The instructional value of fostering social capital in the classroom.

Social capital is a term that encompasses social networks, emotional bonds and habits of reciprocity that promote trust and cooperation. Course designs in political science, economics and psychology that promoted social interaction were more effective in achieving their desired results than those that utilized traditional atomized classes in which students pursue parallel rather than interactive learning. Political caucus participation, learning intuitively about the money creation process and spontaneous psychological group leadership and participation were the outcomes sought and achieved.

**********

Social capital consists of social networks, habits of cooperation and bonds of reciprocity that serve to generate benefits for members of a community. These exchanges have lower opportunity costs than comparable market-mediated services partly because they are voluntary (non-contractual) and are likely to involve more intrinsic satisfaction. They embody the emotional bonds of group support and trust (Fukiyama, 1995; Putnam, 2000).

Social groups that encourage honest self-presentation, have few incentives for competition, and have a reason to exist, often result in the development of bonds of friendship and trust among its members. We believe it is because such bonds are intrinsically rewarding to members. Examples are small participative religious congregations, regular attendees of Narcotics Anonymous meetings, local participative political caucuses and psychotherapy groups. These groups tend to possess social capital.

When strangers gather for purposes of convenience with little incentive to communicate, they are likely to engage in parallel rather than interactive behavior. Examples would be spectators at a sporting event, library patrons and members of a lecture-style college class. These collections of individuals can be described as atomized. They provide no gratuitous emotional support to each other although some may converse with strangers or acquaintances (perhaps partisans of the same team attending an athletic event).

College Instruction and Social Capital

All universities in Texas require six semester hours in American Government and Politics to qualify for a bachelor's degree. One three-hour course focuses on the American Constitution and national, state and local political processes. The other, on institutions of national, state and local government. This is imposed by state law and has been for decades.

Clearly the Texas Legislature views the understanding of government as important. This can only be an attempt to promote intelligent citizen political participation. A standard "lecture-textbook-term paper" course, however, provides content that is detailed, dull, atheoretical, inherently trivial, and is instantly forgotten by students after exams. This is particularly true when objective tests are employed.

We believe that a textbook course can't provide information relevant to participation in politics, will be viewed only as a source of college credit, and can't enhance the enjoyment of political involvement. An interactive, real world oriented course represents an alternative more likely to promote informed political participation if the instruction embodies rehearsals, feedback from social interaction, and information generated in the form of cognitions and emotions combined. The desired outcome, after all, is voluntary behavior.

To this end, an experiment was performed in the spring of 1976 to determine whether an experimental course that incorporated the incidental development of social capital would produce higher political participation rates than a conventional course and higher rates than adult demographic groups known for traditionally high participation.

The Experiment

Hosen (1978) designed and taught three experimental sections of the course that covered the U.S. Constitution and American politics. Both the experimental course and the standard textbook course which served as a control were taught during the spring semester of 1976. The experimental students, in contrast with the control students, were required to:

1. render written and oral decision on hypothetical constitutional cases that embodied contemporary issues, rather than simply memorizing the history of the Constitution as in conventional courses. This involved an interactive process in which students debated, consulted, compromised and sometimes forged a consensus, as the U.S. Supreme Court does in real life;

2. understand policy content and objectives and the realities of political decision-making as applied to policy questions. This required familiarity with various proposed policies, alternatives to each proposal, and the array of possible outcomes, intended or unintended. Students were free to apply their own values in selecting policy options if they could demonstrate an awareness of the high degree of predictive uncertainty involved. They could not substitute ideology for an empirically-based predictive model, however crude, approximate, or conditional. The standard courses focused on lifeless historical processes rather than living content; the rules of elections, rather than political dynamics; the formal role of national political conventions, etc.

3. track the nationally publicized and closely fought 1976 campaigns for the Republican and Democratic presidential nominations. The experimental course material consisted of public information on policy proposals by candidates; application of models for understanding the consequences of policy alternatives; assessment of the likelihood that a candidate's proposed policy would achieve the stated objective if implemented; and estimation of the probability that a given candidate would be capable of implementing the policy. Topics also included the operation of parties; electoral mechanics; party primaries and caucuses; candidate strategies and issue selection; and the constitutional basis for all of the above. The non-ideological orientation to issues proved congenial to students.

4. use class time to debate campaign-relevant topics, pooling any specialized knowledge they possessed; rehearsing caucus procedures and forms of participation such as bargaining, consulting, initiating policy proposals, formulating resolutions, and promoting one's own candidacy for delegate or caucus chairperson. Students in the standard courses discussed issues as a minor part of their course work rather than as the core.

