. A task to accomplish early in the process, then, is to assess with a client relevant concerns and goals. This study was an effort to develop a method for examining an individual's complete hierarchy of personal goals. Counseling procedures often focus on specific goals related to clients' presenting problems. In contrast, the method described here allows counselor and client to examine the client's full array of life concerns and goals. The "presenting problem" can then be seen in the context of the many interwoven aspects of a person's life. More broadly, the method is useful for clarifying objectives in a number of life domains and thereby for facilitating self-direction.

In this article, we discuss the importance of studying personal goals for understanding human functioning and for counseling applications. As a context for this study, the more general topic of self-direction is also addressed in terms of its theoretical and practical relevance. Self-direction is used here to refer to the many aspects of human functioning that collectively indicate that human beings actively influence their own destinies. Within the living systems framework used, the concept more specifically refers to the directive function of cognitive processes that serve to guide the rest of behavior. Personal goals  refer to all of the consequences toward which a person's behavior is directed. That is, in addition to the everyday meaning of the word goals, which implies achievement-oriented striving, we are referring to any and all desired satisfactions ranging from eating lunch to creating world peace.

If it is granted that self-direction and personal goals are important, how do we go about studying them? The study described here represents an initial attempt to interview people about their goals and then to assess each individual's complete goal hierarchy. A new instrument, the AIMS (Adult Intentional and Motivational Systems), Chart is presented, along with validity and reliability information. Research implications are discussed along with anecdotal information and suggestions for counseling applications. It is believed that the procedure can be fruitfully used in counseling for diagnosing problems, setting goals, increasing motivation, and examining progress. The method also has potential for studying the subject of self-direction and for observing developmental changes over time.

Theoretical Rationale

Humans have the unique and remarkable ability to anticipate and symbolically represent the future and to construct plans for achieving their desired goals. A basic proposition of this study is that understanding a person requires understanding the consequences toward which his or her behavior is directed-that is, his or her purposes or goals. The other aspects of the person's functioning can then be understood in relationship to that set of variables. After an extensive review, Boden (1972) concluded that research in this area of purposive behavior is sparse, reflecting the historical behavioral bias in psychology. However, despite that bias, interest in personal goalseeking has been continually present in some aspects of psychology. In certain personality theories, the ability to be self-directing has been associated with healthy or mature functioning (Ford & Urban, 1963). For example, Gordon Allport (1961) considered the ability to make decisions and set personal goals a sign of psychological maturity. Psychoanalytic theory gives a prominent role to ego functions. In existential psychology, the "authentic" individual is able to form values and make goal-oriented decisions independent of the "crowd" (Hall & Lindzey, 1970).

In recent years, theory and research about humans that is based on a model of the human as an active, self-directing organism has become increasingly significant (e.g., Carver & Scheier, 1981; de Charms, 1968; Deci, 1980; G. Miller, Galanter, & Pribram, 1960; J. Miller, 1978; Powers, 1978). It is argued that goal-seeking behavior is a necessary attribute for a living system's survival and functioning (Klinger, 1977; Locke, Cartledge, & Knerr, 1970). Some scholars have even suggested that it is impossible to conceive of human functioning or to rationally explain human behavior without concepts of' purpose or goal orientation (e.g., Tolman, 1932; von Bertalanffy, 1968). Thus, Powers (1973) has suggested that Modern investigation should strive to "restore purposes and goals to our concept of human behavior in a way that doe-, not violate direct experience or scientific methods." (p. 351)

Within this stream, 1). Ford (Note 1) has been evolving a new conceptual framework for understanding individuals as dynamically organized structural-functional units. That framework uses a living -systems model (e.g., Maturana, 1975; J. Miller, 1978) as a way to understand behavioral organization and functioning. Among the basic attributes of this view is a characteristic of humans called the directive function.  Ernest Hilgard (1977) has referred to this same aspect of cognitive ability as the executive function. This directive or executive function is one aspect of general cognitive functioning.

Cognitive psychology is one of the "hottest" areas of study in psychology right now, and for good reason. It has long been a basic premise in the natural sciences that different forms of matter and energy are the basic stuff of which the universe is composed. Biology has come to represent organisms as open, living systems that continue to exist, grow, change and function as physical entities through the exchange of energy and material with their contexts. Many scientific views of the "nature of man" have implicit within them the notion that "that is all there is."

However, since World War 11 another idea has begun to come into focus. Something called information is viewed as equally fundamental and different from, but related to, energy and material. Information is what organizes and controls the functioning of physical entities (Jantsch, 1980). This powerful new idea has spawned technological advances that are of major consequence to human societies, as illustrated by servomechanisms, engineering control systems, and computer technology. The rapid expansion of interest in cognitive psychology is one important manifestation of the spread of this new idea through our culture. Stated simply, human development and functioning requires not only continual material and energy transaction with the environment but continual information transaction as well.

