Psychology Journal Article

ORGA.NIZATIO~NAL BEItAVIOR A1N'D ttU~IA~2~ PERFORMANCE

4, 142-175 (1969)

An Empirical Test of a New Theory of Human Needs CLAYTON P. ALDE~FER1

Department o] Administrative Sciences, Yale University This study was concerned with developing and testing an alternative to Maslow's theory and to a simple frustration hypothesis for the problem of relating need-satisfaction to strength of desires. The alternative theory is based on a three-fold conceptualization of human needs: existence, relatedness, and growth (E.R.G.), It does not assume lower-level satisfaction as a prerequisite for the emergence of higher-order needs. It does include propositions relating the impact of higher-order frustration to the strength of lower-order needs. Empirical tests of differential predictions among Maslow's theory, the simple frustration hypothesis, and E.R.G. theory were conducted by a questionnaire study with 110 employees at several job levels from a bank. The results tended to support E.R.G. theory more than Maslow's theory or the simple frustration hypothesis. Maslow's (1943, 1954) hierarchical theory of motivation has had a m a j o r influence on the thinking and research of m a n y writers in the field of organizational behavior. I n Argyris' (19'64) work on the conflict between the individual and the organization, the concept of self-actualization played a central role. McGregor's (1960) formulation of the now famous managerial styles, theory X and theory Y, relied heavily on the idea t h a t h u m a n motives were arranged in a hierarchy of prepotency. Porter (1962, 1963) based his national survey of managerial job attitudes on Maslow's conceptualizations. Beer (1966) also utilized M a s low's concepts for his empirical work on the relationship among employee needs, leadership, and motivation. I n addition, several writers have proposed modifications of Maslow's (1943, 1954) original framework. One such modification was suggested (though not explicitly stated as such) by Maslow (1962) himself when he wrote about deficiency and growth motivations. Barnes (1960) pro1The writer would like to thank Thomas M. Lodhal, Martin G. Evans, Benjamin Schneider, J. l~ichard l=[ackman, Douglas T. Hall, and Chris Argyrls for their thoughtful and helpful comments on an earlier version of this paper and the Graduate School of Business and Public Administration at Cornell for financial support for the study. Appreciation is also expressed to the management, staff, and employees from the organization where the study was conducted for their very high degree of cooperation. 142

A N E W THEORY OF H U M A N

NEEDS

143

posed a two-step hierarchy consisting of physiological needs at the base and a higher level made up of self-esteem, esteem of others, and belongingness. Harrison (1966) also offered a modification consisting of two levels, similar to that of Barnes. In his model, physiological-economic needs were at the base. Upon satisfaction of these needs, then a higher level of social or ego needs would be sought. Clark (1960) produced a review of the literature organized around Maslow's original hierarchy. In that paper he pointed out the need for a direct empirical test of Maslow's theory. However, until the recent work by Hall and Nougaim (1968), that theory has not been tested empirically in literature known to this writer. Hall and Nougaim (1968) designed a longitudinal study to test key propositions in the Maslow theory. Using five annual interviews from each of 49 managers in A.T. & T., they developed operational definitions to test Maslow's predictions by both static and change analyses. They report some difficulty developing operational definitions for the Maslow system (1968, pp. 19, 30). Their results provide almost no support for the Maslow theory. Indeed, one aspect of their data which they do not discuss at length was the ~endency for the satisfaction of a need to correlate with the intensity of the need itself. This finding not only does not support Maslow's (1943, p. 393) dictum, % satisfied need is not a motivator," it seeems to contradict it. However, the finding might also be reviewed with questions about the adequacy of operational definitions when there were recognized difficulties with that part of the study. Another interpretation would be that the finding is consistent with a twostep hierarchy, since there does appear to be somewhat more of, a tendency for higher order need satisfaction to correlate with need intensity than for lower order need satisfaction to correlate with need intensity.2 The basic problem to which Maslow's theory addresses itself might be stated as, "How does need satisfaction affect the strength of desires?" A very simple solution to this problem might be to say that. for any need, frustration results in increased desire, while satisfaction results in decreased desire2 ,~ The writer is indebted to M. G. Evans for calling attention to this point. 3For the purposes of the discussion in this paper frustration and satisfaction are used to refer to opposite ends of the same continuum. For other purposes, one might wish to distinguishbetween the two concepts more fully. Rosenzweig (1944, p. 380) has approached the same conceptual problem from the point of view of frustration theory. He made a distinction between primary and secondary frustration. In his view "primary frustration involves the sheer existence of an active need...Secondary frustration more strictly embraces the definition given above, emphasis being placed upon supervenient obstacles or obstructions in the path of the active need." In the present field study, where the

144

CLAYTON P. ALDERFER

The simple frustration hypothesis requires no set of relationships among different types of needs, as does Maslow's hierarchy. Neither does it require a theory which takes a stand by specifying categories of needs, as Maslow does when he suggests the five need categories of physiological, safety, love, esteem, and self-actualization. While Maslow did not say so explicitly, his theory might be viewed as an alternative to the simpler notion that need-satisfaction reduces desire. By specifying the core problem which Maslow's theory addresses, one is able to see that there are many possible frameworks to deal with the same question. E.R.G. theory was developed by the writer because of his impression that a more adequate understanding could be achieved for the kinds of phenomena to which Maslow's theory was addressed. An early test of the theory suggested that, while all of its predictions were not confirmed, it did present a viable alternative to Maslow's point of view (Al.derfer, 1966). At the time of this initial test of E.R.G. theory, there were no direct empirical tests of Maslow's work known to the writer. However, in the meantime, Hall and Nougaim (1968) have reported their results which tended not to support Maslow's views. Conant (1947, p. 36) has commented, during his review of the history of physical science, that contradictory facts are generally not a sufficient reason for the abandonment of a particular theory. A new theory needs to be present in addition to facts contrary to the old one. It is probably also a fair statement to say that no theory is ever final or fully comprehensive. Certainly that is true of E.R.G. theory, based upon the initial results of empirical work testing it. Consequently, one of the major tasks of science becomes that of comparing theories to find which one does a better (or best, if there are more than two) job of accounting for the empirical data in its domain at any given point in time. The objectives of this paper are to present E.R.G. theory, to show its similarities and differences with Maslow's theory and the simple frustration hypothesis, and to test predictions made by all three frameworks. In contrasting E.R.G. theory with Maslow's theory, there are both differences in the need categories and differences in the ways that need satisfaction is seen relating to desires. Differences in need catemeasure of frustration was subjective, no a t t e m p t was made to distinguish between the two operationally. However, were an experimental manipulation undertaken, then different procedures might be used to induce frustration and satisfaction. ~E.R.G. theory is phrased in terms of relationships between satisfaction and desire. Desire should be read as synonymous with other terms such as waat, preference, intensity, need strength, and motive strength. The term "desire" was chosen because it stays most closely to the operational definitions. (See METHODS section.) Satisfaction m a y also be read as synonymous with the term fulfillment.

A NEW

THEORY

OF HUMAN

NEEDS

145

gories will be dealt with first, and differences in major propositions next. Then the specific hypotheses tested in this study will be presented, followed by the empirical resulN. DIFFERENCES IN NEED CATEGORIES Maslow's theory deals with five sets of needs: physiological, safety, love, esteem, and self-actualization. At a purely conceptual level it is not clear where safety needs depart from physiological needs, on the one hand, and love needs on the other. In his discussion of the safety concept, Maslow mentions issues such as physical illness, pain, and assault as one set of issues and parental outbursts of rage, name calling, and speaking harshly as another. It seems that physical threats might be usefully put together with the other needs which Maslow terms physiological, and the safety issues involving interaction with other people included with love needs. A similar point might be made regarding esteem needs. In this ease the overlap seems to be between love needs and self-actualization needs. Maslow indicates that his concept of esteem includes the regard a person receives from others. In this sense, it is very similar to love needs. However, he goes on to refer to that aspect of esteem which depends on the internal cues which a person provides himself from real capacity, achievement, and independence. This element in esteem needs seems to be more like self-actualization in which a person is able to be what he is and do what he is capable of doing. E.R.G. theory assumes that a human being has three core needs that he strives to meet. They include obtaining his material existence needs, maintaining his interpersonal relatedness with significant other people, and seeking opportunities for his unique personal development and growth, These needs provide the basic elements in motivation. Often people express their wants in the form of complex goals which may include mixtures of the basic needs. One such "compound" need would be for a promotion, where as a result of the promotion the person would obtain more material rewards in the form of pay, a different constellation of interpersonal relationships, and new opportunities to develop and use his talents. Existence needs include all the various forms of material and physiological desires. Hunger and thirst represent deficiencies in existence needs. Pay, fringe benefits, and physieM working conditions are other types of existence needs. One of the basic characteristics of existence needs is that they can be divided among people in such a way that one person's gain is another's loss when resources are limited. If two people are hungry, for example, the food eaten by one is not available to the

