ABSOLUTE FINAL DUE DATE. Early papers ARE accepted.




EDP5285-01          SPRING 2009




Some 30 years ago, my brother and his first wife separated. I was--and am--very fond of them both. Anguished, I asked my about to be former sister-in-law what had happened--or, as we say in Social Psychology, for "her account". "Well, " she said, "Nobody was the boss in this family. Somebody has to be boss."

Do they? Must groups have a "boss" or a leader? Does this differ by type of group? If so, what kind of a leader must a family have? Would this kind of "boss" differ from a friendship group or a school principal?

Leaders influence? Leaders guide, structure, support? Leaders provide vision? Leaders dominate? Leaders dictate? There are almost as many definitions of leadership as there are situations. Is leadership part of one's personality, a set of traits? Is leadership a set of skills that we can learn? Is leadership something like math--some people "catch on" faster but nearly everyone can learn the basics?

The answers are critical because so many situations require leadership. While we often glorify or vilify vivid individuals--Adoph Hitler, Franklin or Eleanor Roosevelt, Martin Luther King, Mohammed, Queen Elizabeth I, or other memorable figures, there are police officers, fire fighters, teachers, parents, and other everyday persons who provide leadership in myriad situations. If leadership is a set of skills, then we all have the potential to "lead" at some point.



Leadership involves an asymmetrical influence upon group members in the direction of collective, interdependent goals. It also means coordinating member behavior in pursuit of group goals. Leaders use resources and a repertoire of  styles, i.e., recognizable and regular constellations of behaviors directed at influencing others.

Although leadership is asymmetric, i.e., influence flows more from leaders to group members than vice-versa, leadership still has reciprocal properties, involving exchange and negotiation (LMX models) between leaders and group members. One cannot lead without acceptance in some form among group members. In efficient societies and among more informal or volunteer groups in any society, leadership invokes conformity rather than compliance.

Common layperson references to leadership usually focus on personal behaviors rather than social roles. When someone expresses solely in-role behaviors that are legitimately designed to influence others, we typically call this authority. Examples of expressing legitimate authority are:

Occupants of an authoritative role typically exert highly situation-specific influence in modern industrial societies. At a party, your dentist will either refrain from asking you to "open wide" or only joke about it. Upon meeting your physician at the supermarket, she will not ask you to disrobe. In some institutions, such as the military or in most K-12 schools, the influence of authority figures generalizes to many situations.

However, we distinguish between the military officer who delivers orders and the officer who inspires soldiers to follow, or between the Chief Executive Officer who "manages" daily activities and one who presents a bold plan for the company's future (i.e., who "leads"). Despite the daily necessity on many occasions for the former, we typically reserve the term "leadership" for the latter.

The socially naive assume a leader orders others about, enforces obedience and expects compliance. When my 20 year old son became a junior counsellor at Y camp, he had a curious complaint when we picked him up for Camp Break: "None of the kids do what I tell them to do," he moaned (as we cackled under our breath). However, externally-imposed compliance requires surveillance. Dictators or bullies cost societies valuable resources to ensure that members comply. Recall Spain's Guardia Civil under dictator Francisco Franco. Spain, an under industrialized country at the time in the early 1970s, could ill afford to place a soldier on every street corner in Madrid.


In Lewin, Lippett and White's classic experiments, sets of older elementary school boys rotated among three "leaders" in hobby groups.

Students were actually the most productive under the autocratic leader--as long as he stayed in the room! Members of the laissez-faire group actually worked more on their hobbies when "the leader" left the room. The democratic leader was the most popular. The groups with the laissez-faire leader were typically unproductive and dissatisfied. Student hostility was highest under the autocratic leader.

Later research, such as Vroom's, has replicated many of these early findings with some variations. Vroom suggests that the effectiveness of participatory leadership depends on the situation, e.g., the relationships among group members. Hersey and Blanchard single out the level of group development (situational leadership theory). Mature groups may function quite well under a "delegating" style, which resembles laissez-faire leadership.

Interestingly enough, the behaviors of the democratic leader are very close to what developmental research label "the authoritative parent" as opposed to the authoritarian parent (comparable to the autocratic leader above). Both the democratic leader and the authoritative parent are typically warm, friendly, empathetic and concerned, while at the same time holding and enforcing high standards for group [family] members. Children of authoritative parents tend to have higher school achievement.