One basic role of the human nervous system is to collect and use information to regulate one's behavior vis-a-vis the environmental context. In Ford's framework, a significant aspect of this human informational capability is the directive function. People use a continual flow of information from inside and outside their bodies. With experience, they begin to abstract and store information about relationships among events. These events vary in their consequences for a person and take on varying values as a result. As infants, individuals first come to recognize, then to anticipate, and finally, to "intend" the occurrence of various events. These informational intentions selectively trigger coordinated patterns of other behavioral functions to produce the intended consequences.

This use of informational abstracts, derived from a person's history of experience, to "intend" behavior to produce specified consequences is what we mean by the "directive" function. In servomechanisms or engineering control theory, this function is identified by two terms-namely, the  reference signal and feed forward. In humans, this capability is an innate attribute resulting from evolutionary history. It has been displayed by infants in studies of sucking behavior. For example, a 6-week old infant can learn to suck to focus a picture on a backlighted screen or reduce sucking if the picture begins to go out of focus (Bruner, 1968).

Through years of experience, every person accumulates abstract representations of many kinds of consequences to be sought or avoided. We will call these "goals." Moreover, these multiple goals evolve in organized patterns such that some come to be seen as stepping stones toward the achievement of others. We refer to this as being hierarchically organized. The content of a particular person's goals are abstracts from his or her particular experience. Wide individual differences are thus evident in people's goal hierarchies.

Personal goals are becoming a more important part of theorizing about human behavior. As humans are increasingly conceptualized as active beings who influence their own lives, the role of self-directing processes must be taken into account. To understand an individual, a key component would appear to be understanding that person's unique hierarchy of goals, from day-to-day satisfactions to vague dreams and aspirations.

Applied Rationale

Areas of applied psychology have long recognized the value of fostering skills involved in clarifying and making progress towards personal goals-that is, educational efforts to motivate independent learning (Gardner & Gardner, 1978); programs to enhance worker motivation (Locke, Shaw, Saari, Latham, 1981); clinical and counseling procedures to promote self-direction, selfcontrol, self-help, or self-fulfillment (Bandura & Simon, 1977; Kanfer, 1979; Kiresuk & Sherman, 1968; Kolb, Winter, & Berlew, 1968; Mahoney & Thoreson, 1974; Rogers, 1951). It is interesting to note that although the purposive characteristics of human life have been largely rejected as epiphenoma by scientific psychology, fields that have had to focus on people with "real problems" in "real life" settings have continued to consider these characteristics important. When one has to deal with problems such as divorce, suicide, delinquency, and worker productivity, it appears impossible to ignore people's self-directing capabilities. Research has related beliefs about personal causation to happiness and personal effectiveness (Bandura, 1977; Lefcourt, 1976; Seligman, 1975; Brim, Note 2). Studies show that people who can list more sources of satisfaction (i.e., goals) rate their lives as more meaningful (Klinger, 1977). Personal goals appear to provide stability, continuity, personal identity, and satisfaction in a person's life (Buhler, 1968; Levinson, Darrow, Klein, Levinson, & McKee, 1978; Neugarten, Havighurst, & Tobin, 1961), and to be significantly involved in personally satisfying adult development and aging (Atchley, 1975; Lowenthal, 1971).

Concern with personal goals is apparent in many counseling procedures. In behavior therapy, for example, specific goal clarity has been emphasized (Mahoney & Thoreson, 1974; Weigel & Uhlemann, 1975). Goal Attainment Scaling has been developed as a procedure for evaluating results of therapy (Kiresuk & Sherman, 1978). In vocational guidance, consideration of an individual's objectives is primary. Psychodynamic counselors focus on "ego strength," one attribute of which is self-direction (Kolb, Winter, & Berlew, 1968). Existentialists attempt to foster courage and responsibility in making "choices." Humanistic counselors attempt to provide the conditions that allow proactive forces to emerge in individuals, permitting them to pursue their ideals (Rogers, 1951).

Counseling and psychotherapy efforts are proceeding to recognize the need and the ability people have to direct their own lives. Self-attributed change appears to be maintained to a greater extent than change attributed to an external agent or force (Kanfer, 1979). Hence programs of self-help or self-directed change have been designed (e.g., Kolb, Winter, & Berlew, 1968). Identifying important personal goals is often the first step in the counseling process. Only then can people efficiently design strategies for growth or change.

This study is part of a program of research designed to better understand and cultivate the important human attribute of self-direction. The first step has been to develop a method for assessing personal goal hierarchies. Although some instruments have previously been developed for the study of related subjects (e.g., see Buhler, 1965, on "life's basic tendencies"; Erickson, Post, & Paige, 1975, for a "Hope Scale"; Klinger, Barta, & Maxeiner, 1980, on "current concerns"; Rokeach, 1968, on values; and Strong, 1943, on vocational interests), little has been done to thoroughly assess the goals that people have, and their interrelationships. This study, Project AIMS (Adult Intentional and Motivational Systems), has been an effort in this direction. The AIMS Chart was developed to portray the substance and pattern of a person's goals, ranging from immediate satisfactions to long-range ambitions and abstract ideals in basic domains of life. The remainder of this article describes this study, with particular attention to the new instrument, the AIMS Chart. Other procedures and measures used in the study are also presented briefly to provide an overall understanding of Project AIMS and because other measures were used in studying the convergent/discriminant validity of the AIMS Chart.

