DATE: March, 1990
EDITION: Volume VI, Number 2
David Trimble, Editor
Jodie Kliman, Associate Editor
CONTENTS
-Mansell Pattison died in September, after two years in a coma following an automobile accident. He is best known among network therapists for his research on networks and mental health, for which he developed the Pattison Psychosocial Kinship Inventory, still in fairly widespread use in research on social networks and mental health. Before he became more involved in academic administration, he also conducted network interventions in clinical practice, favoring the smaller scale meeting and consultation to network members. Dr. Pattison's last position was as Chair of the Department of Psychiatry and Health at the Medical College of Georgia. We have lost an important champion of network therapy, a clinician familiar with our practice who was also very productive in research on social networks and mental health.
-In the last NETLETTER, I mentioned a group from Connecticut who attended the workshop at Ortho. They are the Network Therapy Project from Norwich, Connecticut, directed by Art Wood and including Linda Anderson, Carol Dooley, Dru Hagar, Nancy Mosher, and Nancy Rossi. It is an interagency effort including William W. Backus Hospital, Community Mental Health Services of Southeastern Connecticut, the Connecticut Department of Mental Health, Norwich Hospital, Reliance House, and United Community Services. I remember expressing some nostalgic yearning for such a "barter" exchange of labor among agencies, which was the model for work I did in the 1970's in Lynn, Massachusetts. In a note this summer, Art Wood spoke of the group "still trying to get out from under the unfunded model we started with." Ah, the challenge of transition from informal to formal organization! Art reports there is a possibility of a one year grant, and that the local Department of Mental Health Regional Director invited the team to the DMH Commissioner's Office to speak with all the other Regional Directors in the state about their project. On June 20, 1989, the team convened an interagency professional training meeting to acquaint local colleagues with the network approach. Larry Ruhf from Mount Tom and I were invited to present. Larry did a role-play simulated assembly using the Mount Tom model; I took notes and commented from the Speck-Attneave perspective. A brief report on the simulation follows in the VIEWS section of this issue.
-Shirley Rodecker has moved on from her position with the Neighborhood Parenting Support Project in Winnipeg; she is looking for opportunities to apply the network model in other domains of social work. Julia Halevy, in their neighborhood for other purposes, dropped in to visit the project in August. The Winnipeg folks have produced some progress reports, which are noted in CORRESPONDENCE below.
-The Stockholm Emergency Network Centre is up and running; Eva Bergerhed sent along an English-language copy of their publicity release, which is included as the final page of this issue.
-The Stockholm folks and Ross Speck presented their approaches to network therapy at an international conference in Brussels, "At the Frontiers of Family Therapy," organized by Mony Elkaim through the Institute for Family and Human System Studies, in May, 1989.
-Julia Halevy and the Mount Tom team are scheduled to present at next year's American Association for Marital and Family Therapy conference. Kathryn Kaminsky tells me that they will probably combine a theoretical presentation with a role play.
-American Orthopsychiatric Association has approved two presentations for their 1990 conference (April 25 -29, in Miami, Florida). On April 26, Gunnar Forsberg is moderating a workshop including the Stockholm group and Larry Ruhf from Mount Tom, who will provide a role play. I will be chairing a panel on "key ideas" in network therapy on April 27. Ortho was quite strict this year in limiting the number of panel members; I regret that I could not have included more people on my panel, which includes Anne Coppard (COTA, Toronto), Kathryn Kaminsky (Mount Tom), and Don Fuchs (Winnipeg's Neighborhood Parenting Support Project). Literally the night before printing of this issue of NETLETTER, I got word from Live Fyrand, who had planned to be on the panel, that she has just been funded for a 3 - year research program on the relationship between patients with chronic somatic disease and their social network. These new responsibilities have made it impossible for her to attend; she has suggested a couple of her Norwegian colleagues as potential replacements. They are Bibbi Bjellan, who works with a residential psychiatric clinic for adolescents, and Ellen Dalland, who runs an aftercare institution for adult psychiatric patients. I have reserved the Champagne Room at the convention hotel (The Fontainebleau) for an informal gathering Thursday, April 26, from 5 to 7 PM. It will be listed in the program as follows: "NETWORK THERAPY GATHERING We invite people interested in social network intervention in mental health to assemble informally with conference presenters on network therapy for social exchange and 'networking.'" I hope to see many NETLETTER readers there.
-Mary Youngquist in Minneapolis-Saint Paul reports that she plans to work on the draft article which appeared in NETLETTER (III,(1), 1988) for publication. She will be collaborating with Dennis Jaffe, an old network therapy hand who has done work with Ross Speck. She reports that Paul Ruff has moved to Kodiak Island, Alaska for two or three years' work at a Native American mental health center.