While the initial Lewin, et al. results were interpreted as a kudo for democratic leadership (CLICK HERE for a sidenote on Zeitgeist, or the spirit of the times), remember that students in groups with "autocratic leaders" outperformed the others while they were supervised. Other research confirms this finding, leading some teachers and executives to ponder whether you must be "nasty" to get results. Probably not. Early research confounded several behaviors in the "autocratic leader":

  1. Initiating and maintaining group structure.
  2. Issuing directives about sequences of actions.
  3. Maintaining on-task behavior.
  4. Monitoring productivity.
  5. One-way flow of communication and influence from group leader to members.
Both democratic and autocratic leaders initiate and maintain structure. Autocratic leaders may be more aloof or insistent on hierarchy--but we know that occupying either a leadership or an authority position may induce these behaviors in many members even if the leader her/himself does not insist on it. Freed Bales maintained that to execute points one through four will almost inevitably irritate at least some group members and create friction.

However, autocracy is usually reserved for #5, a downward only  flow of communication and influence from leaders to "followers." In theory, at least, one can be both structured and more democratic.


Well, we already know that leaders aren't bullies, although an autocratic style may be effective under some circumstances. We often either follow leaders because we want to or because social norms of legitimacy say that we should. But is leadership a set of traits or a set of skills?

Many philosphers, social scientists and the "person in the street" endorse the idea that "leaders are born, not made." In this perspective, shown in the recent book "Outliers", individuals possess stable personal characteristics or "traits" that earmark them for leadership positions. The sources of these traits (genetic? family training? an excursion to Bethel, Maine for the training labs?) may not be described. What ARE these traits? Scholars such as Lord have studied implicit leadership theories, or the traits that group members believe leaders possess. They mention qualities such as dominance, task performance, determination, a "commanding presence," intelligence, honesty, or humor. These traits tend to be the same ones that are cited in social commentary and in some social science writings.

How accurate are these professional and layperson theories? Systematic research attempts to find correlations between personal traits and leadership typically fail. Worse yet, the most talkative person, the tallest person, the middle-aged person, or the White male, rather than the group member with the most task or social skill, often becomes the emergent group leader (i.e., a leader selected by the group after an intial interaction period). As a result of group leaders who resemble commercials for expectation states theory (diffuse and status specific characteristics) more than they do a reliable personality syndrome, the "trait-situation" combination approach as opposed to a "pure trait" approach has become much more influential.We will return to this idea shortly, under laboratory created groups.

However, there are some consistent results. Emergent leaders tend to be more intelligent than the "average group member"--although not vastly so. They tend to be conscientious, friendly and socially sensitive (probably high emotional IQ), with a good sense of humor. Merei, who studied children's groups during the 1950s, found that children who repeatedly emerged as leaders when shifted across groups quickly discovered important norms in the group and conformed to them. After their initial conformity, these children would initiate changes. These results suggest that social flexibility is important in leaders.

Some would call this constellation of attributes Machiavellian. The child leaders Merei studied were more interested in ultimately directing the group than in particular interests or activities. Michael Macobby, in his book The Gamesman, describes a corporate or organizational leader who has little interest in job skills (that's the Craftsman), little ideological commitment to any position or in vindicating his or her views (that tends to be the "Jungle Fighter"), or, indeed, any interest in anything at all except becoming the boss. A cool amoral aura surrounds the Gamesman.

Some characteristics mentioned such as self-confidence, dominance, and commitment to a vision may be true of charismatic leaders who lead by inspiring followers (rather than "group members"). Intense and persuasive, charismatic leaders seek to convert others to the group's views and goals. Charismatic leaders are more common among emergent leaders than in bureaucracies. Differences in situations probably contribute to the overall low correlations between personal traits and leadership.

In addition, the low systematic research support for the "trait" approach has led some social scientists to suspect that bosses and higher status individuals support ideologies of personal leadership traits to justify their elevated positions (see last part of Guide 8). After all, isn't it only equitable that such "superior people" receive superior rewards?