Study Design and Methods

Subjects

 Table 1
Two groups of male adults composed the nonprobability sample used in this study: (a) 25 undergraduate seniors at Pennsylvania State University, randomly selected from a list of male students within three terms of graduation with a baccalaureate degree, and (b) 28 blue-collar workers at Pennsylvania State University between 55 and 62 years of age selected from a list provided by the University Personnel Office. Seven of the original 60 subjects were dropped from the study because of incomplete data. These two very different groups were chosen to provide major differences in personal goal hierarchies, the first group being ready to launch their careers and adult lives and the second group approaching retirement. These differences were thus expected to help test the construct validity of the new measure, which should distinguish the goal hierarchies of such disparate groups. Table I summarizes some characteristics of the two groups. The college seniors were majors in most colleges of Pennsylvania State University, that is, Agriculture, Arts, Business, Earth and Mineral Sciences, Engineering, Human Development, Liberal Arts, and Science. The blue-collar workers were in technical, maintenance, janitorial, or food service jobs in eight different administrative units of the university. In this study, the sample was limited to males to control for possible gender differences.

Variables and Measures

This report focuses primarily on variables representing goal hierarchies and a few other variables from the larger study. Data from most of the other measures are not examined here.

The AIMS interview. Since goals are informational abstracts about intended or desired consequences, they cannot be observed directly. Therefore we must assess them through their manifestations, as physicists do in studying nuclear particles. Two types of manifestations were used in this study. First, on the principle that if one want,; to know what people are thinking-ask them, we did so. Second, recognizing that sometimes people cannot or will not report their thought processes fully, we assumed that the consequences they sought would be manifested in what they did. Therefore we sought a behavioral sample of their activities.

After informal pretests of several approaches, it was concluded that a three-part structured personal interview was most. appropriate for eliciting the in-depth information needed. First, the subject-, were asked to talk about what was generally important to them, or what they wanted out of life. Second, they were asked to systematically describe their activities day by day for the previous 4 days. This was considered an approximation of a behavioral observation. After their activities were recounted, subjects were asked to explain their goals in relation to each activity. Third, they were asked to describe what they sought in six domains of life-that is, family, work and school, social life, leisure, personal growth and maintenance, material/environmental, and other. A few general questions then remained concerning additional goals, satisfactions, priorities, and related beliefs and attitudes.

Information from the three parts provided rich material for the primary coding procedure of constructing a goal hierarchy. The first part tended to bring out broad, long-term goals belonging in the "upper" part of the hierarchy. This part also revealed thoughts that were most spontaneous-that is, subjects named life goals "off the tops of their heads." The behavioral section revealed day-to-day events, short-term goals, and some medium-term goals. The third section provided more detail about satisfactions, activities, goals, and plans for each life domain, thus filling in the gaps remaining on all levels of generality. By presenting the same questions in different ways, subjects were given several opportunities to respond and thus to provide a rather large representative sample of'  activities and goals. This thoroughness was considered important to ensure the content validity needed to depict the structure and content of each subject'- entire goal hierarchy. The format also made it possible for less verbal subjects to provide the needed information.

AIMS Chart coding procedures. The interviews were tape recorded, transcribed, and coded by a group of five coders who had been trained for about 35 hours. This portion of Project AIMS most represents the effort to make an innovative contribution in assessing personal goals. The coding procedure involved two quite difficult and different steps. The first amounted to a content analysis of the AIMS interview to obtain a list of personal goals. The second involved constructing the AIMS Chart from that list of goals. Coding rules for making these two kinds of judgments were developed and refined during the preliminary work and the coder training period. More specifically, the coders listened to a tape recording of the interview while reading a typescript of the interview and underlining phrases that represented goals. They then produced a list of goals for each person. The second step was to place each goal statement on the AIMS chart in one of 28 categories representing different kinds of goals as specified in the coding criteria. As this was done, arrows were drawn between items that had a means-end connection mentioned by the subject. Thus, there were three kinds of coder agreement or reliability problems: (a) agreement on the list of goals, (b) agreement on the goal content, and (c) agreement on the relationships among the goals in the personal goal hierarchy. Cohen's kappa (Cohen, 1960) was used as the primary index of intercoder agreement in this study. This kappa statistic is an agreement index for categorical data that is corrected for chance agreement. Therefore it is a more stringent test of agreement than the simple percentage agreement typically reported. Suggested benchmarks for evaluating the strength of kappa are .41-60 = moderate; .61-.80 = substantial; .81 -1.00 = almost perfect (Landis & Koch, 1977). Figures I and 2 are examples of two such charts.

 click to enlarge
 As can be seen, the AIMS Chart is divided into seven broad life domains and four goal levels, ranging from short, quite specific goals to very broad, long-term goals (designated as A, B, C, D levels). The life domains were selected and defined by considering domains used in of  her research (e.g., Andrews & Withey. 1976; Klinger, Barta, & Maxeiner, 1980), what might be all appropriate categorization for this kind of instrument, and what might he useful in counseling. The Four levels represent what actually amounts to a continuum with dimensions of both time and generality of goals.