146

CLAYTON P. ALDERFER

other. When a salary decision is made that provides one person or group of people with more pay, it eliminates the possibility of some other person or group getting extra money. This property of existence needs frequently means that a person's (or group's) satisfaction, beyond a bare minimum depends upon the comparison of what he gets with what others get in the same situation. However, this comparison is not "interpersonal" in the sense of necessitating comparison with known significant others. The interpersonal aspect of equity is not an issue for existence needs. The comparison process for material goods is simply among piles of goods, without necessarily attaching the added dimension of knowing who the others are who would obtain smaller or larger shares. It turns out in our society that people have learned to state such comparisons in interpersonal terms. Consequently in developing operational definitions, some existence need comparisons were stated as interpersonal comparisons. However, this reflects a realistic limitation of the measures, reflecting certain aspects of our culture for which the theory, not being a learning theory, does not account. Relatedness needs include all the needs which involve relationships with significant other people. Family members are usually significant others, as are superiors, coworkers, subordinates, friends, and enemies. One of the basic characteristics of relatedness needs is that their satisfaction depends on a process of sharing or mutuality. People are assumed to satisfy relatedness needs by mutually sharing their thoughts and feelings. This process markedly distinguishes relatedness needs from existence needs because the process of satisfaction for existence needs prohibits mutuality. The exchange of acceptance, confirmation, understanding, and influence are elements of the relatedness process. It is not necessary for the formal power between two people to be equal, or nearly so, for relatedness need satisfaction to be possible. The essential conditions involve the willingness of both (or all) persons to share their thoughts and feelings as fully as possible while trying to enable the other(s) to do the same thing. Argyris (1962) has termed this quality of a relationship authentic. Furthermore, the outcome in satisfying relatedness needs need not always be a positive affectual state for both or either person. The exchange or expression of anger and hostility is a very important par~ of meaningful interpersonal relationships, lust as is the expression of warmth and closeness. Thus, the opposite of relatedness satisfaction is not necessarily anger, but it is a sense of distance or lack of connectedness. Growth needs include all the needs which involve a person making creative or productive effects on himself and the environment. Satisfac-

A NEW THEORY OF HUMAN NEEDS

147

tion of growth needs comes from a person engaging problems which call upon him to utilize his capacities fully and may include requiring him to develop additional capacities. A person experiences a greater sense of wholeness and fullness as a human being by satisfying growth needs. Thus satisfaction of growth needs depends on a person finding the opportunities to be what he is most fully and to become what he can. The business of categorizing or developing lists of human needs and motives is not a new activity. Henry Murray's (1938) list is reasonably well known, and it is longer than Maslow's set. Langer (1937) and Sehein (1965) have also suggested three category sets which bear considerable similarity to the one proposed in E.R.G. theory. In addition to definitional differences of varying degrees, these concepts differ in purpose from those utilized in ~laslow's or E.R.G. theory. Classification is not the same as explanation and prediction, but some categorization (or variable definition) is necessary in order to formulate a theory. This threefold categorization of human needs represents the first way E.R.G. theory departs from Maslow's scheme. It is one way to deal with the problems posed by the overlapping nature of his safety and esteem categories. In this new framework, those aspects of safety needs which deal with physical or material desires belong to the existence category while those aspects which have to do with interpersonal processes fit the relatedness ca£egory. The same kind of point applies to esteem needs. Those aspects of esteem which depend upon reactions from others fit the relatedness category, while those which represent autonomous self-fulfilling activity belong to growth needs. The different categorization allows one to refer to the new framework as E. R. G. theory. By itself the change in the way needs are characterized implies different predictions about how satisfaction is related to strength of desire. In the case of relatedness needs, the change in categoriza£ion speaks to a problem with Maslow's theory that he himself noted (1943, p. 308) : "There are some people in whom, for instance, self-esteem seems to be more important than love. This most common reversal in the hierarchy is usually due to the development of the notion that the person who is most likely to be loved is a strong or powerful p e r s o n . . . " By recognizing that part of self-esteem which depends on regard from others as part of relatedness needs, the issue of whether love desires precede or follow desires for the esteem of others vanishes. MAJOR PROPOSITIONS AND EXPLANATORY CONCEPTS IN E.R.G. THEORY Seven major propositions in E.R.G. theory provide a basis from which empirically testable hypotheses can be logically derived. The form of

148

CLAYTON P. ALDERFER

this derivation is as follows. If A is an operational indicator of an E., R., or G. satisfaction and B is an operational indicator of an E., R., or G desire, then A should show an empirically verifiable relationship to B in such a way as predicted by one of the E.R.G. propositions. If empirical results provide support for the A to B relationship, then one can have more confidence in the theory. If the empirical results do not provide support for the A to B relationship, one would have less confidence in the theory. The structure of this theory is such that the results can provide support for some propositions while not for others. The major propositions in E. R. G. theory are as follows: P1. The less existence needs are satisfied, the more they will be desired. P2. The less relatedness needs are satisfied, the more existence needs will be desired. P3. The more existence needs are satisfied, the more relatedness needs will be desired. P4. The less relatedness needs are satisfied, the more they will be desired. P5. The less growth needs are satisfied, the more relatedness needs will be desired. P6. The more relatedness needs are satisfied, the more growth needs will be desired. PT. The more growth needs are satisfied, the more they will be desired. These propositions indicate that any desire can have several types of satisfaction (including some outside its particular category) affecting its strength. Any satisfaction also affects more than one type of desire (including some outside its particular category). This multiple determination property is shown in Fig. 1, which gives a summary of the propositions in diagrammatic form. An additional aspect of the theory, however, concerns providing explanatory concepts or mechanisms which lie behind the various propositions. These explanatory concepts are intended to help answer the "why" questions for the various propositions. As such, they add richness to the theoretical framework. In a loose sense they may be seen as analogous to axioms which provide a basis from which the main propositions can be derived. In proposition 1, which deals with the impact of the lack of existence

149

A NEW THEORY OF HUMAN NEEDS

FI~. 1. Major propositions in E.R.G. theory? Need Frus tration ~

(lack of satisfaction)

Need ,gatis]action ~

Strength o] Desires E

(pl) 3 e

(P4)

~R

~

G

~

(P7)

rs

gs

Numbers in the diagram refer to the proposition numbers in the text. bSee footnote 3 of the text. c Each of the arrows i~ this diagram has its "dual" which could also be drawn. Thus, it would not be necessary to have both a satisfaction and a frustration column. However, for purposes of clarity in presentation and to emphasize the theoretical reasoning behind the propisitions, this particular format was employed. need satisfaction on existence desires, there is an assumption of the i n t e r c h a n g e c & i l i t y of various existence needs. P a y and fringe benefits

are obvious examples where an organization actually makes choices about how to compensate its employees. The investigations of Nealey (1963.) and others have been based on this assumption. Where the interehangeability assumption m a y not be as obvious is with such things as physical working conditions or physical demands of the job. However, most job evaluation systems contain provisions where an employee is paid more because he has dirty, hazardous, or physically taxing duties in his iob (Strauss and Sayles, 1967). Moreover, Jacques (19'61) has formulated a view of p a y m e n t based upon the time span of discretion. According to his view, there is a correspondence between equitable p a y m e n t and discretion. H e reported t h a t there is evidence to confirm the view t h a t the length of discretionary time span corresponds with the financial loss which would be caused by substandard discretion (1961, p. 84). There is a point of view voiced by Dunnette (19'67) and by Lawler and Porter (1963) t h a t pay can stimulate and satisfy needs other than for material goods. Lawler and Porter pointed to correlations between