Control students did receive extra course credit for participating as a campaign volunteer while the experimental students were explicitly told they would receive no credit for participation in the actual voting and caucus activities in the real world. The experimental course grades were entirely determined by written exams on policy determination and constitutional problem-solving; follow up individual oral exams; contributions to class deliberations.

The experimental courses built social capital among classmates by its nature and design while control class participants remained atomized, even as they engaged in verbal learning about political participation, ironically an activity in which social capital is nearly indispensable.

Results

In Texas, state and local party primaries are important. Most districts are dominated by one party whose candidate almost always wins the general election. Since this candidate is selected in a party primary many months before the general election, the primary is the defacto election.

In 1976 Texas didn't have a presidential primary. The April party primaries selected only candidates for the U.S. Congress plus candidates for state and county offices. Delegates to party presidential conventions were chosen by precinct caucuses who chose delegates to district caucuses who chose delegates to state conventions who chose delegates to the national convention. Precinct caucuses were held after the polls closed and participation was confined to those who had voted in their party primary earlier in the day. Few citizens even knew such a process existed.

Without a presidential primary to attract voters, a 50% voter turnout was considered quite high. Factors that correlate positively with the propensity to vote include years of education, number of college degrees earned, membership in a politically active occupational organization and years of age. Participation in a highly publicized 1975 Houston mayoralty campaign (a nonparty election) was a good proxy for prior political interest among the college students, many of whom were in their late 20's and 30's.

A useful reference point or baseline was obtained by surveying 60 public school teachers who had a minimum of a master's degree, taught at the same school, and included many who were active in a teacher's union. They shared at least a moderate amount of social capital. The percentage who voted in the primary was 51%, a very respectable turnout by Texas standards.

The student experimental and control sections were filled by routine computer assignment, a blind process. The control group consisted of six sections of the same course (n = 150) taught with standard textbooks supplemented by discussions and taught by six different instructors. The experimental group (n = 100) filled three classes all taught by the experimenter. The experimental group had a 71% voter turnout; the controls, 37%. The experimentals exceeded both the student controls and the teacher group in voter turnout at the .01 level of statistical significance (see original 1978 article). Experimentals with prior interest turned out 80%; controls, 49%; difference significant at .05 level. Experimentals with no prior interest, 57%; controls with no prior interest, 25%; significant difference at the .05 level.

The act of voting required far less time, energy and inconvenience than caucus participation. Voting also was a solitary, secret activity while caucuses represented a public, transparent group activity usually involving strangers. There was, for students (or any other participants), no clear instrumental motivation to participate in either. No individual could affect the outcome and no further academic credit was at stake. There were classes for several weeks beyond election day and a student could, if she wished, report her caucus participation experiences to her classmates. Only the experimental classes, however, had laid the groundwork for such spontaneous sharing. Exchanging political stories probably yielded intrinsic satisfaction based on the previously developed social capital.

Teacher caucus participation was 2.6% (far above the general population average); control students, 3.8%; experiments, 26.3%. Caucus participation can be passive (attend, perhaps speak but probably not; nothing more) or active (submit, read and/or speak on behalf of a resolution; run for chairperson of caucus; seek to be part of delegation to the district convention; seek seat at state convention; seek a special committee assignment at the district convention). All teacher participation was passive. For student controls, 3.1% (out of 3.8%) were active. For experimentals, 18.4% (out of 26.3%) were active. Experimental participation was 600% greater than for either student or teacher controls.

The experimentals produced a state Republican delegate, a Democratic caucus chairperson and two foreign students (part of experimental class but not the experimental group) who asked to attend as observers (an Indian national and a Peruvian national). Resolutions were written, submitted, read and debated by 10.8% (of 26.3%). Many were elected as delegates to district caucuses.

There was a possibility that the experimental course had produced a high level of temporary motivation which would quickly fade. Six months after the courses ended, the Presidential (General) Election was held. The experimenter had had no contact with experimental students during that time. An independent research team recontacted experimental and control students and found that participation in the Presidential election was 88% for experimentals; 51% for controls; significant at the .01 level.

Voting in the U.S. resembles buying a lottery ticket, both in terms of powerlessness of the individual voter and in terms of voter psychological involvement. Caucus participation, however, resembles democratic activity and a sense of societal self-determination.

Social Capital and College Economics Instruction

Cameron (1997) developed a money supply creation simulation adapted to the classroom to bring the process to life for introductory and intermediate college economics students. The innovation was quite effective but required modifications to offset the atomized class structure that college economic courses promote. The simulation could only approximate reality if students could be induced to initiate contact with each other. Thus the development of social capital became part of the standard instructions Cameron provided other instructors. Financial intermediaries, it seems, can best be represented by active humans rather than graphic representations in a textbook.