During the coder training period, pairs of' coders worked on each case by completing certain steps separately and then collaborating for final results. It was determined I hat this procedure produced higher levels of' agreement than with single coders. Intercoder agreement (between pairs) was established during training for both detailed and aggregated aspects of the list -making step and the charting. Cohen's kappa was sufficiently high, ranging from k = .75 to k = .98.

On the study sample itself, the coding was done by single coders because of limited resources. Therefore, a subset of 14 randomly selected interviews was recoded by the project coordinator and coder-expert reliability calculated. This was done for (a) the step of making a list of goal statement,;, (b) the task of charting each goal, and (c) the pattern of goals displayed on the chart by several methods of aggregated frequencies.

 click to enlarge
 Validity information of three kinds was also obtained. First, an assessment of content validity was made by meeting with eight seniors and six blue-collar workers to show them the charts that were constructed from their interviews. They were asked to comment on the accuracy oft he goals I listed, to suggest deletions or additions, and to evaluate the adequacy of the chart for representing their system of personal goals.

 Second, convergent and divergent validity were examined by correlating variables derived from the AIMS Chart with variables from the other instruments, using the full study sample. "Goal directedness" variables were selected for convergent validity and the others for divergent validity. Two of the variables derived from the AIM" Chart were the total number of 'longer-term goals (all the goals categorized oil the C and D levels of' generality) and the total number of shorter-term goals (A and B levels). The first, (longer-term goals), was interpreted as a measure of' goal-directedness in the sense of planning ahead to achieve desired objectives. The second, shorter-term goals, was thought to measure an orientation toward more immediate satisfactions. In the convergent/divergent validity matrix, additional measures of goal-directedness, were expected to correlate with the number of longer-term goals, and three other variables (to he described in a moment) were not expected to correlate with goal-directedness.

Third, construct validity was studied by testing five hypotheses concerning ways the measure should distinguish between the older and younger men. These included:
    1      The proportion of total goals that are coded as longer-term goals (Levels C + D) on the chart will be greater for students than for the blue-collar workers.
    2      The proportion of total goals that are in the family domain will be greater for the blue-collar workers.
    3      The proportion of total goals that are in the material/environmental domain will be greater for the blue-collar workers.
    4      The proportion of total goals that are in the work-and-school domain will be greater for the students.
    5      The proportion of total goals that are in the social domain will be greater for the students.

These predictions were based on priority rankings made by the subjects on the written questionnaire and on considerations of lifestyle and developmental periods represented by these two groups.

Interviewer Rating Scale (IRS). Qualitative ratings were made by the person who interviewed the subject and again later by a different person, who listened to the tape and coded the interview. Global dimensions were rated, such as life satisfaction, goal -directedness, and perceived control over life. These questions were similar to those asked of the subject in the final part of the interview so that the relationships between subjective and objective ratings on the same variables could be examined. Each item of the 15-item IRS was evaluated for coder-interviewer agreement using kappa. In addition, intercoder agreement was calculated for 26 cases that were coded again by a second person. For the majority of items on the IRS, agreement levels were substantial, having a mean of k = .70, and ranging from k = .62 to  k = .90.

Self-administered questionnaire. The final form of data collection was a paper-and -pencil questionnaire completed by each subject at home and returned by mail. The package included measures of life satisfaction, perceived control, perceived ability, goal-directedness, social role involvement, activity level, and others. The response format used a Likert-type scale ranging from I to 5. Established instruments were used in this battery as well as scales pretested in a previous study. Repeated measures reliability for these over a 2-week interval ranged from r = .58 to r = .86, with a mean of r = .73. Internal consistency (coefficient alpha) ranged from a = .58 to a = .70.

Data Collection Procedures

 Potential subjects were contacted by phone. the study explained, and their cooperation sought as volunteers. If they agreed, an appointment for data collection was scheduled at their convenience for 'approximately 2 hours.

Four trained interviewers, three women and one man, conducted 15 interviews each, with each one interviewing approximately equal numbers of older and younger men To maintain consistency among the interviewers and across time, the project coordinator reviewed a random selection of taped interviews during the data collection period and provided appropriate feedback to the interviewers. Subjects were seen privately in consulting rooms on campus or in their homes, depending on their preference, for a period averaging 80 minutes. The self-administered questionnaire was left with them for later completion and mail-in. Telephone follow-up was used to encourage return of the questionnaire. The IRS was completed by the interviewer shortly after the interview and by the coder at the time of coding.

Results

Reliability

The results of calculating coder-expert reliability for the AIMS Chart were encouraging.