Other social scientists support a "positional" approach: leaders are made, not born. They assert that leadership is a collection of skills (e.g., supporting group members, providing motivation, keeping the group on task). While some individuals develop these skills prior to assuming a leadership role,  others have leadership thrust upon them, developing skills "on the job." (Of course, as the Johnsons note, one does not magically acquire leadership skills simply by assuming a  position of authority.)

Alvin Zander's research on leadership in field situations strongly supports the situational position (although his findings suggest that at least some leaders import their skills). In several studies, Zander worked with naturalistic groups in the field, but he also randomly assigned leaders to teams. He found that some individuals engaged in leadership behaviors regardless of whether they occupied central or peripheral group positions. However, virtually all persons assigned to central positions became more directive and showed other leadership "traits." Consistent with Zander's results, studies of group communication structures typically report that persons located in the central or "hub" position of centralized communication structures disproportionately emerge as group leaders.

A middle ground is the idea of "skill transport" or "skill generalization." People may take skills that have worked well in one type of leadership situation (such as giving orders) and transport or generalize these to a second situation. Thus skill that were learned in situation one appear as "personality traits" in situation number two because they appear effortless and "ingrained." Again, we will revisit this notion below.


Why such inconsistency in trait studies of leadership? One problem is that exponents of the trait approach seem to assume that most leaders are emergent, i.e., groups select the leader. Although this may be true in many informal groups or in volunteer organizations, it is probably not the case for businesses, schools, churches and synagogues, or politics. Thus, traits that scholars and laypeople assume characterize leaders in implicit leadership theories may be instead the traits that group members wish that their leaders had, and which leaders indeed may have in disproportionate amounts in emergent groups.

But, instead, leaders are often imposed upon groups through appointment rather than through internal selection. Many behaviors of these appointed individuals reflect authority, not leadership. Indeed, "leaders" may be chosen for authority roles because they appear to embody the goals of the organization or because they "manage" rather than "lead."

In any event, authority figures may be--or may not be--emergent leaders. Upper management chooses most middle managers, although input from subordinates may be encouraged. Upper management, in turn, is often appointed by company presidents, boards of directors, church elders, school boards, business owners, high school faculty, or people in other relatively elite positions. New middle or upper managers may be even recruited from outside the organization. Political leaders typically are selected by their parties before their names go on the ballot (except sometimes in local races).

These circumstances may explain why studies in naturalistic settings generally find little correlation between personal traits and leadership position (let alone leadership success). Pleasing status superiors rather than working well with group members may determine one's appointment to authority roles. Relational skills such as ingratiation or conformity may influence leadership selection in formal groups. Outstanding individual task achievement may also impress status superiors, even when individuals with these accomplishments have poor relational skills.

Most studies which identify "emergent leaders" occur in laboratory settings in which the study participants are typically undergraduates taking introductory psychology, sociology, business, or communications courses. The assumption is that these students are pretty much the same (they are all undergraduates, aren't they?) and thus the individuals who emerge as leaders in these created groups have some type of "special quality" about them. It might be charisma, it might be organizational, it may even be "foot in the babbling mouth" disease, whatever the personal trait, researchers usually assume that the group began as a level playing field of equals, and emergent leadership reflects stellar personal characteristics.

Unfortunately, at the same time, laboratory researchers collect little background data on participants. They may think they do by administering batteries of "personality tests." And many study directors will probably collect data on race, ethnicity, gender, and maybe age.

However, the background characteristics that I'm interested in for this context aren't in "personality tests." What is the grade level of each participant: freshman or senior? Does the participant belong to a prestigeous fraternity or sorority? What's the student's major field? Is the student a recognized campus leader? Does the student's parent(s) occupy a leadership position? To researchers, faculty or graduate students, these are characteristics that belong in their own "distant pasts." To them, these undergraduates seem pretty much alike: YOUNG! Excepting, of course, easily observable traits such as gender or race--or maybe height.

However, it is very likely that the students themselves in these groups notice the fine distinctions. If Bob belongs to a prestigeous fraternity or is a senior, if Mary is the President of the college student council or is a major in math, rather than English, almost certainly students in these created groups will discover these characteristics fairly quickly in the chit-chat among participants before the official experiment begins. If students have had experience with the task before (Mary on a mathematical task; Bob on campus knowledge), they will probably head the queue as "emergent leaders."