150

CLAYTON P. ALDERFER

managers' pay and esteem and autonomy-need satisfactions. Reasoning from expectancy theory, Dunnette suggested that pay can be used as a reward for those who seek power, status, and achievement. Neither the concepts nor the propositions of E.R.G. theory would dispute this view. Using pay as an example of an existence need does not contradict this view. What E.R.G. theory attempts to do that neither expectancy theory nor Lawler and Porter's (19'63) empirical results do is explain why pay (and other existence needs) are sought to satisfy other than material needs. The discussion of proposition 2 below deals directly with this issue. The structure of proposition 4 is similar to that of proposition 1 when it states that lack of satisfaction of relatedness needs leads to higher relatedness desires. The explanatory mechanism in this proposition is transyerabiIity of significant others. Persons who lack a basic sense of connectedness and sharing in their emotional lives with significant others will seek to obtain that need satisfaction. If they are unable to obtain the satisfaction with the original person where the satisfaction is missing, then they will tend to transfer the desire to others. Some of the earliest clues about the operation of this process are found in Freud's (1963, p. 106) work on transference. "Expectant libidinal impulses will inevitably be roused, in anyone whose need for love is not being satisfactorily gratified in reality, by each new person coming upon the scene, and it is more than probable that both parts of the libido, the conscious and the unconscious, will participate in this attitude." Aspects of both propositions 1 and 4 subsume what has been termed the simple frustration hypothesis. In this respect the two propositions are similar. A specific need in either category that is not being satisfied will result in that specific desire being heightened. Thus, the propositions are also consistent with the conclusion reached from a literature review by Lawson and Marx (19:58) that frustration often increases motivation. Propositions 2 and 5 follow from a concept which played an important part in Lewinian field theory. From that framework, investigators attempted to create regression by frustration (Barker, Dembo, and Lewin, 1943). By regression they meant a more primitive, less mature way of behaving, not necessarily behavior that had been produced earlier in life (Lewin, 19'51). The concepts of existence, relatedness, and growth needs were presented as separate and distinct categories. There was nothing about the definitions of the needs that implied an ordering among them. One of the ways in which the needs can be viewed as being on a continuum is in terms of their concreteness (Harvey, Hunt, and Schroder, 1961). Exist-

A NEW

T H E O R Y OF H U M A N

NEEDS

151

ence needs are the most concrete. Their presence or absence is the easiest for the person to verify due to the fact that their objectives can be reduced to material substances or states. Relatedness needs are less concrete than existence needs. Their presence or absence depends on the state of relationships between two or more people. To verify the state of relatedness needs depends on the consensual validation of the people involved in the relationship. Finally, growth needs are the least concrete. Ultimately their specific objectives depend on the uniqueness of ea.eh person..At the most precise level, the actual state of growth of a person can be known only to the person, and it can be known to him only when he is not deluding himself. The continuum from more to less concreteness is also a. continuum from more to less verifiability and from less to more potential uncertainty for the person. The sense in which frustration regression is employed in E.R.G. theory concerns the tendency of persons to desire more concrete ends as a consequence of being unable to obtain more differentiated, less concrete ends. Thus a person is thought to desire existence needs when relatedness needs are not satisfied because he is using them as an easier, more concrete way of establishing his connectedness with other people. He seeks relatedness needs when he is unsatisfied with his growth because he is searching for opportunities for more clarity and support in the quest to stretch, develop, and expand himself. Thus propositions 2 and 5 are based on the idea that when a person is not satisfied in attaining less concrete, more uncertain ends, he "regresses" to needs which are somewhat more concrete and less uncertain as to their attainment. It is in this sense that a person may use the size of his pay check as an indicator of the esteem in which he is held by his boss, colleagues, or organization. According to E.R.G. theory one would expect him to do this less, the more open, trusting, and mutually respectful his relationship was with those significant others. Given the increasing amount of data showing the lack of relatedness-need satisfaction in organizational life, it is not at all surprising to find that persons rely on pay to assess the esteem by which they are held. If propositions 2 and 5 follow from a frustration-regression mechanism, the explanation behind propositions 3 and 6 might be termed satisfaction progression. These propositions are intended to be in the same spirit as .1VIaslow's original hierarchy, except that the impact of relatedness satisfaction on growth desires does not presume satisfaction of existence needs. Satisfaction-progression reasoning is based on the premise that as a person fulfills the more concrete aspects of his desires, more of his energy becomes available to deal with the less concrete, more personal, and more uncertain aspects of living. As he is able to fulfill existence

152

CLAYTON P. ALDERFER

needs he needs to spend less of his energy in search of material things. Consequently he is more able to attend to interpersonal issues. Realistically, he no longer needs to fear other human beings as competitors for scarce material resources. As he is able to find fulfillment of his relatedness needs, he no longer lacks authentic social support. He has found relationships where it is possible for him to share the complex and subjective parts of his emotional life. Again he is more free to have his energy be turned to the most personal and unique parts of living, being and becoming what he can most fully. Satisfaction of relatedness needs has relieved him of the anxiety of social uncertainty. More of his energy, therefore, is available for expanding and utilizing the fullness of his person. Proposition 7 states that the more growth needs are satisfied the more they will be desired. This proposition follows from an expansion of the concept of setting aspiration levels (Lewin, Dembo, Festinger, and Sears, 1944). By their very nature, growth needs are intrinsically satisfying. The more a person grows, the more he wants to grow; the less he grows the less he desires to grow. These statements translate readily into the classical aspiration mechanism if one is willing to talk about success and failure a part of the growth process. Classical aspiration level theory states that persons tend to raise their aspiration level when they reach self-determined goals and lower aspiration levels when they fail to attain their goals. Certainly success and failure are part of the growth process as it is conceived by E.R.G. theory. But growth as here conceived includes more than success and failure in attaining predetermined goals. As Zaleznik (1968) has shown quite persuasively, failures and disappointments may often provide very rich opportunities for growth providing the person is able to face and cope with the failure. Thus growth satisfaction, in this theory, occurs when a person does and experiences things that tend to enhance, enrich, and enlarge the functioning of his personhood. This process can occur as a result of success or failure in attaining predetermined goals. The crucial criterion is when a person sees himself as learning from any experience and emerging as a fuller, more differentiated, more competent human being. It may seem paradoxical that a person should fee] more competent following failure. However, if one assumes that some failure is part of living, then coping with it in a way that enlarges one's personhood can enhance a person's sense of competence. One expanded aspiration mechanism states that the more a person experiences himself as a full and differentiated human being, the more he will aspire to be a whole and rich person. Both success and failure can be growth satisfying experiences. The more a person finds added wholeness in his living, the more he will seek additional opportuni-

A NEW

T H E O R Y OF H U M A N

NEEDS

153

ties to grow both on the same dimensions where he has already grown and in new arenas in his life. DEFICIENCY AND GROWTH MOTIVATION: A "DERIVATION" In one of his later papers, Maslow (19'62) introduced two newer concepts of motivation, deficiency motivation and growth motivation. One way to view these concepts is as a way ~o collapse the need categories into two, rather than three or five categories. Hall and Nougaim (1968), for example, took this tact. However, another view is also possible. E.R.G. theory is intended to be a dynamic theory even though the major propositions as presented here are stated in more or less static terms. With a dynamic theory, one can seek to investigate equilibrium points. The mathematical derivation of such equilibrium points is beyond this paper's scope and the writer's competence. However, it is possible to present a verbal argument leading to an hypothesis of two equilibrium points, one for deficiency motivation and one for growth motivation. For deficiency motivation, the major cycling of desires is between existence and relatedness needs. A person whose motives operate mainly in this cycle seeks existence needs and attains suftieient satisfaction that his existence needs decrease in desire and he turns to relatedness needs as his prime desires. However, ha is unable to attain satisfaction of these needs. While these needs remain active, but unsatisfied he tends to turn back to existence needs, seeking material gratification. As before, he is able to attain material satisfaction and turns once again to relatedness needs, which remain unsatisfied and become increasingly difficult to satisfy because of the intensity of demand. Thus he is forced back to material demands, and on the cycle goes. For growth motivation, the major cycling of desires is between relatedness and growth needs. A person, whose motives operate mainly in this cycle, seeks growth needs but finds he is not always able to satisfy these needs. As a consequence he also seeks relatedness needs which he is able to attain. As a result of the social support, he is provided with more energy for grouch, which he then seeks. He is not always able to attain growth satisfaction and "regresses" periodically to relatedness needs, which he is able to attain, he turns again to growth, and so the cycle continues. The reasoning behind the key concepts and propositions in E.R.G. theory is relevant for the empirical test of the theory presented in the latter section of this paper. For an investigator who may wish to conduct his own empirical tests of the theory, the additional reasoning may be important for developing operational definitions. It also shows that E.R.G. theory consists of more than some conceptual definitions and

