Personal Influence. Ordinal Position and Purchasing Behavior

Citation:

D. F. Kirchner (1971) ,"Personal Influence. Ordinal Position and Purchasing Behavior", in SV - Proceedings of the Second Annual Conference of the Association for Consumer Research, eds. David M. Gardner, College Park, MD : Association for Consumer Research, Pages: 82-98.

Proceedings of the Second Annual Conference of the Association for Consumer Research, 1971     Pages 82-98

PERSONAL INFLUENCE. ORDINAL POSITION AND PURCHASING BEHAVIOR

D. F. Kirchner, San Fernando Valley State College, Northridge, California

[The author is an Associate Professor of Marketing at San Fernando Valley State College, Northridge, California.]

Though the opinion leader concept has occupied an important niche in the literature of consumer behavior for many years, it has remained largely nonoperational and has been of greater significance to academicians than to practitioners. This condition may stem from a number of practical as well as conceptual difficulties which have been cited by Nicosia (1964), Arndt (1967) and recently by King and Sommers (1970).

The main thrust of this paper represents an attempt to examine some of the behavioral foundations of personal influence, and at least tentatively to suggest that they can be partially organized by psychological variables such as gregariousness, affiliation and dependence, factors which appear to be summarized by individual differences in order of birth.

The view of opinion leadership adopted here is based on the observation that this phenomenon is widely distributed in the population (King & Sommers, 1970). It is a view which focuses on the transactional nature of personal influence and explicitly recognizes the interaction of influencer and influence noting that it is possible for these roles, however subtly, to be exchanged perhaps repeatedly in the course of a single conversation. Finally, it considers the possibility of modes of leadership and followership which occur within as well as among individuals.

MODES OF PERSONAL INFLUENCE BEHAVIOR

Rogers (1962) has suggested in connection with the adoption process the existence of active and passive adopters and rejecters, people who play strong or weak leader roles with respect to the acceptance or rejection of new ideas or practices. Kirchner (1969) applied this concept to behavior rather than to individuals and hypothesized two modes of leadership (active and passive), two parallel modes of followership and a final category, independence behavior.

Active leadership manifests itself in an outgoing, gregarious manner and is characterized by the volunteering of information or advice when it appears needed. Passive leadership is more taciturn and restrained. Information may be offered when requested, rarely volunteered. Similarly, active followership implies an overt search for information from personal sources. ln contrast, the passive mode is of a more deferential nature and is not characterized by an active search of personal sources, but rather by a willingness to accept advice when offered. Finally, independence behavior suggests a preference for non-personal sources of information when contemplating purchase decisions.

Scales were developed to measure a propensity to perform these kinds of behavior. [Five separate scales totaling 39 six-point items comprise the Personal Influence Scales. The results suggest that the constructs are scalable. A variety of reliability and validity evidence is available in Kirchner (1969). Sample scale items are reported in the Appendix.] Intercorrelation of scale scores suggests that the gregarious elements of personal influence behavior, active leadership and active followership, positively covary and that these are negatively associated with independence behavior. Of the passive modes, passive leadership is positively associated with active leadership and active followership but not significantly associated with passive follower or independence behavior. Passive follower scores positively covaried with active follower scores, were negatively correlated with independence, and were not associated at all with any of the leader scales.

The inference tentatively drawn from these results was that individuals who are gregarious leader types are also likely to be adroit followers, seeking out and susceptible to personal influence as well as offering it to others. This finding is consistent with earlier leadership studies which suggest that those viewed by their colleagues as desirable leaders were also preferred followers (Hollander, 1964). As expected, the gregarious leader-follower pattern is not always compatible with independence behavior.

AFFILIATION, DEPENDENCE AND ORDINAL POSITION

Consistently the personal influence literature suggests that those in leader positions tend toward gregariousness. Studies of affiliation indicate that under conditions of ambiguity or anxiety, some individuals tend to seek out and are more susceptible to information from personal rather than impersonal sources.