And, of course, the invidious distinctions among students who supposedly operate on a "level playing field" in the laboratory are mirrored 100 times over out in the "real world." Juries, who typically elect their foreman (advisedly, most are male), tend to pick better educated White males with prestigeous jobs. Congregation presidents tend to be better educated (and mostly male to boot).

So, are we looking at the personality characteristics that make individuals rise like cream to the top of the milk jug as emergent leaders--or are we looking at status characteristic composites--whether diffuse statuses (gender, ethnicity) or situation-specific qualities--that specific individuals possess and that group members recognize as indicating experience or stereotypically suggesting someone with "more ability"? Are what some researchers call "personality traits" instead simply the transfer of skills from one set of experiences to another?

Without research designed to explicitly test these general hypotheses, we simply do not know. Do "leaders" tend to dominate the group conversation? So do individuals with high status characteristics, and they interrupt other members more, too. If Bill had leadership experience in a congregation Youth Group, is Bill a "leader," or simply more experienced with Robert's Rules of Order than most of the other "group" members?

I don't think it would be too difficult to test these hypotheses (limit pre-group small talk; present bogus information about group members; measure those background experiences and characteristics), and someone here may want to follow up (great topic for an Educational Leadership person!) The point is that most research using college students just doesn't check these hypotheses out, and as a result, I think we need to question the material on "personality" as a factor in emergent group leadership too.

I am not negating the idea that "some leaders are born" or that personal qualities can make a difference (see below for one of my "epiphany examples")--I would just like to rule out some of the alternative hypotheses that I raise in this section FIRST.

I was all set to be a positional or situational "leaders are made not born" type until I studied religious congregations. There I found that leadership style could make a huge difference, particularly when a congregation was unaffiliated with a national denomination. One striking example was a large, local church with no national affiliation. At least 1800 parishioners attended each of its two Sunday morning services. It encompassed a highly rated private academic school (pre-k through 12th grade), a television station, and its own homeless shelter. This entire conglomerate was started by its first pastor. While this future minister served as an American soldier in Korea, he promised God that if he survived he would begin a church. Talk about no atheists in foxholes! One man's vision created it all.

What is vision? Is it charisma? But the literature on charismatic leaders discusses leadership style (intense, confident) rather than content. A key issue, of course, with vision IS content! Johnson and Johnson recognize its importance but don't define it clearly (I'm not sure how close I'll come either.) Grappling away, I see a leadership vision   as an individual's original created script and schema for a new organization or set of interrelationships.

This particular pastor possessed vision, holding a clear blueprint for, first, his church, then its later extensions. He worked tirelessly to implement his vision. He told the project interviewer that he called 50 different parishioners each week. I believe him! He called the observer twice in four months. Each time he introduced himself, mentioned his happiness that Traci attended services, then asked what the church could do for her. Because the congregation had no national affiliation, he could not borrow from prefabricated sermons or Sunday School outlines, thus he had to work harder. He ministered to his flock for over 20 years. Toward the end of his tenure the Board of Elders bestowed "dictatorial powers" upon him for life ! Instead, he happily retired, with a new minister at the helm, knowing that he had initiated one of the largest, most complex religious organizations in North Florida.

In general, I still support the positional perspective on leadership, but I have unbent enough to recognize that there are exceptions such as this ambitious--and skilled--religious enterpreneur.



One gigantic consistency in the twentieth century literature on informal groups is that two types of leaders typically emerge in goal-oriented groups that interact over time. Whether we call it a task leader, a low LPC leader, a goal oriented leader, or a bureaucratic leader, the term addresses someone who is highly task oriented, directive, who initiates and coordinates group activity. The socio-emotional leader, high LPC, supportive, people-oriented, or social relations leader interjects humor, is friendly, complimentary, lowers friction, and generally contributes to a positive group climate. Freed Bales, who was one of the first to systematically observe these two emergent forms through his Interaction Process Analysis and groups that met for several weeks, believed that both were necessary: the task leader to move the group toward goal achievement, and the socio-emotional leader to promote interpersonal group cohesion. Further, he believed it was difficult to combine these two skills, since some friction would occur as a task leader urged goal completion or criticized achievement efforts. Further research indicates that the emergence of either or both kinds of leadership depends on the task, characteristics of group members, organizational characteristics. The effectiveness of each basic type of leader is examined under contingency models in the next section.