154

CLAYTON r . ALDERFER

several statements, which may appear to be no more than a series of testable empirical generalizations. There are explanatory mechanisms behind the propositions, even though this part of the theory has more surplus meaning and is generally looser logically than the propositions and measures alone. COMPARISON OF E.R.G. AND MASLOW PROPOSITIONS E.R.G. theory retains the notion of a need hierarchy without requiring it to be strictly ordered. Maslow says that for a chronically hungry man "life tends to be defined in terms of eating. Anything else will be defined as unimportant" (1943, p. 374). E.R.G. theory would say that a chronically hungry man can recognize whether he feels connected to primary groups and to society and whether he is able to engage in activities which enable him to use his skills and talents. E.R.G. propositions 3 and 6 have an emphasis similar to Maslow's hierarchy, but they do not require lower-level gratification as an additional condition as Maslow's theory would. Furthermore, proposition 7 also bears on a criticism of Maslow's theory that he himself raised (1943, p. 386) : "There are other, apparently innately creative people in whom the drive to creativeness seems to be more important than any other counter-determinant. Their creativeness might appear not as self-actualization released by basic satisfaction, but in spite of lack of basic satisfaction." Without a strictly ordered hierarchy, the case which Maslow raises as a limitation to his theory need not exist. When a hierarchy is no longer viewed as strictly ordered, it does represent a change of a basic assumption. Maslow made the point that "needs cease to play an active determining or organizing role as soon as they are gratified" (1943, p. 393). E.R.G. has a different point of view. One way in which a satisfied need can remain a motivator is if it is activated through serving as a substitute for some other need which itself is not being satisfied. In particular, E.R.G. theory contains two propositions to this effect. They involve the need hierarchy principle working in reverse; if a higher-order need is frustrated, the next-lower-order need is activated. For E.R.G. theory, this leads to propositions 2 and 5: The less relatedness needs are satisfied, the more existence needs will be desired. The less growth needs are satisfied, the more relatedness needs will be desired. These two propositions are not found in Maslow's work, and therefore represent another point of departure from his theory. In reviewing the Hall-Nougaim (1968) work, it was emphasized that

A NEW THEORY OF HUMAN NEEDS

t55

their data seemed to run contrary to the point of view that a satisfied need is not a motivator. E.R.G. theory also departs from that view for the reasons implied by the Hall-Nougaim (1968) study. The departure is shown in proposition 7 which states that growth needs are desired the more they are satisfied. In Maslow's (1943) original statement he did not discuss the consequences of satisfaction of self-actualization needs. However, in a later statement (1962, p. 31) he did say: "Growth is instead a continued, more or less steady upward or forward development. The more one gets, the more one wants, so that this kind of wanting is endless and can never be attained or satisfied." Thus E.R.G. theory departs from Maslow's earlier statement but is in esseatial agreement with his later position. Maslow's basic hypothesis i s tha~ a certain degree of satisfaction of lower-level needs is a prerequisite for t h e appearance of higher-order needs. Operating within the context of that hypothesis, threefold categorization implies some different predictions that the fivefold system. Since in E.R.G. theory, safety needs in part belong with existence needs and in par~ with relatedness needs, it was necessary to make temporary use of two new concepts to state the different predictions. Let these be termed safety-existence needs and safety-relatedness needs. Then, according to E.R.G. theory: The more safety-existence needs are satisfied, the m o r e relatedness needs will be desired. The more safety-relatedness needs are satisfied, the less relatedness needs will be desired. According to Maslow's theory, the prediction would be: If physiological needs are satisfied, then the more safety needs are satisfied, the more love (relatedness) needs will be desired. Thus, in the case of safety needs, the change in categorization by itself implies a different prediction. Analogously, in the case of esteem needs, one can see that in part they belong with relatedness needs and in part with growth needs. Again, it is necessary to make temporary use of two new concepts to explore different predictions. Let these be termed esteem-relatedness needs and esteemgrowth needs. Then, according to E.R.G. theory: The more esteem-relatedness needs are satisfied, the more growth needs will be desired. The more esteem-growth needs are satisfied, the more growth needs will be desired.

156

CLAYTON P. ALDERFER

According to Maslow's theory the prediction would be: If physiological, safety, and belongingness needs are satisfied, then the more esteem needs are satisfied, the more self-actualization (growth) needs will be desired. In the case of the relationship between esteem and growth needs the categorization change by itself does not imply substantially different predictions. But, in the case of the relationship between esteem and love needs the conceptual change does imply different predictions. According to E.R.G. theory: The more love (relatedness) needs are satisfied, the less esteem-relatedness needs will be desired. But according to Maslow's theory: If physiological and safety needs are satisfied, then the more love needs are satisfied, the more esteem needs will be desired. In summary, E.R,G. propositions differ from Maslow's propositions in three major ways. First, they include the hierarchical idea, but the hierarchy is not strictly ordered. Second, t h e y include taking account of higher need frustration on lower-level desires. Third, the different category system results in different predictions by itself. D I F F E R E N T I A L P R E D I C T I O N S T E S T E D IN THIS STUDY

There were two major areas where this study was designed to test differential predictions from Maslow's theory. First, there were the propositions that followed from E.R.G. theory having a single category for interpersonal issues rather than the overlap among safety, love, and esteem needs. Second, there were the propositions concerning movement down the hierarchy which were made by E.R.G. theory but not by Maslow's theory. Respect from co-workers and respect from superiors were the relatedness needs based on E.R.G. theory, while belongingness and status needs were the needs based on Maslow's theory. In order to test hypotheses about belongingness needs by Maslow's theory, one should show that physiological and safety needs are relatively satisfied. To test hypotheses about status needs by Maslow's theory, one should show that physiological, safety, and belongingness needs are reasonably well satisfied. Data bearing ou these assumptions will be presented below. Given these initial assumptions, then Maslow's theory would lead to the following hypotheses about interpersonal needs2 In the phrasing of these hypotheses, "belongingness" needs refer to the operations developed for Maslow's concept of love needs and "status" for Maslow's concept

A NEW THEORY OF HUMAN NEEDS

157

H1. The m o r e belongingness needs are satisfied, the m o r e status needs will be desired. H2. The less belongingness needs are satisfied, the m o r e they will be desired. status needs are satisfied, the m o r e they will be desired.

H 3 . T h e less

Proposition 4 in E.R.G. theory lead to an analogous set of hypotheses: H4. The less respect from co-workers is satisfied, the m o r e respect from superiors will be desired. H5. The less respect from superiors is satisfied, the m o r e respect from co-workers will be desired. H6. The less respect from co-workers is satisfied, the m o r e it will be desired. HT. The less respect from superiors is satisfied, the m o r e it will be desired. Hypotheses 1-3 from Maslow's theory and 4-7 from E.R.G. theory result from each theory's way of dealing with interpersonal needs. If E.R.G. theory does a better job of predicting than Maslow's theory, then there should be some evidence (though perhaps not strong evidence) tha~ proposition 4 from E.R.G. theory operating on Maslow type needs receives empirical support. One difference between the theories with regard to interpersonal needs is that E.R.G. theory has one category for interpersonal needs rather than several as Maslow's theory does. In addition, Maslow's interpersonal needs depend on a different conceptualization than E.R.G. needs; they do not specify need satisfaction with regard to significant others. As a result, if E.R.G. theory did a better job because of having a single category, one would expect suppo~% for "cross theory" predictions. But one would not expect this support to be strong because none of the Maslow interpersonal needs specify significant others, while the E.R.G. relatedness concept does. Hypotheses 8-15 follow from using combinations of E.R.G. and Maslow interpersonal needs in E.R.G. proposition 4. H8. The less belongingness needs are satisfied, the m o r e respect from peers will be desired. of esteem needs. These terms follow from the P o r t e r (1962) and Beer (1966) operational definitions of Maslow needs used in this study. Respect from coworkers and respect from superiors are operational indicators of relatedness needs in E.R.G. theory.

:158

CLAYTON P. ALDERFER H9. The less belongingness needs are satisfied, the m o r e respect from superiors will be desired. H10. The less status needs are satisfied, the m o r e respect from co-workers will be desired. H l l . The less status needs are satisfied, the more respect from superiors will be desired. tt12. The less respect from co-workers is satisfied, the m o r e belongingness needs will be desired. H13. The less respect from co-workers is satisfied, the m o r e status needs will be desired. H14. The less respect from superiors is satisfied, the m o r e belongingness needs will be desired. H15. The less respect from superiors is satisfied, the m o r e status needs will be desired.