Differential Instructional Outcomes Attributable to Social Capital

During the two year period of 1993-94, Hosen performed an instructional experiment using all students in six community college general psychology classes (n = 180) to which he had been assigned as instructor. He performed a similar instructional exercise with members of six group counseling classes (n = 60) that he also taught during that period.

The general psychology classes each consisted of approximately 30 students, mostly sophomores. Each student had no more than two or three friends in the same class and usually the number was zero. A typical student might know four or five classmates as casual acquaintances. Seating was unassigned and the chairs faced the front of the room. The format of the course involved teaching theory by examining current or recent events known to most students, fleshing out media reports with details and then attempting to fit a variety of psychological principles to the data.

The course generally embodied an evolutionary perspective on psychological functioning in which the individual is very loosely guided by a subjective hedonic calculus. This perspective is detailed in Hosen (1993); Hosen, Stern & Solovey-Hosen (2001); and Hosen, Stern & Solovey-Hosen (in press).

Two sample course topics come to mind. A report published around 1993 indicated that 16% of female military personnel stationed at sea for lengthy periods prior to, and during the Gulf War, became pregnant. A variety of plausible circumstances were discussed in class with respect to possible motivation (loneliness, desire for motherhood, sexual deprivation, fear of war, boredom, rape, seduction, etc.) and execution (when, where, whether or not in privacy, etc.) The presence of numerous male and female Gulf War veterans in class provided an alternative explanation. They said it was a means of escaping further duty. This was typical of student involvement in the general psychology classes when they had specialized knowledge.

Another topic was a highly publicized mass murder in a cafeteria in central Texas that occurred during the course. Student unfamiliarity with mass murder led them to misconceptions which yielded to facts and useful theory. They assumed that the time that elapsed during which a gunman killed 17 elderly people and wounded perhaps twenty more within a single large room would be determined by the firing speed and capacity of his two weapons. They hypothesized that an enraged man bearing familiar weapons would use up less than two minutes. The elapsed time was actually 40 minutes, incompatible with uncontrollable rage. The gunman actually sought to extract satisfaction from the exercise of total power over his potential victims, approaching each and deliberately deciding whether or not to execute the person he was facing. This was closer to sociopathy and narcissism than to an impulsive psychotic response to perceived injustices.

The instructor did not solicit information about personal problems but he listened if told privately. Some students participated in a study that involved submitting diaries or activity logs but these were confidential. In short, there was a relaxed class atmosphere, not too different from that of the previously described experimental course in American politics. The psychology students did not, however, constitute an intimate, interdependent social group.

The comparison group (n=60) consisted of six group counseling classes taught over the same period. The counseling classes were limited to an enrollment of 10 students. They were popular and always filled rapidly. The purpose of the class was to prepare college freshman or sophomores who planned to serve as addiction counselors to conduct small therapy groups. The average age of such students was about 30 years; many had had tumultuous lives. The course provided no lectures and used no text. It was designed as "learning by doing with corrective feedback followed by more doing."

Students who attended all classes, which met one night a week for three hours over 15 weeks, automatically received a grade of "A" if they participated. There were no exams. The students took turns leading group sessions with the instructor conducting some sessions and participating as an ordinary member in all others. The instructor introduced theory and techniques when they promised to add value to student skill development and comprehension.

The important tasks involved becoming comfortable initiating group interactivity, reducing stage fright when serving as leader, learning to read meanings of non-defensive and supportive actions by members, and detecting inauthentic disclosures and actions motivated by anticipated social approval. It is in the nature of small voluntary purposive groups to provide unexpected learning about human nature. The course was billed as a laboratory of interpersonal learning, a facet of human nature.

The counseling course was experiential but was not intended as psychotherapy. A student assigned to conduct a session invented exercises, suggested topics or solicited topics of interest from others. The instructor informed students that they were not obligated to disclose intimate information about themselves but were obligated to be truthful if they did disclose. Invariably, there was considerable disclosure by members who believed they were indirectly benefitting other members.

This course appeared to provide therapeutic benefits to those who sought it. A 24 year old woman had moved away from her 'family two years before into her first apartment. She had been raped by an intruder when she was 23 and wanted to tell her parents but couldn't bring herself to do so. Informing her parents was not a group suggestion; she had wanted to do this since the rape. She had become very uncomfortable and ill at ease when visiting her parents and suffered chronic tension. She and her parents had previously been close. After telling the class about being raped, she was able to tell her mother who later helped her tell her father. She later reported that the closeness had been restored and her tension greatly diminished. She seemed more relaxed in class.