Lists of goals. To answer the first reliability question concerning list making, the 14 recoded interviews were used to compare the lists of goals derived from the interview protocol. Each page of the protocol was treated as 10 units-that is, 10 potential opportunities for identifying a goal statement. The mean percentage of agreement for listing the same goals was 86.2%, ranging from 72.6% to 95.3%. Cohen's kappa was also calculated, using .5 as the proportion of judgments in which agreement was expected by chance. This resulted in a mean of k = .72 and a range of from k = .45 to k = .91. Thus, coder agreement for identifying subjects' goals was, on the average, substantial. The number of goals on these lists averaged 65, ranging from 17 to 108.

Classifying goal content. Second, reliability was calculated for the step of classifying each goal in one of 28 categories on the AIMS Chart produced by the seven domains and four levels. The goal statements examined were those that had been identified by both coders in the list-making step of the coding process. Charting agreement was analyzed for the life domain dimension (work, family, etc.), the level of generality (using either absolute agreement or agreement within one level), and agreement on both dimensions (with absolute agreement on level or within one level). Table 2 provides percentages of agreement and the Cohen's k statistic for each analysis. Chance agreement was considered 1/7, 1/4, and 1/28 for dimensions of domain, level, and both, respectively. Note that the lowest agreement figure on Table 1 is 58.6% for both domain and level. That means that 58.6% of the time two independent coders placed the same statement in exactly the same box on the chart with 28 boxes to choose from.

 Table 2

Goal patterns. As a third method of addressing reliability, the frequencies of goals listed in the seven domains and four levels were aggregated to provide a less stringent and more pattern-oriented way of examining the charting agreement. Other frequencies were also compared, including the number of goals listed on Levels A and B (shorterterm), the number on Levels C and D (longer-term), the total number of goals identified, and the total number of meansend arrows drawn. A Pearson productmoment coefficient was used to compute coder-expert reliability for the 14 cases that were recoded. As seen in Table 3, these correlations were quite high in most instances. The mean number of goals in each frequency group is also reported to show the nature of the variable being correlated. It appears that in general, higher frequencies produced higher correlations. Domain #7, Other/General, was essentially a leftover group with goals that did not fit the othercategories. With such a vague definition and a low frequency, this reliability coefficient did not turn out to be significant.

Validity

Content validity. To begin with, the presentation of completed charts to 14 subjects resulted in the unanimous opinion that they were generally very accurate. A number of subjects noted that some of their short-term goals had changed since the interview simply because of their brief nature (e.g., write a paper, play golf, etc.). As expected, the medium and longer-term goals were more stable. Nevertheless, the subjects considered the pattern of goals to be an accurate representation of their goal hierarchy, with some substitution of the more immediate goals. Viewing the charts was very interesting for most of the subjects as well. For those more introspectively inclined, it was a thought-provoking experience to see the pattern of their daily activities together with broader goals and values. One person remarked that it was amazing to see one's life laid out on a 24 X 36 in. piece of paper.

 Table 3

The interviews in this study indicated that people were able to discuss their personal goals and that each subject had a unique goal hierarchy. No one was without knowledge of consequences that were personally important, and each hierarchy was amenable to our coding procedure. The examples in Figures I and 2 demonstrate some characteristic we expected of personal goal hierarchies. Some goals in these interviews were described as subgoals leading to larger goals, indicated by arrows on the AIMS Chart. For example, in the case of the college student depicted in Figure 1, Domain #5 (personal growth and maintenance) includes running, which leads to keeping in shape, which is connected to being healthy in general. These life domains are not unrelated, of course. This student is interested in getting passing grades partly to advance his career and partly to please his parents, a part of his family life domain. Theoretically, goal hierarchies are complexly organized, often with one subgoal serving more than one larger goal (as illustrated here) and with multiple subgoals leading to the same larger goal. This latter case is illustrated by the student's depiction of his social life-that is, the fraternity dinner and his attendance at a party were both for the purpose of "getting to know new guys."

When the goals from an entire interview have been coded on the AIMS Chart, we found that is is possible to examine a number of interesting attributes, both concerning the pattern of the entire hierarchy and the specific goals. One could study, for example, the extent of a person's goals, their diversity, content, immediacy, generality, -abstraction, and means-end organization. Among some of our subjects, contrasts could also be made between people who emphasized subjective consequences like "peace of mind" versus those who talked about concrete, observable goals like "buy a home." Priorities among life domains were also apparent. In the examples depicted here, the older, blue-collar worker is clearly more concerned with the happiness of his family life than climbing a career ladder. Moreover, he makes very few connections between subgoals leading to larger goals, as did many college students in their plans for school and career. Perhaps, nearing the latter part of his life, he does not plan years in advance, preferring to focus on day-to-day satisfactions.

 Click to enlarge Table 4
Discriminant validity. The result of the convergent/divergent discriminant validity analysis was that, in general, the correlations supported the predictions made concerning goal-directedness, (see Table 4). For convergent validity, three other measures of goal-directedness were correlated with the number of longer-term and shorter-term goals from the AIMS charts. The measured used for convergent validity were (a) a Goal-Setting Scale (developed by M. Ford, Note 3); (b) three items from the questionnaire filled out by the subjects (Using a Likert-type scale of 1-5, these were Do you think it is better to plan your life a good way ahead, or would you say life is too much a matter of luck to plan very far? How much do you tend to set goals for yourself that. you try to achieve? How much do you think you plan your activities with certain goals in mind?); and (d) six items from the Interviewer Rating Scale consisting of similar Likert-type questions (e.g., To what extent does this person seem to deliberately plan to obtain the consequences they value?).