E.R.G. propositions 2 and 5 deal with the impact of higher-order need frustration on lower-order desires. These propositions are not found at all in Maslow's theory. They lead to hypotheses 16-21. H16. The less growth needs are satisfied, the m o r e respect from co-workers will be desired. H17. The less growth needs are satisfied, the m o r e respect from superiors will be desired. H18. The less respect from co-workers is satisfied, the m o r e pay will be desired. ttl9. The less respect from co-workers is satisfied, the more fringe benefits will be desired. It20. The less respect from superiors is satisfied, the m o r e pay will be desired. H21. The less respect from superiors is satisfied, the m o r e fringe benefits will be desired. METHOD SUBJECTS

Approximately 110 persons employed in ~ bank located in a city in upstate New York served as subjects for this study. They represented

A NEW THEORY OF HUMAN NEEDS

159

all job levels in the organization below vice president. With the exception of those people who were absent during the study, they represented all the employees in the organization. PROCEDURE

The employees met during working hours to complete questionnaires for the study, They were infoTmed of the study through a memo signed by the president of the bank, and they were contacted for their participation by the personnel officer. The questionnaire itself was administered by the writer meeting with groups of 12-15 employees at a time. Each questionnaire took about 30 minutes to complete. The data were gathered as part of a morale survey conducted for the organization. The employees were told that a report of their responses by groups would be made to the top management in the organization. The attitude survey was conducted as part of a project initiated by the bank to revise their wage and salary schedule. Prior to the time when the study was conducted fringe benefits in the organization had been reviewed and increased. Recently the organization had successfully begun using electronic data processing equipment. OPERATIONALDEFINITIONS Maslow's work has been predominantly a contribution to the theoretical literature. As a result there is no established paradigm or set of operational definitions for testing the theory. To this writer's knowledge there are three instances in the published literature where writers have attempted to develop operational definitions for Maslow's concepts. Porter (1962, 1963) and Beer (1966) developed questionnaire measures of needs from the five-category system, and Hall and Nougaim (1968) used an interview coding method. Neither Porter nor Beer has actually reported tests of Maslow's theory; they have used their measures for different purposes. Hall and Nougaim, however, did attempt tests of Maslow's theory. In their longitudinal study they employed both static and dynamic analyses. They correlated need-satisfaction in one year with need intensity of that year, and they correlated change in need satisfaction between two years with change in need intensity between the same two years. The present writer has used both interview and questionnaire methods to test E.R.G. theory (Alderfer, 1966). Convergent and discriminant validation of these measures has been reported (Alderfer, 1967). To date, the methodological approach has been eorre/ations between measures of need satisfaction and desire at a given point in time. This approach does

160

CLAYTON P. ALDERFER

not allow for firmly establishing causal direction. Both longitudinal and experimental methods could be profitably employed in future studies. T h e present study utilized static correlations from questionnaires to test between the theories. Satisfaction measures were based on six-point L i k e r b s c a l e items, while the need intensity measures were ratings of "desire." T h e Porter (1962, 1963) and Beer (1966) studies measured need intensity b y " i m p o r t a n c e " ratings or rankings. I n the interviewquestionnaire validity study it was found t h a t while importance and desire ratings were highly correlated for existence and growth needs, t h e y were not for relatedness needs (Alderfer, 1967). T h e desire items used in this study were based on desire ratings and are shown in T a b l e 1. T h e satisfaction items are shown in T a b l e 2. I t e m content for the operational indicators of Maslow belongingness TABLE 1 D~smE I~EMS Instruction: Tell how much more of the following factors you would like to have in your job. 1 means No More 2 means Slightly More 3 means Somewhat More 4 means Much More 5 means Very Much More Existence Needs Pay: I. Good pay for m y work

2. Frequent raises in pay Fringe Benefits: 1. A complete fringe benefit program 2. Frequent improvements in fringe benefits Relatedness Needs

Respect from co-workers: 1. Cooperative relations with my co-workers 2. Respect from my co-workers 3. Openness and honesty with my co-workers Respect from superiors: 1. Respect from my boss 2. Openness and honesty between my boss and me 3. Mutual trust between my boss and me Belongingness (love) Needs

1. The opportunity to be helpful to my co-workers 2. The opportunity to develop close friendships at work Status (esteem) Needs

1. The status my job gives me 2. The feeling that my job is regarded as important.

A NEW THEORY OF HUMAN NEEDS

161

and status needs were taken directly from scales developed by Porter (1962, 1963) and Beer (1966). This was done in order to reduce the possibility of bias because this writer held an alternative point of view. The items were presented in the same format as the E.R.G. items, however, which meant that their format was different from that employed by either Porter or Beer. RESULTS SATZSFACTION OF LOWER-LEvEL NEEDS

The data in Table 2 indicate that for the group as a whole pay satisfaction is moderate at best. An item of interest is ##5, which states that pay is adequate to provide for the basic things in life. The mean response to this item is 3.19, or approximately mildly agree on t h e attitude scale. Sixty-five percent of the respondents indicated that much or more agreement with the item. On the othe r hand, there are the responses to item 3, which states that the person does not make enough money to live comfortably. The mean response to this item is 3.37, or approximately mildly agree on the attitude scale. Forty-nine percent of the respondents indicated that much or more agreement with the item. These responses are consistent with the fact that the organization was in the process of examining its wage and salary schedule. The data on fringe benefits, however, shows a different picture. Item 2, which states that the fringe benefits provide nearly all the security the person wants, has a mean response of 2.40, or approximately agree on the attitude scale. Eighty-six percent of the respondents showed mild agreement or more with this item. These answers are consistent with the fact that the organization had recently revised its benefit program. Also from the data in Table 2 one can see that, overall, belongingness needs seem to be reasonably well satisfied. Ninety percent of the respondents showed some agreement that they had developed close friendships at work. Eighty-eight percent showed some agreement that they had opportunities to help their co-workers. The mean response for both of the belongingness items was nearest to agree on the scales. The level of satisfaction of lower level needs bears only on the tests of Maslow hypotheses. In this study, these tests refer only to the outcomes of hypotheses 1-3. From the data given above one could reasonably conclude that the needs for security (in the form of fringe benefits) and for belongingness were reasonably well satisfied in this organization. The need for pay was less well satisfied. It is not altogether clear tha~ either pay or fringe benefits are adequate operational concepts for physiological and safety needs. It might be argued that fringe benefits are a

162

CLAYTON P. ALDERFER , TABLE 2 MEANS AND STANDARD DEVIATIONS OF NEED SATISFACTION ITEMSa •

(n

=

110)

ExiStence Needs Pay: 1. Compared to the rates for similar work here my pay is good. 2. Compared to similar work in other places my pay is poor. 3. I do not make enough money from my job to live comfortably. 4. Compared to the rates for less demanding jobs my pay is poor, 5. My pay is adequate to provide for the basic things in life. 6. Considering the work required, the pay is what it should be. Fringe Benefits: 1. Our fringe benefits do not Cover many of the areas they should . . . . . 2. The fringe benefit program here gives nearly all the Security I want. 3. The fringe benefit program here needs improvement. 4. Compared to other places, our fringe benefits are excellent. Relatedness Needs Respect from Superiors: 1. My boss will play one person b against another. 2. My boss takes account of my wishes and desires. 3. My boss discourages people from making suggestions. 4. It's easy to talk with my boss about my job. 5. My boss does not let me know when I could improve my performance. 6. My boss gives me credit when I do good work. 7. My boss expects people to do things his way. 8. My boss keeps meinformed about what is happening in the company. Respect from Peers: 1. My co-workers are uncooperative unless it's to their advantage. 2. I can count on my co-workers to give me a hand when I need it. 3. I cannot speak my mind to my co-workers. 4. My co-workerswelcome opinions: different from their own. '

Mean

Standard Deviation

3.69

1.53

3.13

1.62

3.37

1.59

3.15

1.67

3.19

1.50

4.21

1.35

4.61

1.33

2.40

1.00

4.40

1.13

2.36

1.07

4.76 2.46

1.59 1.28

4.72

1.38

2.42 4.18

1.36 1.36

2.47 3.30 3.06

1.28 1.39 1.43

4.82

1.20

1.96

1.10

4.31 2.98

1.35 1.22

A

NEW THEORY OF HUMAN ~'EEDS

163

TABLE 2 (Contin~,ed) 5. My co-workers will not stick out their necks for me. Maslow Interpersonal Needs Belongingness (Love) : 1. I have developed close friendships in my job. 2. I have an opportunity in my job to help my co-workers quite a lot. Status (Esteem) : 1. I have the feeling that my job is regarded as important by other people. 2. My job gives me status. Growth 1. "I seldom get the feeling of learning new things from my work. 2. I have an opportunity to use man;¢ of my skills at work. 3. In my job I have the same things to do over and over. 4. My job requires that a person use a wide ~ange of abilities. 5. My job requires making one or more important decisions everyday.