Schachter (1959) suggested that the tendency to define reality via personal referents is an important determinant of gregariousness. He also discovered that those who seemed to display the greatest inclination to personal referents and who were most susceptible to them were those who were first born or only children. Presumably, earlier borns are raised in an environment significantly different from that of their later born siblings. It has been suggested that with earlier borns, parents are less certain as to their child raising abilities, are more solicitous, spend more time with these children, and respond to more cues more quickly. Hence, they provide a more personalized kind of anxiety reduction for earlier born children not enjoyed by later borns. As subsequent children appear, parents are more experienced, have greater demands placed upon their limited time, and perhaps become more blase' about the child rearing task with the result that the earlier borns typically receive ministrations different in number and kind than their later born siblings.

Other factors have been proposed as accounting for ordinal position differences in addition to the relative inexperience of the parents. These include a period of concentrated exposure to adults without the influence of competing siblings and the greater achievement expectations of parents for their first born.

Selected Ordinal Position Findings

In recent years, the ordinal position literature has grown substantially. The results of seemingly similar studies are not always in agreement, a state that can be partially explained by sampling variation, differences in working definitions, research designs and environmental dissimilarities. The ordinal position tradition extends back more than forty years and encompasses a variety of dependent variables. Findings pertaining to affiliation, dependence, conformity and leadership tend to be more consistent than those of other variables. A sampling from the literature will illustrate this point.

Dependence, Affiliation and Conformity

A variety of field and laboratory studies indicate that in an anxiety arousal condition, earlier borns choose the company of others or indicate a preference for the presence of others more 80 than later borns (Gerard & Rabie, 1961; Zimbardo & Formica, 1963; Miller & Zimbardo, 1966; Suedfeld, 1969; McDonald, 1969a, 1970). Eisenman (1966) found that first borns in group therapy were more likely to speak and ask questions than later borns, which was interpreted as an indication of seeking others for assurance or cognitive clarification. Earlier borns anticipated greater psychological distress prior to an eight-hour sensory deprivation experiment (Zuckerman & Link, 1968), showed less anticipated stress when given personal assurances (Helmreich, et al. 1968), and actually experienced more stress as a result of this type of experiment (Dohrenwend & Dohrenwend, 1966).

Similar results were found under low anxiety conditions but were not always quite so clear cut. They include reports of mothers and teachers as to the dependence of earlier borns (Sears, 1950, 1957); evidence that they belong to more social groups {Schachter, 1959; Smart, 1965), that they are more likely to volunteer for group rather than individual experiments (Dohrenwend, et al., 1967) and are more inclined to discuss areas of disagreement with others (Radloff, 1961).

Psychometric evidence as to dependence, affiliation and conformity is available from a number of sources. Conners (1963) reported stronger affiliative needs among earlier borns, Dember (1963) and Rosenfeld (1966) reported similar ordinal position results as did Allman and White (1968). Sampson and Hancock (1967) used the Edwards Personality Preference Schedule and found statistically significant differences regarding conformity, achievement and autonomy but not for affiliation although for the most part the results were in the predicted direction.

Susceptibility to Social Influence

A variety of evidence suggests that earlier borns are more socially sensitive than later borns. Wrightsman (1960) found greater anxiety reduction among earlier borns who waited with others for an experiment to continue than earlier borns waiting alone. In an autokinetic study (Staples & Walters, 1961), in Asch-type experimental environments (Becker & Carroll, 1962; Becker, Lerner & Carroll, 1964, 1966), in a field study comparing slum dwellers with persons from middle class households (Dohrenwend & Dohrenwend, 1966), earlier borns appeared to be more susceptible to social stimulation. Dittes (1961) demonstrated that earlier borns were highly responsive to peer rejection while later borns "were virtually invulnerable". S'totland and Walsh (1963) indicate that first borns are more likely to use others as a guide in evaluating themselves. Earlier borns tended to reject deviates and bring their own opinions more rapidly back into line with the group than later borns (Arrowood & Amoroso, 1965). Rhine (1968) demonstrated that first born girls were more conforming while McDonald (1969b), showed earlier borns to be more normative with respect to religious behavior.

Ordinal Position and Leadership

A host of evidence, however disparate, suggests that the first born are frequently overrepresented in the population of those formally or informally defined as "eminent". From Galton's (1874) English Men of Science through Stewart's (1970) review of the U. S. Presidency the evidence appears to be accumulating in support of this view. Excellent summaries are provided in Kammeyer (1967) and Altus (1966).