Although Lord has proposed that the above typology is a socially constructed fiction, occurring because of what group members believe leadership to be (implicit leadership theories) and thus selectively perceive, I respectfully disagree. The emergence of these two types of leaders has been documented so thoroughly, by researchers using very different typologies for at least 50 years, in so many different groups that we can safely accept it as a generalization.


Although I suspect that the human preference to concentrate on task versus social leadership is probably a timeless phenomenon, the socio-political climate plays an undeniable role in the success of different leadership styles. Picture, for example,  societies characterized by "the divine right of kings," in which political rulers enjoyed nearly total control. Or totalitarian societies in which any political choice was conspicuously absent. In these cases, socio-emotional leadership was a "luxury," nice but not necessary, since the titular leader could pick and choose [typically] his leadership style. The United States factory boss, until the early twentieth century, could choose to be a tyrant secure in the knowledge that the government, the clergy, and the "Pinkertons" were there to support him if workers complained, rebelled, or attempted to form unions. In these cases leadership was clearly general, diffuse and not situation specific.

The 1950s, in contrast, was marked by a labor shortage of male workers in their 20s because birth rates during the 1930s and early 1940s had been so low. Recruiting and retaining good workers was more difficult. Should we then be surprised that the 1950s ushered in an era of "enlightened management," in which managers were expected to be more sensitive to worker feelings and the rise of "democratic leadership"? I think not! With control less than total, companies were forced to develop new, less autocratic, styles of leadership.

We see the emergence of more participatory and democratic styles of leadership when competition increases across groups and organizations for members. For example, viewing "college students as consumers" and more as partners in the educational enterprise gained credence during the 1980s as colleges which had expanded (or emerged) during the 1960s and 1970s, when baby boomers were late adolescents, faced the contracting "birth dearth" generations. Informal friendship groups, which rely on socio-emotional bonds to survive, rarely support dictatorial leaders.


Leaders use what is available to them to influence group members. This means using bases of power (French and Raven) such as:

For example, a coach may be an attractive figure who has a background in professional sports, thereby drawing on referent, legitimate and expert power bases simultaneously.

Bases of power transform into styles of influence (Kelman) such as

Clearly compliance draws heavily on reward and coersion power while conformity draws on referent, legitimate and expert power. In identification processes, group members accept influence because they identify with the leader and want to be like her or him. Typically this means the leader uses personal attributes such as warmth, attractiveness, "spectacular achievements" or expertise rather than positional attributes such as an authoritative role.

By this time it should be clear that the same styles, or bases of power and influence, are not equally effective across situations. Rather a combination of leadership style and group characteristics should be more predictive of group productivity and satisfaction. The most famous theory combining leadership style and situational factors is Fred Fiedler's contingency model. He assessed leader motivational style through a semantic differential scale called the Least Preferred Co-worker (LPC) Scale, individuals who rate their least preferred coworker in a relatively positive manner are more relationship oriented. Low LPC scorers are considered more task oriented.

Fieldler identified three major group factors:

The extreme combinations of these three factors are "good" when leader-member relations are good, the task is highly structured, and the leader's position power is high. The "bad" extreme occurs where leader-member relations are poor, the task is unstructured, and position power is low. Under both extreme combinations, Fiedler believes Low LPC leaders will be more effective. Human relations skills (the High LPC leader) become more important in the ambiguous (and more frequent) middle combinations.

Blake and Mouton's Ohio State Model creates a Leadership Grid to measure task and socio-emotional leadership style preferences. Their perspective, as well as Hersey and Blanchard's situational leadership theory, discusses various combinations of these styles. While the Ohio State Model generally argues for a high level of both kinds of leadership, situational leadership theory, which considers the level of group development and the type of task, extends Fieldler's ideas. Both situational leadership theory (which is newer) and Fieldler's Contingency Model (in particular) have received considerable empirical testing and a reasonable amount of research support.