6. I do not have the opportunity to do challenging things at work.

4.22

1.32

2.08 2.32

1.02 1.05

2.52

1.20

2.94

1.20

4.27

1.50

2.58

1.30

2.78

1.50

2.68

1.34

2.73

1.45

3.79

1.42

a For all satisfaction items: 1 -- Strongly Agree; 2 -- Agree; 3 = Mildly Agree; 4 = Mildly Disagree; 5 -- Disagree; 6 = Strongly Disagree. Words in italics represent abbreviation for items used in later tables. more reasonable indicator of security needs t h a n p a y is of physiological needs. B o t h are existence needs as defined b y E.R.G. theory. I f one grants t h a t an acceptable level of lower need-satisfaction exists, then it is possible to consider the results for H y p o t h e s e s 1-3, concerning M a s l o w t h e o r y predictions a b o u t higher order needs. MhsLow THEORY BELONGINGNESS AND STATUS HYPOTHESES T h e results of the tests of H y p o t h e s e s 1-3 are given in T a b l e 3. H y pothesis 1, predicting a relationship between satisfaction with belongingness needs and desire for status, received no statistically significant support. Moreover, the direction of the observed relationship was as would h a v e been predicted b y E.R.G. t h e o r y for the s u m m e d scales. H y p o t h e s i s 2, predicting a relationship between lack of satisfaction with belonginghess needs and desire for belongingness, also received no statistically significant support. One of the items showed statistically significant supp o r t for H y p o t h e s i s 3. Persons who did n o t see their job regarded as

164

CLAYTON P. ALDERFER TABLE 3 CORRELATIONS TESTING HYPOTHESES 1-3 (n = 110) Desire for:

Belongingness satisfaction Close friendship Help my co-workers Sum of belongingness items

Belongingness

Status

.055 .029 .074

--.078 -.079 -.156 ~

Status satisfaction Job regarded important Job gives status Sum of status items

.158" ,IIi --.094

Not significant according to Maslow predictions; direction is wrong. According to E.R.G. predictions, would be significant at p < .05, 1-tail test. * p < .05, 1-tail test.

important tended to express more desire for status than persons who saw their job as being regarded important. The other status item showed a correlation in the predicted direction, but it did not attain an acceptable level of statistical significance. None of the need satisfaction scales, formed by summing the individual items, showed significant correlations with the desire measures in the direction predicted by Maslow's theory. One might wonder whether failure to find much support for the Maslow interpersonal hypotheses was because lower-level need-satisfaction was not adequate. This question cannot be answered definitively. However, Hypothesis 3 received the strongest support of any of the Maslow hypotheses, and this hypothesis--concerning the impact of status satisfaction on status desires--depended most on lower-level need-satisfaction. Its prior conditions required reasonable satisfaction of physiological, safety, and esteem needs. Moreover, the lack of lower-level need-satisfaction as a possible explanation for the little support shown for the first three hypotheses may be kept in mind when observing the results for Hypotheses 4-15. E.R.G. THEORY RELATEDNESS HYPOTHESES The correlations shown in Table 4 provide strong support for all of the E.R.G. relatedness need predictions. Both individual items and the scales summing the satisfaction items show significant correlations between lack of need satisfaction and strength of desires. Hypothesis 4 is supported by the correlations relating lack of satisfaction with respect from co-workers to higher desires for respect from superiors. Hypothesis 5 is supported by the correlations relating lack of satisfaction with respect from superiors to higher desires for respect from co-workers.

165

A NEW THEORY OF HUMAN NEEDS TABLE 4 CORRELATIONS TESTING HYPOTHESES 4--7 Desire for:

Satisfaction with respect from co-workers Uncooperative Give a hand Cannot speak m y mind Welcome opinions Will not stick out necks Sum of respect from co-workers

Respect from co-workers

Respect from superiors

--.360** .235* --.336** .205* --.247* --.376**

--.204* .077 --.175" .090 --.060 --.163"

- . 207* .143 - . 091 .169" - . 022 - . 247* - . 159" .266** - . 222**

- . 464** .421"* - . 302** .368** - - . 095 - . 060 - . 302** .252** - . 491"*

Satisfaction with respect from superiors Play one person Takes account of wishes Discourages suggestions Easy to talk to Does not improve performance Gives me credit Do things his way Keeps me informed Sum of respect from superiors * p < .05, 1-tail test. ** p < .005, 1-tail test.

Hypotheses 6 and 7 are supported by correlations showing that lack of satisfaction with respect from co-workers and lack of satisfaction with respect from superiors are related to higher desires for those respective needs. The strong support for these hypotheses tends to rule out lack of lower-level need-satisfaction as an explanation for the little support found for Hypotheses 1-3. If lower need-satisfaction was not adequate for interpersonal needs (whether as defined by Maslow or by E.R.G. theory) to be activated, one would not expect to find such strong support for Hypotheses 4-7. This is not to say that lower-level need satisfaction is not required for interpersonal needs to be activated. That issue is not addressed in this study. It is to say, however, if lower-level need satisfaction is required for interpersonal needs to become salient, it was attained in this study. E.R.G.

THEORY PREDICTIONS WITH MASLOW NEEDS

The correlations shown in Table 5 show the results of testing Hypotheses 8-15. Hypothesis 8 received some support. One of the belongingness satisfaction items was significantly correlated with desire for respect from co-workers in the predicted direction. Hypotheses 9, 10, and 11

166

CLAYTON P. ALDERFER TABLE 5 CORRELATIONS FOR TESTING HYPOTHESES 8--15

(n = 110) Desires for:

Belongingness satisfaction Close friendship Help my co-workers Sum of belongingness items Status satisfaction Job regarded important Job gives status Sum of status items

Respect from co-workers

Respect from superiors

.234* .014 .105

.113 --.028 .133

.059 .154 .066

•085 . . . . .139 •133 Desires for:

Satisfaction with respect from co-workers Uncooperative Give a hand Cannot speak my mind Welcome opinions Will not stick out necks Sum of respect from co-workers Satisfaction with respect from superiors Play one person Takes account of wishes Discourages suggestions Easy to talk to Does improve performance Gives me credit Do things his way Keeps me informed Sum of respect from superiors

Belongingness - . 096 .112 - . 175" .036 .035 .085 - . 157 .161" .056 .195" .068 - . 032 - . 067 .157 - . 105

Status -.

163" • 100

- . 143 - . 046 - . 085 -.118

- . 280** .192" .067 .198" ~0Ol - . 014 - . 232** .260** - . 224**

* p < .05, 1-taft test. ** p < .005, 1-tail test. received n o s t a t i s t i c a l l y s i g n i f i c a n t support. T h e r e were n o significant correlations b e t w e e n b e l o n g i n g n e s s need s a t i s f a c t i o n a n d desire for respect from superiors, s t a t u s n e e d s a t i s f a c t i o n a n d desire for respect from co-workers, or s t a t u s need s a t i s f a c t i o n a n d desire for respect from superiors. H y p o t h e s e s 12, 13, a n d 14 received some s t a t i s t i c a l l y significant s u p p o r t . F o r each of these hypotheses, one or more s a t i s f a c t i o n i t e m s were s i g n i f i c a n t l y correlated with desires in t h e p r e d i c t e d direction. One of t h e items c o n c e r n i n g s a t i s f a c t i o n w i t h respect f r o m co-workers was

A NEW THEORY OF HUMAN NEEDS

167

related to desire for belongingness. One of the items concerning satisfaction with respect from co-workers was related to desire for status. And two of the satisfaction-with-respect-from-superiors items were related to desires for belongingness. Hypothesis 15 received strong statistically significant support. Five satisfaction with respect from superiors items and the scale summing all the items were significantly correlated with desires for status in the predicted direction. E.R.G. THEORY PREDICTIONS ABOUT FRUSTRATION OF HIGHER-ORDER NEEDS

The correlations in Table 6 show the results of testing Hypotheses 16-21. Hypotheses 16 and 17 received strong statistically significant support. Several growth-satisfaction items and the scale summing all the items indicated that lack of satisfaction with growth was related to higher desires for respect from co-workers and to higher desires for respect from co-workers and to higher desires for respect from superiors. Hypotheses 18 and 19 received no statistically significant support. There were no satisfaction-with-respect-from-co-workers items that were significantly related to desire for pay or desire for fringe benefits. Hypotheses 20 and 21 received strong statistically significant support. Several respect-from-superiors-satisfaction items and the scale summing all the items indicated that lack of satisfaction with respect from superiors was related to higher desires for pay and to high desires for fringe benefits. SOME ~/~ETHODICALNOTES