A sample of the evidence is reviewed here. It consistently indicates that earlier borns are overrepresented among superior high school students, in college populations and especially in graduate and professional schools (Altus, 1965; Bradley, 1969). A variety of studies of scientists (Cattell, 1947), attorneys (Very & Prell, 1970), Rhodes Scholars (Apperly, 1939) and National Merit Competition finalists (Altus, 1966) indicate an overrepresentation of earlier borns. Earlier borns have been shown to have higher I.Q. scores (Burton, 1968), are more articulate and have higher verbal and quantitative skills than later borns (Lunneborg, 1968). Evidence that first borns are overrepresented in formal leader positions is available in Smith and Goodchilds (1963) and Stewart (1970). Earlier borns have been found to be among the sociometrically chosen (Alexander, 1966) and indicate preferences to work with ideas and direct others based on Kuder Preference data (Oberlandin, et al., 1970).

The potential connection of these findings to personal influence is direct. If opinion leaders, especially the active variety, and active followers tend to be gregarious, and if earlier borns also tend toward gregariousness and occur in unexpectedly large numbers in various leader positions, it then seems reasonable to suggest the possibility of an overrepresentation of the earlier born among those who display a tendency toward more socially oriented modes of personal influence behavior. Likewise, they might be underrepresented among the less gregarious elements of the model.

HYPOTHESES

H1: Earlier borns are overrepresented in the opinion leader group.

H2: Earlier borns are overrepresented among those classified as active leaders.

H3: No significant birth order differences exist among passive leaders.

H4: Earlier borns are overrepresented among those classified as active followers.

H5: No significant birth order differences exist in the passive follower group.

H6: Earlier borns are underrepresented among those classified as independents.

METHOD

The Personal Influence Scales along with the leadership scale reported by Rogers (1962) were administered to two samples of undergraduate subjects at San Fernando Valley State College. Hypotheses were tested based on the number of first born and only children who ranked above and below the mean value of the distribution of scores for each scale. The results were controlled for sex and family size. Both samples excluded Ss who were raised in homes in which parents were divorced or deceased, who were adopted or who were twins. This procedure resulted in the elimination of 13 Ss from the original sample and 21 Ss from the second sample for a total of 12.3%.

RESULTS AND DISCUSSION

To provide a basis for comparison with the results of the Rogers' scale the active and passive leader scales were scored as a combined instrument. The data are in Table 1.A and appear to support the general hypothesis that first borns are overrepresented among those classified as leaders. Table 1.B separates out the earlier born into first born and only children e The data indicate that only children contributed heavily to this outcome in both samples e

The results for the Rogers' scales (Tables 2.A and 2.B) indicate no significant differences exist in either sample but here as in the previous case only children appear to be overrepresented among leaders.

Hypothesis II receives substantial support from Tables 3.A and 3.B. Earlier borns are overrepresented among active leaders. The predominance of only children is again evident.

Hypothesis III suggests that since passive leadership does not presuppose the same degree of gregariousness, there is no basis for expecting an ordinal position effect. This hypothesis is confirmed in Tables 4.A and 4.B, but as before the only children appear to show up more heavily among passive leaders but not at a statistically significant level. It is noteworthy that the passive leader results parallel those obtained with the Rogers' scale.

Because of the overrepresentation of only children in these results, a reexamination of the literature was undertaken to determine if a plausible explanation was available. Published studies only infrequently distinguish between specific ordinal positions and in particular between first born and only children. A few, however, were helpful.

Sears (1957) describes the world of the only child as one which is fundamentally adult oriented, the most readily available model being the parents. If earlier borns receive an abundance of adult attention because parents lack confidence and have more time, then in the absence of later children which serve to redirect parental energies, only children should be the object of solicitous and anxious parents for a far longer period. Schachter (1959) reported a study by Haeberle (1954) which indicated that only children were more dependent than first born. Conners (1963) indicated that only children possess greater affiliative needs than first and later born. Sampson and Hancock (1967) suggest that only children are more conforming and are lowest of the birth ranks studied with respect to n autonomy. Though the evidence is sketchy, it tends to support the speculation of a hierarchial relationship as to dependence, affiliation and conformity. What one might expect for the first born on these dimensions, one might expect even more so from only children.