For divergent validity, three other variables not expected to correlate with goal-directedness were used. They were (a) activity level, as measured by an Activity Check-List (based on the Pleasant Events Schedule by MacPhillamy & Lewinsohn, Note 4) with 50 items such as "buying things for myself," "giving a party," and "being with my spouse"; (b) busyness, as measured by six items from the. written questionnaire, which asked "How busy would you say you are in this area?" about each of six life domains; and (c) social role demand, as measured by questionnaire items concerning time spent on fulfilling roles such as parent, friend, student, and spouse.

The first variable from the chart, the number of longer-term goals (Levels C & D), was correlated with the other measures of goal -directedness. These correlations were higher than for the variable of shorter-term goals (Levels A & B), which, because of the more immediate nature of short-term goals, was expected to be less related to the construct of goal -directedness. In addition, both longer-term goals and shorter-term goals had low correlations with the unrelated variables of activity level, busyness, and social role demand. These correlations were also produced for the two subject groups separately (n = 25 and 28, respectively). The result was that some correlations were quite different while the general pattern remained the same.

Construct validity. Finally, the five hypotheses used to test construct validity were confirmed in each case. The expected differences between the two subject groups were tested using a t test procedure. As seen in Table 5, these differences were found to be highly significant. Students expressed a higher proportion of longer-term goals than did the blue-collar workers. They placed a greater emphasis on work and school life and social life, whereas blue-collar workers had a higher proportion of family goals and material/environmental concerns. In addition, the domains of life not expected to differentiate the groups (personal, leisure, and other) were indeed nonsignificant.

 Table 5

It should be noted here that proportions were used instead of raw numbers of goals because of confounds such as verbal ability. However, given that such confounds did not appear to exist, certain expected differences provide additional evidence of construct validity: (a) Students reported a greater total number of goals, with a mean of 106.3 versus 81.8 for the blue collar workers, t = 3.48, p < .001; and (b) students had a greater number of longer-term goals with a mean of 53.6 versus 36.5 for the blue-collar workers, t = 4.21, p < .001.

Discussion and Conclusions

This study demonstrates that people are aware of and can describe personal goals and satisfactions that guide their behavior. Moreover, this initial development of the AIMS Chart appears to indicate that important attributes of a personal goal hierarchy can be measured and compared with other aspects of personal functioning. The coder reliability was encouraging for such a difficult task and a new instrument. The more pattern -oriented level of analysis (Table 3) was the most, reliable and perhaps most important. For example, the reliability coefficient for total goals on the chart was r = .87 (p < .001), indicating that coders had a high level of agreement on the overall extent of subjects' goals. Most of the domain categories and levels of generality were also highly reliable, implying that although coders disagreed on the coding of some individual goals, they depicted similar hierarchy patterns. To illustrate, consider the goal "be with grandchildren" in Figure 2, Box 2C. A separate coder might consider this a B-level goal in the continuum of levels while still depicting the same overall picture-that is, with family life emphasized and having a balance of short- and long-term goals.

Other cases in the study produced patterns differing in variables such as number of goals and distribution among domains and levels. The possibility thus exists for using variables such as these derived from the charts to construct typologies of goal hierarchies. It may be, for example, that some people have many short-term goals for behavior (immediate satisfactions) and little thought for long-range objectives, whereas other people have a number of vague longterm goals and little involvement with subgoals with which to reach them. Still others may have a balanced distribution and a high level of organization in their goal hierarchy. These differences could be predictive of other variables such as life satisfaction. Continued research with the instrument might clarify these relationships.

The validity tests employed in this study also support the future use of the instrument. The hypothesized group differences were verified, thus indicating that the method can detect distinguishing attributes of a goal hierarchy. Certain chart variables were found to be correlated in expected ways with other variables related or unrelated to goal -directedness. Finally, the most interesting validity test and the most relevant to applied uses for the instrument was strongly supportive. That is, subjects who viewed their charts considered them to be accurate. Certain comments were also suggestive of future research developments. For example, one man emphasized his negative feelings about work, and another man stressed how important his family was to him. In future work, it may be useful to involve subjects in rating the goals on their charts with respect to attributes such as importance, valence, and expectations about achieving goals. Another 'study, for instance, found that psychiatric patients who placed high importance on goals preselected by the researcher but had little hope of achieving them were more prone to depression (Erikson, Post, & Paige, 1975). In this study, one blue-collar worker had one longterm goal that he had dreamed of seeking most of his adult life (i.e., it had high valence for him), but it was so drastically different from his general life pattern that it never seemed feasible to achieve it. To find this conflict between high desirability and low apparent feasibility is not uncommon in career counseling. In an adult approaching retirement, high valence goals that have been dormant because they seemed impossible may provide a potential source of new meaning in a retiree's life. Variables such as these would add to the information available from the AIMS instrument.