One of the questions that naturally arises in a study that utilizes correlations between different parts of the same instrument concerns the possibility of response bias. This possibility was anticipated when the questionnaire was designed, and certain procedures were followed in order to reduce the likelihood of developing support for hypotheses as a consequence of response bias. In addition, there is some empirical data from the study which can be used as a check for response bias. In the questionnaire all satisfaction items were located in one part of the instrument, while all desire items were located in another. The two types of items appeared in different types of formats. Satisfaction items appeared as six-point Likert scales, while desire items appeared as five-point rating scales. Furthermore, among the Likert-scale items some were written so that need satisfaction would be coded from "agree" (low-number) answers, while others were written so that need satisfaction would be coded from "disagree" (high-number) answers. In addition, in neither part of the questionnaire did the several items

168

CLAYTON P. ALDERFER TABLE 6 CORRELATIONS FOR TESTING HYPOTHESES 16--21

(n = 110) Desires for:

Growth satisfaction Seldom learning new things Opportunity to use skills Same things to do over and over Use range of abilities Making decisions Challenging things Sum of growth items

Respect from co-workers

Respect from superiors

--.203* .159" --.230** .204* --.062 --.236**

-.200* .199" -.100 .147 .081 -.080

--.234**

-.191"

Desires for: Satisfaction with respect from co-workers Uncooperative Give a hand Cannot speak my mind Welcome opinions Will not stick out necks Sum of respect from co-workers Satisfaction with respect from superiors Play one person Takes account of wishes Discourages suggestions Easy to talk to Does not improve performance Gives me credit Do things his way Keep me informed Sum of respect from superiors

Pay

Fringe Benefits

-.032

--.062 .039 .138 --.065 .031 --.043

- . 129 .177" .047 .198" .005 .100 - . 153 .167" - . 170"

- . 259** .096 .030 .132 - . 168" .039 - . 046 .134 - . 173"

--.084

.066 --.057 --.090

.004

* p < .05, 1-tail test. ** p < .005, 1-tail test. w h i c h m a d e u p a p a r t i c u l a r need s a t i s f a c t i o n or desire scale a p p e a r n e x t t o each other. F o r t h e E . R . G . items, more o f t e n t h a n not, the w o r d i n g of s a t i s f a c t i o n i t e m s was l i t e r a l l y different f r o m t h e analogous desire items. H o w e v e r , i n t h e case o.f t h e i t e m s used to t e s t M a s l o w predictions, t h e s a t i s f a c t i o n a n d desire i t e m s h a d a l m o s t i d e n t i c a l wording. F u r t h e r more, all t h e s a t i s f a c t i o n i t e m s were s t a t e d positively. T h u s a lown u m b e r a n s w e r i n t h e s a t i s f a c t i o n p o r t i o n of t h e q u e s t i o n n a i r e coupled w i t h a l o w - n u m b e r a n s w e r i n t h e desire p o r t i o n of t h e q u e s t i o n n a i r e

A NEW THEORY OF HUMAN NEEDS

169

could provide support for the Maslow Hypotheses 2 and 3 due to response bias alone. The data show no support at all for Hypothesis 2 and very weak support for Hypothesis 3. The summed satisfaction and desire scales correlate .074 for belongingness needs and -.094 for esteem needs. If these scales are used as the most valid test of the hypotheses, then one can say that essentially no support was found for these hypotheses. When so little support was found for the hypotheses where response bias was loaded in favor of finding the relationships, then one can be more confident that when significant relationships were found, they were not due to response bias. Another methodological point concerns the relative reliability of the E.R.G. versus the Maslow satisfaction and desire scales. Perhaps the E.R.G. relatedness predictions came out better than the analogous Maslow predictions because the scales measuring them were better. The correlation between the Maslow belongingness-satisfaction items was .377 and between the Maslow status satisfaction items was .500. For the Maslow desire scales, the respective inter-item correlations were .400 and .511. For the respect from peers and respect from superiors scales, the median inter-item correlations for the satisfaction scales were .422 and .294, respectively. For the E.R.G. desire scales, the median inter-item correlations were .605 for respect from peers and .749 for respect from superiors. Spearman-Brown estimates of internal consistency of the scales, based on the using of the median inter-item correlations and number of items in the Spearman-Brown formula, showed that the E.R.G. satisfaction scales had slightly higher reliabilities (.79'8 and .645 versus .548 and .667) than the Maslow scales. The E.R.G. desire scales had substantially better reliabilities than the Maslow desire scales (.819 and .861 versus .571 and .676). Therefore, one should question whether the poorer support given to Hypotheses 1-3, as compared to that for Hypotheses 4-7, was because poorer measures were used to test them. SUMMARY OF HYPOTHESIS TESTS

It is possible to provide an overall assessment of the hypothesis tests by forming a chart indicating the degree of support shown for the various hypotheses. We adopt the following convention: (1) if an hypothesis had no items reaching statistical significance it is termed receiving "no" support; (2) if an hypothesis had one or more items, but not the sum of items, reaching statistical significance it is termed receiving "some" support; and (3) if an hypothesis had several items and the sum of the items reaching statistical significance, then it is termed receiving "strong" support. Table 7 shows the summary of sup-

170

CLAYTON P. ALDERFER TABLE 7 SUMMARY OF SUPPORT FoR HYPOTHESES

Masl0w-theory belongingness and status predictions HI. None (Belongingness satisfaction and status desires) H2. None (Belongingness satisfaction and belongingness desires) H3. Some (Status satisfaction and status desires) E.R.G.-theory relatedness predictions H4. Strong (Respect-from co-workers satisfaction and respectfrom-superiors desires) H5. Strong (Respect-from-superiors satisfaction and respectfrom-co-workers desires) H6. Strong (Respect-from-co-workers satisfaction and respectfrom-co-workers desires) H7. Strong (Respect-from-superiors satisfaction and respectfrom-superiors desires) E.R.G. theory with Maslow belongingness and status~needs predictions HS. Some (Belongingness satisfaction and respect-fromco-workers desires) H9. None (Belongingness satisfaction and respect-from-superiors desires) H10. None (Status satisfaction and respect-from-co-workers desires) Hll. None (Status satisfaction and respect-from-superiors desires) tt12. Some (Respect-from-co-workers satisfaction and belongingness desires) H13. Some (Respect-from-co-workers satisfaction and status desires) H14. Some (Respect-from-superiors satisfaction and belongingness desires) H15. Strong (Respect-from-superiors satisfaction and status desires) E:R.G. theory frustration of higher-order needs predictions H16. Strong (Growth satisfaction and respect-from-co-workers desires) H17. Strong (Growth satisfaction and respect-from-superiors desires) H18. None (Respect-from-co-workers satisfaction and pay desires) H19. None (Respect~from-co-workers satisfaction and benefits H20. Strong (Respect-from-superiors satisfaction and pay desires) H21. Strong (Respect-from-superiors satisfaction and benefits desires)

Fraction of items significant at ~.05 0 0 1/2 2/4 5/8 5/5 6/8

1/2 0 0 0 1/5 1/5 2/8 5/8 5/6 2/6 o 0 3/8 2/8

port received by the various hypotheses. Hypotheses 1, 2, 9, 10, 11, 18, and 19 received no support. Hypotheses 3, 8, 12, 13, and 14 received some support, and hypotheses 4, 5, 6, 7, 15, 16, 17, 20, and 21 received strong support.

A NEW THEORY OF HUMAN NEEDS

171

DISCUSSION E.R.G. theory was proposed to deal with the problem of how needsatisfaction was related to need strength. In order to be a viable alternative to current conceptual schemes, the new theory should provide greater conceptual clarity and show more empirical validity. These comparisons may be made in two ways: (1) with the simple frustration hypothesis which implies that any frustrated need will increase in strength but implies no connection between needs of different types, and (2) with Maslow's hierarchical theory of motivation which includes five categories of needs and includes the major assumption that the categories are arranged in an order of prepotency such that a certain degree o f satisfaction o f lower-level needs is required for higher-order needs to increase in strength. The strong support given to hypotheses 4, 5, 16, 17, 20, and 21 for example, suggest that the simple frustration hypothesis alone is not adequate to account for the relationship between satisfaction and desires. Each of these hypotheses represented a situation where the degree of satisfaction of one need was related to the strength of desire for a di#erent need. In the case of hypotheses 4 and 5, the different desire came from the same overall need category, namely relatedness needs. But in the case of hypotheses 16, 17, 20, and 21 the different needs belonged to other general need categories. For hypotheses 16 and 17, lower satisfaction of growth needs was related to higher desires for relatedness needs. For Hypotheses 20 and 21, lower satisfaction of relatedness needs was related to higher desires for existence needs. The strong support for hypotheses 6 and 7 was consistent with the simple frustration hypothesis. However, since these and other hypotheses follow from proposition 4, one might suggest that the E.R.G. formulation accounts for the phenomena predicted by the simple frustration and more. The outcomes on hypotheses 16, 17, 20, and 21 also provide a basis for favoring E.R.G. theory over Maslow's hierarchical theory. These results support hypotheses generated by E.R.G. theory that were not part of Maslow's theory. They show a place where that theory was incomplete and where E.R.G. theory was not. It should be recognized, however, that two other hypotheses, 18 and 19, were not supported by the empirical results. Both of these hypotheses show that it was satisfaction with respect from co-workers that did not show a relationship to existence desires. In contrast, satisfaction with respect from superiors did show the predicted relationship to both desires for pay and fringe benefits. If this pattern of results is replicated, it suggests that E.R.G. may have to be