The fourth hypothesis suggests an overrepresentation of earlier borns in the active follower distribution. The evidence in Tables 5.A and 5.B does not provide support for this hypothesis. The only children in the first sample appear to behave according to expectations though not at a statistically significant level, but the results of the second sample show a shift in direction.

Several explanations are possible. First, a number of previous works suggest that ordinal position differences are most apparent under anxiety arousal conditions. In this instance, the testing environment may have been insufficient to stimulate these differences. The non-significant results and in particular the reversal may also be at least in part due to a deficiency in this particular scale. Of the five scales developed, only the active followership scale showed a multiple pattern of factor loadings which was difficult to clearly interpret.

TABLE 1

BIRTH ORDER AND OPINION LEADERSHIP (COMBINED)

TABLE 2

BIRTH ORDER AND OPINION LEADERSHIP (RC)

TABLE 3

BIRTH ORDER AND OPINION LEADERSHIP (ACTIVE)

TABLE 4

Birth ORDER AND OPINION LEADERSHIP (PASSIVE)

TABLE 5

FOLLOWER (ACTIVE) AND BIRTH ORDER

The results for the passive follower category are as expected. No significant differences exist. The data in Table 6 are consistent in both samples. With respect to independence behavior, the results are not quite so clear. The hypothesis suggests that earlier borns will be underrepresented among independents. The data in Tables 7.A and 7.B indicate an overrepresentation of first born, an underrepresentation of later born with no clear relationship for only children.

Interpretation of this finding is speculative at best. The work reviewed earlier by Sampson and Hancock (1967) offers some insight. In addition to the conformity findings, Sampson and Hancock report EPPS results which indicate an ordinal position ranking on n autonomy. Higher scores were reported for first borns followed in order by later borns and only children. This result suggests that first borns possess a stronger inclination to resist coercion or domination. The authors note that "it is almost as though n autonomy reflects a rebellion against one's siblings relationships."

The work of Becker, Lerner and Carroll (1964, 1966) may also be instructive. First born subjects appear to be more susceptible to personal influences which were "normative" rather than informative. Later borns were more sensitive to personal sources which were informative. Normative influences are socially supportive and pertain to the need to meet the expectations of others whereas informative influences provide worthwhile information regarding one's environment. The questions on the independence scale may have been interpreted by respondents as being mainly informative.

Each of the scale results were controlled for sex and family size. The analysis by family size indicated nothing of significance, however, the number of individuals in family sizes with 1, or more children was too small to provide a meaningful examination. No significant departures from the results previously reported were observed when the male-female dimension was examined except in the case of the independence scale. The analysis indicates nothing significantly different from that indicated earlier for first and later borns. Male only children are overrepresented and females are underrepresented. The female results are consistent with the high dependence, low n autonomy explanation. The male results are difficult to interpret .

In sum, the results appear to support the leadership hypotheses particularly in the case of only children. The follower and independence findings presented some difficulties which require methodological as well as conceptual refinement.

The results reported here provide an alternative approach potentially useful in organizing the personal influence process. The value of this study does not reside in the immediate field relevance of the findings but rather in the explicit recognition of some of the complexities of the personal influence process. It points to the necessity for further theorizing as to the behavioral factors underlying leader, follower and independence behavior and underscores the admonition that future research proceed from a more sophisticated theoretical model than is now available.

TABLE 6

FOLLOWER (PASSIVE) AND BIRTH ORDER

TABLE 7

INDEPENDENCE AND BIRTH ORDER

APPENDIX

PERSONAL INFLUENCE SCALES

The following are examples of the questions that appear on the five subscales:

To answer the questions on this page, choose one of the following options and write the number in the space provided.

1) Agree strongly.

2) Agree moderately.

3) Agree slightly.

4) Disagree slightly.

5) Disagree moderately.

6) Disagree strongly.

Leader (Active): I am often influential in sending my friends to stores where they can obtain "good buys".