In viewing the charts, a few subjects also described important life transitions that had taken place since the initial interview (e.g., marriage, graduation), They noted that these changes would be reflected in the content and structure of a new chart. This implies the potential use of the procedure for longitudinal study of individual development in terms of an evolving system of personal goals. Particularly at a time of transition, the chart could be a useful measure of a person's shifting organization of activities and priorities. For certain therapeutic interventions as well, this instrument might measure change processes related to goal-directedness.

Although the focus of the present study was instrument development rather than counseling, there were indications of potential applicability throughout. In one case, a man about to retire faced an unhappy home life and had few leisure interests. His chart made it. quite apparent that the loss of work interests would leave him with a very empty goal hierarchy. This pattern, clearly, has implications for retirement. counseling. A study by Reichard, Livson, & Peterson (1968) has, in fact, found that people who have nonwork interest,-, that are developed before retiring have a better adjustment to a retirement lifestyle. Another man with a marital relationship he anticipated might become unpleasant if he and his wife spent too much time together after retirement was already planning ways of pursuing things of interest to him that would keep him "out of the house" much of the time after he retired. People can anticipate both positive and negative potential consequences and seek to maximize the one and to minimize the other.

Using the AIMS Chart to examine an individual's full set of complex (and perhaps contradictory) goals, vague ideals and dreams, and actual behavior patterns would be of use in academic or career counseling. For example, the subject in Figure I talked at length about short-term activities and goals that were important to him. These goals were almost exclusively social and leisure, involving many parties, sports, and entertainment almost every night. When he then talked about his long-range goal of being successful in business, he was struck by the incongruence of his daily behavior. This discrepancy became apparent to him by the process of the AIMS interview and then even more graphically when he viewed the chart.

The potential for use of this instrument in counseling thus appears to be strong. The interviews themselves were said to be enlightening and beneficial by a number of subjects. Comments were made that the process of verbalizing one's goals and related activities was helpful for clarifying direction and raising motivation. Some of the subjects who viewed the coded charts had similar comments and remarked that the interview experience had influenced their behavior in positive ways. One student reported fairly dramatic changes in the way he had reorganized his activity priorities to achieve his objectives more successfully. Another subject realized during the interview that he had some previously unarticulated goals in his family life domain. By clarifying what they were, he was able to think of ways to improve his family life. These consequences of the data collection procedures were not intended but also not surprising. Within the systems theory framework of this research program, the directive function is very important to effective living. A procedure that heightens consciousness about one's goals can be expected to enhance the attainment of such personally valued outcomes.

This article sought to bring attention to the subject of personal goals and self-direction as a relevant concern in counseling. The interview study presented appears to indicate that personal goal hierarchies can be reliably and validly assessed. Qualitative data from the project suggest research directions for studying self-direction. The method also appears to have promise for counseling applications. Follow-up research is being directed towards developing an alternate, self-adminstered form of the measurement procedure that would be less costly but equally valid. The coding of lengthy interviews requires both intensive and extensive use of staff time. It is hoped that advances in the measurement of personal goals will assist counselors in helping their clients achieve their aims in life.

Reference Notes

1. Ford, D. H. The organization and development of human behavior. Book in preparation. Pennsylvania State University, 1983.

2. Brim,O. The sense of personal control over one's life.  Invited address to Divisions 7 and 8, at the 82nd Annual Convention of the American Psychological Association, New Orleans, September 1974.

3. Ford, M. "Youth Goal-Setting Scale," derived from Vondracek, F. W., Urban, H. B., and Parsonage, W. H. Project CARES Final Report,  1973-1974 (Center for Youth Studies and Social Policy, Report No. 7). University Park, Penn.: Pennsylvania State University, 1974.

4. MacPhillany, D., & Lewinsohn, P. M. The Pleasant Events Schedule. Eugene: University of Oregon, 1971. (Mimeo)

References

Allport, G. W. Pattern and growth in personality. New York: Holt, Rinehart& Winston, 1961.

Andrews, F., & Withey, S. Social indicators of  wellbeing New York: Plenum Press, 1976.

Atchley, R. C. Adjustment to loss of job at retirement.  International Journal of Aging and Human Development, 1975, 6, 17-27.

Bandura, A. Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review. 1977, 84,191-215.

Bandura, A. L., & Simon, K. M. The role of proximal

intentions in self-regulation of refractory behavior. Cognitive Therapy and Research, 1977, 1, 177-19:1.

Boden, M. A. Purposive explanation in psychology. Cambridge, Mass.: Harvard University Press, 1972.

Bruner,J.S. Processes of cognitive growth: Infancy. Worcester, Mass.: Clark University Press, 1968.

Buhler,C. Some empirical approaches to the study of life's basic tendencies. In F. T. Severin (Ed.), Humanistic viewpoints in psychology. New York: McGraw-Hill, 1965.

Buhler, C. The general structure of the human life cycle. In C. Buhler & F. Masserik (Eds.), The course of  human life: A study of goals in the humanistic perspective. New York: Springer, 1968.