172

CLAYTON P. ALDERFER

modified along the lines that specific relatedness frustrations tend to produce specific existence desires. When hypotheses 1 and 2 received no support and hypothesis 3 only some support, one might raise questions about the adequacy with which Maslow's theory deals with interpersonal needs. The predictions which followed from E.R.G. concerning interpersonal issues--namely 4, 5, 6, and 7--all received strong support. Thus, within its own conceptual system, E.R.G. theory did better than Maslow's theory did within its own conceptual territory. Furthermore, it was possible to develop a series of predictions which applied E.R.G. predictions to Maslow-type needs. Any support for these predictions, found in hypotheses 8-15, would tend to favor E.R.G. theory over Maslow's theory. There was some empirical support provided for five out of eight of those hypotheses. If E.R.G. provided the more adequate conceptualization, then it is not surprising that these "cross" theory predictions did not receive very strong support since they did involve mixed conceptualizations. One should recall, however, that the Maslow needs were not measured as reliably as the E.R.G. needs were. This would certainly reduce the likelihood of finding support for the Maslow predictions. But, one cannot be too firm about the implications of not finding much support for hypotheses 1, 2, and 3 from Maslow's theory due to the relatively lower reliability of the Maslow need scales compared to the E.R.G. scales. More effort has been spent de~eloping reliable and valid scales for the E.R.G. needs than for the Maslow scales, and it is not surprising that these scales should show higher reliability. However, it also may be that Maslow's concepts are harder to operationalize because they are not clear abstractions of reality. Recently, Schneider (1968) has factor analyzed a series of questionnaire items built from Maslow's concepts. His factor structure from a sample of 150 nurses did not result in the Maslow items forming factors corresponding to the need categories. Rather there were factors consisting of self-actualization and autonomy items, security and social items, and esteem items. These results tend to support the overlapping nature of categories in Maslow's system discussed above. Schneider was not aware of this writer's work when he designed his scales and collected his data. Moreover, his results contrast with a factor analysis of E.R.G. items, which did tend to show results consistent with the E.R.G. conceptualization (Alderfer, 1967). Another reason why one might not totally explain the lack of support for the Maslow hypotheses as a function of low reliability of measurement stems from the data pertaining to the outcome of Hypothesis 1. This hypothesis, derived from Maslow's theory, was not sup-

A NEW THEORY OF HUMAN NEEDS

]73

ported by the empirical data. Moreover, the sign of the relationship was as would have been predicted by E.R.G. theory. Stemming from having one category for interpersonal needs, that theory would have predicted a negative relationship between belongingness need-satisfaction and desire for esteem. Even though the data reported here generally tends to favor E.R.G. theory over both the simple frustration hypothesis and Maslow's theory, there are several good reasons to view the results tentatively. First, the data were taken from a single organization, and there is no way of knowing what special conditions in that organization may ha~e favored the particular outcomes observed here. Second, there is a legitimate question whether Maslow's constructs were adequately operationalized. Since this investigation was carried out by the author of a competing point of view, Maslow's concepts were operationalized by relying on the work of others (in this case Porter and Beer) in order that no bias might be introduced by this writer inventing measures for Maslow's concepts. But further efforts should be devoted to developing reliable measures for Maslow's concepts. Third, not all the comparative predictions that could be tested between the theories were studied in this investigation. It might turn out that other comparative predictions would favor Maslow's theory. Finally, the methodology used here was only one of several possible ways to test the predictions. It did not allow for tests of the direction of causality which both theories imply. Each of the preceding limitations of the present study suggests directions for future research. If field settings are to be used to test the theories, several organizations with different purposes might serve as sites for future studies. These studies might also be concerned with developing more adequate measures of Maslow's concepts. The additional comparative predictions implied by the two theories might be tested in future work. Finally, it would be especially impolSant to employ better methods than static correlations. Longitudinal studies and both field and laboratory experiments provide better possibilities for inferring causality than static correlations. In short, the: present study represents just a beginning on the problem of developing a more valid theory relating need satisfaction to strength of desires. REFERENCES ALDERFER, C. P. Convergent measures by interviews 1967, 51, 509--520. ALDERFER, C. P. Differential faction obtained in the University, 1966.

and discriminant validation of satisfaction and desire and questionnaires. Journal o] applied psychology, importance of human needs as a function of satisorganization. Unpublished doctoral dissertation, Yale

]74

CLAYTON P. ALDERFER

ARGYRIS,.C.Integrating the individual and the organization. New York: Wiley, 1964. ARGYRIS, C. Interpersonal competence and organizational effectiveness. Homewood, Illinois: Irwin-Dorsey, 1962. BARKER, R. G., DEMR0, T., AND LEV~IN, K. Frustration and regression, In R. G. Barker et al. (Eds.), Child behavior and development. New York: McGrawHill, 1943. Pp. 441-458. BARNES, L. B. Organizational systems and engineering groups. Boston: Harvard Graduate School of Business, 1960. BEER, M. Leadership, employee needs, and motivation. Columbus, Ohio: Ohio State University Division of Research, 1966. CLARX, J. B. Motivation in work groups: a tentative view. Human organization, 1960-61, 13, 198-208. CONANT, J. B. On understanding science. New Haven : Yale, 1947. DUNNETTE, M. D. The motives of industrial managers. Organizational behavior and human performance, 1967, 2, 176-182. ERIKSON, E. Childhood and society. New York: Norton, 1950. FREUD, S. Therapy and technique. New York: Collier Books, 1963. HALL, D. T., AND NOUGAI~, K. E. An examination of Maslow's need hierarchy in an organizational setting. Organizational behavior and human performance, 1968, 3, 12-35. HARRISON,R. A conceptual framework for laboratory training, Mimeo, 1966. HARVEY, 0. J., HUNT, D. E., AND SCHRODER~H. M. Conceptual systems and personality organization. New York : Wiley, 1961. JACQUES, E. Equitable payment. New York: Wiley, 1961. LANGER, W. C. Psychology and human living. New York: Appleton-CenturyCrofts, 1937. LAWLER, E. E., AND PORTER, L. W. Perceptions regarding management compensation. Industrial relations, 1963, 3, 41-49. LAWSON, ~:~., AND MARX, M. H. Frustration: theory and experiment. Genetic psychology monographs, 1958, 57, 393-464 LEWIN, KURT. Field theory in social science. New York: Harper & Row, 1951. LEWIN, KURT, DEMB0, T., FESTINGER, L., AND SEARS, P. Level of aspiration. In J. McV. Hunt (Ed.), Personality and the behavior disorders. New York: Ronald Press, 1944. Pp. 333-378. MASLOW,A. H. Motivation and personality. New York: Harper, 1954. MASLOW, A. H. A theory of human motivation. Psychological review, 1943, 50, 370-396. MASLOW, A. H. Toward a psychology of being. Princeton, New Jersey: D. VanNostrand, 1962. MURRAY, H. A. Explorations in personality. New York: Oxford University Press, 1938. PORTER, L. W. Job attitudes in management: I. perceived deficiencies in need fulfillment as a function of job level. Journal of applied psychology, 1962, 46, 375-384. PORTER, L. W. Job attitudes in management: II. perceived importance of needs as a function of job level. Journal o/applied psychology, 1963, 47, 141-148. ROSENZWEIO, SAUL. An outline of frustration theory. In J. McV. Hunt (Ed.), Personality and the behavior disorders. New York: Ronald Press, 1944. Pp. 379-388.

A NEW

THEORY

OF H U M A N

NEEDS

175

SC~EIN, E. H. Organizational psychology. Englewood Cliffs, New Jersey: PrenticeHall, 1965. SCHNEIDER,BENJAMIN.Personal communication, September 11, 1968. STRAUSS, G., ANn SAYLES,L. R. Personnel, Englewood Cliffs, New Jersey: PrenticeHall 1967. ZALEZNIK, A. Management of disappointment. Harvard business review, 1967, 45, 59-70. RECEIVED: June 28, 1968