Leader (Passive): I am often asked for advice by friends who are not satisfied with some product purchase that they have made.

Follower (Active): For many of the products that I purchase, I find that asking others about them is usually a good way to get information and gain insight.

Follower (Passive): For most purchase decisions, merely observing how others behave is usually of greater value than published sources of information.

Independence: When considering a purchase decision where the choice is not clear, I am inclined first to seek out published sources of information.

Scoring was based on a seven point system with non-responses being assigned a score of four.

REFERENCES

Allman, T. and White, W. F. Birth Order Categories as Predictors of Selected Personality Characteristics, Psychological Reports, 1968, 32, 857-60.

Alexander, Norman C. Ordinal Position and Sociometric Status, Sociometry, 1966, 29, 41-51.

Altus, W. D. Birth Order and Academic Primogeniture, Journal of Personality and Social Psychology, 1965, 2, 872-76.

Altus, W. D. Birth Order and Its Sequelae, Science, 1966, 151, 44-49.

Arndt, J. Word of Mouth Advertising, New York: Advertising Research Foundation, 1967.

Arrowood, J. and Amoroso, D. Social Comparison and Ordinal Position, Journal of Personality and Social Psychology, 1965, 2, 101-104.

Becker, S. W. and Carroll, J. Ordinal Position and Conformity, Journal of Abnormal and Social Psychology, 1964, 69, 318-323.

Becker, S. W., Lerner, M. J., and Carrol, J. Conformity as a Function of Birth Order. Payoff, and Type of Group Pressure. Journal of Abnormal and Social Psychology, 1964, 69, 318-23.

Becker, S. W. Conformity as a Function of Birth Order and Type of Group Pressure: A Verification, Journal of Personality and Social Psychology, 1966, 3, 242-44.

Bradley, R. W. and Sanborn, M. P. Ordinal Position of High School Students Identified by their Teachers as Superior, Journal of Educational Psychology, 1969, 60, 41-45.

Burton, D. Birth Order and Intelligence, Journal of Social Psychology, 1968, 76, 199-206.

Cattell, J. M. Men of Science, Lancaster, Pa.: Science Press, 1947.

Conners, C. K. Birth Order and Needs for Affiliation, Journal of Personality, 1963, 31, 408-16.

Dember, W. M. Birth Order and Need Affiliation, Journal of Abnormal and Social Psychology, 1964, 68, 555-57.

Dittes, J. E. Birth Order and Vulnerability to Acceptance, American Psychologist, 1961, 16, 358.

Dohrenwend, B. S. and Dohrenwend, B. P. Stress Situations, Birth Order and Psychological Symptoms, Journal of Abnormal Psychology, 1966, 71, 215-23.

Dohrenwend, B. S., Feldstein, S., Plosky, J. and Schmeidler, R. Factors Interacting with Birth Order in Self Selection Among Volunteer Subjects, Journal of Social Psychology, 1967, 72, 125-128.

Eisenman, R. Birth Order, Anxiety and Verbalizations in Group Psycho-Therapy, Journal of Personality and Social Psychology, 1966, 30, 521-26.

Galton, F. English Men of Science: Their Nature and Nurture. London: Machillan, 1874.

Gerard, H. B. and Rabie, J. M. Fear and Social Comparison, Journal of Abnormal and Social Psychology, 1961, 62, 586-92.

Helmreich, R., Kuiken, D. and Collins, B. Effects of Stress and Birth Order on Attitude Change, Journal of Personality, 1968, 36, 466-73.

Hollander, E. P. Leaders, Groups and Influence, New York: Oxford University Press, 1964.

Kammeyer, E. Birth Order as a Research Variable, Social Forces, 1967, 46, 71-80.

King, C. W. and Sommers, J. O. Overlap of Opinion Leadership Across Consumer Product Categories, Journal of Marketing Research, 7, 43-50.

Kirchner, D. F. Personal Influence, Purchasing Behavior and Ordinal Position, unpublished doctoral dissertation, Graduate School of Business Administration, University of California, Los Angeles, 1969.

Lasko, J. K. Parent Behavior Toward First and Second Children, Genetic Psychology Monographs, 49, 1, 1954, 3-95.