A frequent benefit experienced by members was a favorable change in beliefs about their likeability and competence from interaction with classmates. The instructor also invited members, if they wished, to set a goal and let the group monitor their progress and perhaps offer encouragement. A typical goal was to stop smoking cigarettes. Both smokers and non-smokers learned much about the near intractability of the addiction for many people. Other common goals related to the development of assertiveness to be applied in their real outside lives.

The net effect of this course was the development of genuine friendship and gratitude among classmates. This is a manifestation of social capital.

The Experiment

An attempt was made to induce subjects to participate in an exercise that often leads to anxiety, embarrassment, a fear of being judged unfavorably by others, and anticipatory shame at performing a task before others at which one has no particular skill. In one condition participation was induced, at least in part, by an offer of an incremental increase of a half letter grade in the final course grade. In the other condition, no tangible reinforcer was offered. The comparison was between the mean hedonic level of the participants under the contrasting conditions.

A typical psychology class had 30 students. The experimenter began the class by announcing that the class would be taking a pop quiz during that session. No exam had previously been given without warning.

They were told that any student completing it would automatically receive a quiz grade of A. This would have the effect of raising a participant's semester grade by half a letter grade relative to grades given to students in other sections of the course taught by different instructors. Failure to complete the "quiz" would carry no penalty; the nonparticipating student would be given a final semester grade based on the average of her other test scores. No one would receive a lower final semester grade for non-participation but participants were assured of a higher final grade than they would otherwise have earned.

All students initially judged this to be a no-lose proposition and had pen and paper ready. They were told that each question would require them to briefly record a description of their emotional state and a rating of their hedonic state (on a 10 point scale from very unpleasant (-5) through neutral (+/-0) to very pleasant (+5) when periodically asked to do so by the instructor. These were to begin with a description for the moments just prior to the announcement of the quiz; the second, just after the announcement.

At that point they were asked to arrange their desks and chairs in a circle. This elicited the first reports of anxiety (3rd hedonic evaluation). Students were then instructed to select an animal(s) that they admired and/or believed they resembled but not to reveal it to others (followed by a 4th hedonic evaluation).

These first and second year college students had learned from prior experience with college courses that academic work was almost always passive and anonymous, usually a lecture with some voluntary participation. They had expected a discussion, unchallenging and unthreatening. Most were stunned and terrified to be instructed to perform imitations of their chosen animals, complete with sights and sounds, in the center of the classroom, surrounded by classmates (5th hedonic evaluation). About 20% of the students were not alarmed but instead pleased. The remaining 80% were anxious; many terrified.

This exercise was performed six times with members of six separate general psychology classes (total n = 180), roughly evenly divided between females and males. Of the 144 who reported extreme discomfort, none left class. They were sitting in a circle so their departure would have been apparent. They were receiving a substantial reward in exchange for experiencing perhaps 20 to 60 minutes of anxiety. They may have hesitated to alienate the instructor, or they may have been watching each other to see if others were leaving.

Students at community colleges in Texas do not usually hesitate to protest instructor actions to administrators. A greater many are indifferent to their grade point averages. There was no way to predict whether there would be a mass exodus or a trickle of defectors followed by official complaints against the instructor.

They were coaxed to participate individually by the instructor. Most resisted strenuously. The instructor-experimenter was unrelenting in his solicitation to participate. His requests were made in a quiet, calm manner but it was clear that he wanted compliance. The "quiz" answers they submitted at the end of class revealed intense, rapid-onset anxiety initially, followed by great elation when the task was completed. After much coaxing and a growing awareness that classmates were slowly volunteering, however reluctantly, all finally participated.

Early in the process, when students protested, the instructor made clear by his manner that they had a right to state their objections, would be listened to and answered. Some asked what purpose the exercise served and were told that it would be explained afterward; it would lose its value if explained before. In each class someone asked if the teacher would participate (and thus share their "humiliation"); he agreed to do so.

As individuals volunteered and began the imitation, which required them to perform around the entire encircled class, their classmates began yelling out forms of good natured kidding. There was much laughter and the person completing an imitation usually reported feeling elated thereafter. When the instructor performed as an equal, he also became the object of good-natured jokes as the other participants had. During or after each imitation, classmates made guesses as to the identity of the animal imitated. Students followed these guesses with explanations for their selections and were listened to respectfully. This brief verbal recitation caused no discomfort. Many found it a relief to intellectualize.