A key finding in the Groupthink studies is that it takes more than [interpersonal] cohesion to create the syndrome that contributes to poor decision-making. An important element is leadership. "Groupthink" is more likely to emerge when leaders:

The role of the leader thus becomes critical in group decision-making. Don Forsyth, for example, contrasts studies of John Kennedy's style at cabinet meetings before and after the Cuban Bay of Pigs invasion.

In reviewing the literature on leadership styles and contingency models of leadership, the role of individual determinism can be overemphasized. Many middle managers have little choice about their leadership style, no matter what their personal preference. For example, in large organizations from factories to universities, quotas may be imposed from above. The middle manager may have relatively little say about what the group goals may be or even how they are to be obtained. In department required courses, the college instructor may have relatively little control over which material is covered. K-12 teachers may keep a nervous eye on state-mandated achievement test content. Task-oriented leadership rules the day! The socio-emotional oriented leader may "sneak in" friendliness, concern, and support but democratic leadership is not an available option.

Does this make a difference? Almost certainly. When group members have a say in group goals and their implementation, they work harder without constant supervision. Perhaps group leaders may be able to increase participation in goal implementation, even if the actual goals are imposed from above. Leaders may also utilize resources such as expertise, legitimacy, or rewards to create a more positive group climate even if they are unable to implement greater group member influence.


Having considered styles of leadership, the "making of leaders," and how leaders exert influence, we are left with an important question: do groups even need leaders? After all, many small friendship groups exist with, at best, rotating informal leadership. In Jewish synagogues, a "minyan" of 10 adult members can conduct a service. Feminist groups have experimented with attempts to prevent emergent leadership, for example, by insisting that each woman in the group speak first before others take another turn. Small companies have tried to eliminate "middle management," using a relatively horizontal structure in which all employees simply report to the business owner. Such experiments have become more common in the current economic landscape in an attempt to control costs.

When tasks are unitary, goals are operational, or individuals strongly identify with their work and have past histories of meeting group goals, an immediate team leader may be unnecessary. For example, university faculty typically report directly to a department or disciplinary chairperson.

However, most groups are much more complex. Consider a public high school which has several departments representing different disciplines. There will be special groupings (choir, the football team). Faculty and professional staff are not interchangable. Further, there are many support staff in specialized positions, such as secretaries, guidance counselors, and security personnel. Group members are interdependent upon all other members filling their jobs well. Thus,  the high school will require considerable coordination and various middle managers will be appointed.

I began this Guide with a long-ago personal example when my brother's first marriage fell apart (he has been happily remarried for nearly 30 years now.) Do we need bosses in the family? And what are they like?

Freed Bales' colleague, the late Talcott Parsons, decided to exterpolate the Bales findings on emergent leaders, from all-male groups, to the family. You may have already guessed this one: daddies and husbands were the task leaders and mommies and wives the socio-emotional leaders in the Parsons schema.

But, wait a minute! With respect to the family, this just doesn't make sense at all. Who typically pays the bills? Who does the grocery shopping? Who typically represents the family at Parent-Teacher meetings, neighborhood meetings, or at church or synagogue? These are not socio-emotional tasks (tough to assign warm fuzzies to shopping for groceries), they are instrumental tasks--and it is usually the wife who performs them. Further, research indicates that fathers do the bulk of playing with children--decidedly socio-emotional material.

The moral of the story is that task or socio-emotional leadership isn't determined by your gender hormones in all probability. The task structure, your leadership style, societal expectations and stereotypes, your experience in those settings will all play a large role. And does the family have to have a boss? In the immortal words of the group dynamicist, I guess it all depends.

When we review the influence of occupying a leadership position on the individual, it can be enough to send us scurrying for cover with a strong admonition to keep group leadership to a minimum!

Whether appointed, elected, or randomly tossed into place, over time leaders tend to:

Couple these actions and perceptions with the tendency for subordinates to ingratiate and the stage is set for problematic decision making.

Given our research knowledge, it is probable that subordinates will keep on flattering and leaders will continue to engage in self-delusion. What can be done to counteract these tendencies?



This page was built with Netscape Composer.
Susan Carol Losh April 5 2009
The boss?