 Carver, C. S., & Scheier, M. F. Attention and selfregulation: A control-theory approach to human behavior. New  York: Springer-Verlag, 1981.

 Cohen, J. A coefficient of agreement for nominal scales.  Educational and Psychological Measurement, 1960, 20,37-46.

 de Charms, R. Personal causation: The internal affective determinants of behavior. New York: Academic Press, 1968.

Deci, E. L. The psychology of self-determination. Lexington, Mass. Lexington Books, D. C. Beath & Co., 1980.

Erickson, R. C., Post, R. D., & Paige, A. B. Hope as a psychiatric variable. Journal of Clinical Psychology, 1975,32,324-330.

Ford, D., & Urban, H. Systems of psychotherapy. New York: Wiley, 1963.

Gardner, D. C., & Gardner, P. L. Goal-setting and learning in the high school resource room. Adolescence, 1978, 13, 439-493.

Hall, C. S., & Lindzey, G. Theories of personality. New York: Wiley, 1970.

Hilgard, E. R. Divided consciousness: Multiple controls inhuman thought and action. New York: Wiley, 1977.

Jantsch, E. The self- organizing universe. New York: Pergamon Press, 1980.

Kanfer, F. H. Self-management: Strategies and tactics. In A. P. Goldstein & Kanfer (Eds.), Maximizing treatment gains. New York: Academic Press, 1979.

Kiresuk, T., & Sherman, R. Goal attainment scaling: A general method for evaluating community mental health programs. Community  Mental Health Journal, 1968, 4, 443-453.

Klinger, E. Meaning and void: Inner experience and the incentives in people's lives. Minneapolis: University of Minnesota Press, 1977.

Klinger, E., Barta, S. G., & Maxeiner, M. E. Current concerns: Assessing therapeutically relevant motivation. In P. C. Kendall & S. D. Hollon (Eds.), Cognitive-behavioral interventions: Assessment. New York: Academic Press, 1980.

 Kolb, D. A., Winter, S. K., & Berlew, D. E. Self-directed change: Two studies. Journal of Applied Behavior Science, 1968, 4, 453-473.

Landis, J. R., & Koch. G. G. The measurement of observer agreement for categorical data. Biometrics, 1977,33,159-174.

Lefcourt, H. M. Locus of control: Current trends in theory and research. Hillsdale, N.J.: Erlbaum, 1976.

 Levinson, D. J., Darrow, C. N., Klein, E. B., Levinson, M. H., & McKee, B. The seasons of a man's life. New York: Knopf, 1978.

Locke, E. A., Shaw, K. N., Saari, L. M., & Latham, G. P. Goal setting and task performance: 1969-1980. Psychological Bulletin, 1981, 90, 125-152.

Locke, E. A., Cartledge, N., & Knerr, C. S. Studies of the relationship between satisfaction, goal setting, and performance. Organizational Behavior and Human Performance, 1970, 5, 35-158.

 Lowenthal, M. F. Intentionality: Toward a framework for the study of adaptation in adulthood. Aging and Human Development, 1971, 2, 79-95.

Mahoney, M. J., & Thoresen, C. E. Self-control: Power to the person. Monterey, Calif.: Brooks/ Cole, 1974.

Maturana, H. R. The organization of the living: A theory of the living organization. International Journal of Man-Machine Studies, 1975, 7, 313-332.

 Miller, G., Galanter, E., & Pribram, K. Plans and the structure of behavior. New York: Holt, 1960.

Miller, J. G. Living systems. New York. McGrawHill, 1978.

Neugarten, B. L., Havighurst, R. J., & Tobin S. S. The measurement of life satisfaction. Journal of Gerontology, 1961, 16, 134-143.

 Powers, W. T. Feedback: Beyond behaviorism. Science, 1973,179, 351-356.

Powers, W. T. Quantitative analysis of purposive systems: Some spadework at the foundations of scientific psychology. Psychological  Review, 1978, 85,417-435.

Reichard, S., Livson, F., & Peterson, P. G. Adjustment to retirement. In B. L. Neugarten (Ed.), Middle age and aging. Chicago: University of Chicago Press, 1968.

 Rogers, C. R. Client-centered therapy. Boston, Mass.: Houghton Mifflin Co., 1951.

 Rokeach, M. Beliefs, attitudes, and values. San Francisco: Jossey-Bass, 1968.  0

Seligman, M. E. P. Helplessness: On  depression, development, and death. San Francisco: Freeman,  1975.

Strong, E. K., Jr. Vocational interests of men and women.  Stanford, Calif.: Stanford University Press, 1943.

Tolman, E. C. Purposive behavior in animals and men.  New York: Appleton -Century -Crofts, 1932.

von Bertalanffy, L. General system theory. New York: Braziller, 1968.

Weigel, R. E., & Uhlemann, M. R. Developing individualized behavior change goals with clients. Journal of Contemporary Psychotherapy, 1975, 7, 91-95,

 Received October 10, 1982

Revision received April 22, 1983