The emotional-hedonic reports submitted as "quiz" papers reported an experience of elation that lasted from ten to twenty minutes upon completing the task. Participants enjoyed the joking, camaraderie and some noticed their classmates for the first time. There was a new sense of connection, playfulness and probably a slightly increased emotional closeness to other students. The experiment began to feel like a party to an individual who had completed her turn. The kidding was never hostile, there was no ridicule, and some tried to help others get through the near-trauma.

Each of the 180 subjects submitted descriptions of emotional states and hedonic values at eight standard points in time. The lowest scores tended to occur just prior to volunteering; the highest, after completion. Each subject had a mean score derived from their eight entries. The range of the 180 means was -4.0 + 4.5. The median of the 180 individual means was +1.5. The median score for the last of the eight entries by the 180 participants was +3.0.

When the same experiment was done with the counseling students (six classes, ten students in each), they felt less coerced, less anxious, and, on balance, less elated at the completion. They were less surprised and anxious so the emotional rebound of relief was less intense. They were more self-directed, less susceptible to the influence of leadership by the instructor and less in need of it to facilitate learning. They did participate readily and enjoyed it.

A mean hedonic score was calculated for each of the 60 participating counseling students. The median of mean hedonic scores for this group was +2.7, higher than +1.5 for the general psychology students. The median of the final (8th) hedonic score (at the experiment's conclusion) was +3.0 for both groups.

The experiment probably seemed unexceptional to the counseling students since they sometimes initiated novel and creative group processes themselves. They had voluntarily taken the course and such exercises were considered part of the course. At this point in their course, they knew their classmates fairly well. Their responses suggest that trust had developed.

The psychology students probably felt coerced because they were being coerced. They had enrolled in a lecture course and were being "paid" to play games. The fact that the instructor and class atmosphere was not experienced as authoritarian may have mitigated the perception of being coerced. Only a few students expressed any anger at the end of the process.

The final 100% participation probably resulted from a predisposition of some to comply with the instructor's wishes, particularly since they had no warning time in which to plan resistance. Some may have been encouraged by seeing other students participate. Since the instructor was willing to participate, the process may have seemed less authoritarian than it might otherwise have. Also, some students gently encouraged others. Some reluctant participants lost their fear rapidly and others could see this more relaxed demeanor. Perhaps most important, the reward in terms of grades was quite large for the effort expended.

Conclusion

The pedagogic experience was intended to demonstrate the power of social approval and disapproval and the intense aversion most participants felt at the prospect of looking foolish and being punished internally by the experience of shame and embarrassment. It was clearly important to a majority that they not perform an act in which they had no proven skill. A minority, however, saw it as fun. These probably thrived on attention, were generally rewarded for it, or were simply not susceptible to shame. Probably there were no obvious social norms for guidance and unplanned public scrutiny may be aversive for most people.

The counseling students, on average, felt some discomfort but did not interpret this as a signal to quit. They were also well protected by emotional bonds sometimes known as social capital. Intrinsic satisfaction was sufficient to justify participation for the counseling students.

The psychology classes required a tangible reward to participate. The experimenter had to play the system to use higher than usual grades, something only the grading system and the college credit mechanism could bestow. The experimenter simply awarded the higher grades without consulting school officials.

Summary

The development of social capital in the instructional process was effective in producing desired outcomes in the domains of government and politics, macroeconomics and psychology. The interactive courses required directed and focused activity to achieve the intended results and atomized classes could not produce meaningful learning except, perhaps, though a massive increase in resources expended.

References

Cameron, N. (1997). Teaching tools: Simulating money supply creation in class. Economic Inquiry, 35 July;1 686-693.

Fukiyama, F. (1995). Trust. NY: Free Press.

Hosen R. (1978). Increasing political participation through instruction. Teaching Political Science, 5 (3), 349-356.

Hosen, R. (1993). A psychology of happiness and well-being and its relation to the psychology of addiction and dependence. Psychology: A Journal of Human Behavior, 30 (2).

Hosen, R., Stern, L., and Solovey-Hosen, D. (2001). The complexity of motivated human choices that affect psychological well-being. Psychology and Education, 38, 314, 3-27.

Hosen, R., Stern, L. and Solovey-Hosen, D. (in press). Proactive risk-taking promotes happiness and subjective well-being. Psychology and Education.

Putnam, R. (2000). Bowling alone. NY: Simon & Schuster.

Ron Hosen, EdD, Instructor at Lee College when experiments were performed. Dina Solovey-Hosen, MSPH, Online Teacher, Virtual School, Houston Independent School District.

Correspondence concerning this article should be addressed to Dina Solovey-Hosen, 5909 CAREW, Houston, TX 77074