Lunneborg, P. W. Birth Order, Aptitude and Achievement, Journal of Consulting and Clinical Psychology, 1968, 32, 101.

McDonald, A. P., Jr. Manifestations of Differential Levels of Socialization by Birth Order, Developmental Psychology, 1969.a, I, {485-92.

McDonald, A. P., Jr. Birth Order and Religious Affiliation, Developmental Psychology, 1969.b, I, 628.

McDonald, A. P., Jr. Anxiety, Affiliation and Social Isolation, Developmental Psychology, 1970, 3, 242-54.

Miller, N. and Zimbardo, P. Motives for Fear-Induced Affiliation: Emotional Comparison or Interpersonal Similarity, Journal of Personality, 1966, 34, 470-503

Nicosia, F. M. Opinion Leadership and the Flow of Communication: Some Problems and Prospects, in L. G. Smith (Ed.). Reflections on Progress in Marketing. Proceedings of the 1964 Educators Conference, December, 1964. Chicago: American Marketing Association, 1965.

Oberlandin, M. I., Frauenfelder, K. H. and Heath, H. Ordinal Position, Sex of Sibling and Personal Preferences in a Group of 16-year Olds, Journal of Consulting and Clinical Psychology, 1970, 35, 122-25.

Radloff, R. Opinion Evaluation and Affiliation, Journal of Abnormal and Social Psychology, 1961, 62, 578-85.

Rhine, W. R. Birth Order Differences in Conformity and Level of Achievement Arousal, Child Development, 1968, 39, 987-96.

Rogers, E. Diffusion of Innovation. New York: The Free Press of Glencoe, 1962.

Rosenfeld, H. M. Relationship of Ordinal Position to Affiliation and Achievement Motives, Journal of Personality, 1966, 64, 155 59.

Sampson, E. E. and Hancock, F. An Examination of the Relationship Between Ordinal Position, Personality and Conformity: An Extension, Replication and Partial Verification, Journal of Personality and Social Psychology, 1964, 68, 453-56.

Schachter, S. The Psychology of Affiliation, Stanford: Stanford University Press, 1959.

Schachter, S. Birth Order and Sociometric Choice, Journal of Abnormal and Social Psychology, 1964, 68, 453-56.

Sears, R. R. Ordinal Position in the Family as a Psychological Variable, American Sociological Review, 1950, 15, 397-401.

Sears, R. R., Maccoby, E. and Levin, H. Patterns of Child Rearing, Evanston, Ill.: Row, Peterson, 1957.

Smart, R. G. Social Group Membership, Birth Order and Leadership, Journal of Social Psychology, 1965, 67, 221-25.

Smith, E. E. and Goodchilds, J. D. Some Personality and Behavioral Factors Related to Birth Order, Journal of Applied Psychology, 1963, 47, 300-03.

Staples, F. R. and Walters, R. H. Anxiety, Birth Order and Susceptibility to Social Influence, Journal of Abnormal and Social Psychology, 1961, 62, 716-19

Stewart, L. The Politics of Birth Order, Proceedings of the Annual Convention of The American Psychological Association, 1970, 5 (pt. 1), 365-66.

Stotland, E. and Walsh, J. A. Birth Order and Experimental Study of Empathy, Journal of Abnormal and Social Psychology, 1963, 66, 610-14.

Suedfeld, P. Sensory Deprivation Stress: Birth Order and Instructional Set as Interacting Variables, Journal of Personality and Social Psychology, 1969, 11, 70-74.

Very, P. S. and Prell, R. W. Birth Order, Personality Development and Choice of Law as a Profession, Journal of Genetic Psychology, 1970, 116, 219-21.

Wrightsman, L. S. Effects of Waiting with Others on Changes in Level of Felt Anxiety, Journal of Abnormal and Social Psychology, 1960, 61, 216-22.

Zimbardo, P. G. and Formica, R. Emotional Comparison and Self Esteem as Determinants of Affiliation, Journal of Personality, 1963, 31, 141-62

Zucherman, M. and Link, K. E. Expectancy and Birth Order as Determinants of Affective Responses to Isolation, Perceptual and Motor Skills, 1968, 27, 279-86.

----------------------------------------