-The Boston Globe (August 3, 1989) ran an interesting news item, reporting on the work of Kirk Williams and Richard Hawkins at The University of New Hampshire. In a study of spouse abuse, they reportedly found that "social ties and standing have an effect on whether men abuse their wives. Nonabusive men tend to have lived longer in their communities... and that bond makes them think it more likely police will be called if they assault their wives. 'Consequently, men who spend more time socializing with friends, doing special things with their partners and becoming involved in church and civic affairs, should be nonassaultive, compared to men who are isolated,' Williams explained."
-Ann Coppard was planning to visit her native England this past summer, and was arranging to meet with Clyde Mitchell, one of the original social network theorists. She planned to share back issues of the NETLETTER with him; it would be delightful if they provoke some response.
-Larry Ruhf's conduct of a simulated full-scale assembly at Norwich (see NEWS) gave me my first opportunity to look at a Mount Tom style 3-stage full-scale assembly through the eyes of a Speck/Attneave six-stage "spiral model" practitioner. The human service workers at the training meeting took the roles of Barbara, a 30 year old borderline patient who had been a client in a service program for ten years, the director of the program, who had been Barbara'a original therapist, Barbara's roommate, her case manager, a program nurse, Barbara'a current therapist, Barbara's brother(who dates her roommate), Barbara's roommate's mother, each of Barbara's divorced and remarried parents with their new spouses, Barbara's sister, and a couple of Barbara's friends. Barbara, who could be quite high- functioning, and was very effective caring for children as a nanny, was regressing as she faced the prospect of moving from her current sheltered residence to an apartment with her roommate (who had moved out shortly before). Larry framed the meeting as a "commencement/transition" meeting, started out by asking each person in turn to say who s/he was and to say "one thing you think is crucial to talk about today." This opening was effective in providing expression for the emotions of hope, pride, frustration, and fear with which people greeted the impending transition, and elucidated the patterns of coalition and conflict within the family and among family and service providers and friends. As people exhorted Barbara to continue her progress by moving into a more independent setting, she expressed more and more resistance to the "abrupt" process, and complained that she was not getting what she wanted in living accomodations. Larry pressed for everyone to "talk about worries," interrupting after a few people talked to say "This sounds too good to me, too polite. Let's get out unvoiced feelings and concerns." After a time of people expressing their concerns about failure, sabotage, loss of support, or trouble trusting Barbara with responsibility, Larry then then asked people to start talking about Barbara's strengths. Some of the people who had expressed the most worry joined others in praising Barbara's honesty, energy, reliability, motivation ("until recently," qualified her therapist), sense of humor, caring, and friendship. By this time, it appeared to me that the initial obstacles of personal reserve, competition, covert and overt conflict, insecurity, and burnout had become surmountable, as the group achieved cohesion through so much communication of intense feelings about Barbara and her situation. Larry began to make some gentle suggestions, noting that relationships with the family of origin seemed distant, and might become less so; that Barbara needed a lot of support at this difficult moment of her life. He said, "Now it's time for people to say, honestly, what you have to offer her." This challenge evoked a somewhat defensive response from the service workers, who began to remind Barbara that she had to help them to help her by complying with treatment. Larry said, " I can understand generating problems to get human contact; how do we deal with this, group?" People, including the formerly distant mother and sister, began to come up with concrete suggestions for action, and the group moved toward consensus on a plan for intensifying support from all sectors of her network as Barbara made the move. Larry proposed that the group convene again after Barbara moved in to her new apartment, and reassured her of his continued availability as a network conductor, stating, "We're not done until you fire us."
I find as I write this summary of the meeting that I can fairly readily articulate the stages of the spiral model in my narrative of the meeting, with evidence of familiar moments of retribalization, polarization, mobilization, depression, and breakthrough. This was not my experience as an observer at the time of the assembly; I was struck then by how productive it was to help keep the group process shifting between the information-processing, problem-solving mode on the one hand and the communication and sharing of emotional experience on the other. The conductor's stance was certainly familiar to me; Larry was clearly respectful of the group's abilities to solve their own problems and heal their wounds, while very active in interrupting and redirecting the group process to mobilize those abilities. The sequence of events showed a shifting from collective and individual rigidity and frustration to hope, support, and practical action. Without using the spiral sequence model as a conscious plan of action, Larry's guidance of the group back and forth between problem-solving and emotional sharing liberated the creativity of the group and its members, a process which Ross Speck and Carolyn Attneave described in Family Networks as "shifting the frame of reference."
In his introduction to the role-play, Larry elaborated on the Connecting phase of Mount Tom's three-phase model of Convening, Connecting, and Shifting the Locus of Responsibility. Uncovering a shared emotional experience facilitates empathy, and generates the impulse to help. Larry urged people to resist the impulse to move directly into a problem-solving stance. "Sadness is a tremendous connecting experience;" other emotions such as guilt, remorse and anger are also connecting.
My experience with the simulation leaves me assured of the continuity between the full-scale assembly as now practiced at Mount Tom and the full-scale assembly pioneered by Speck and Attneave. I'm fascinated to discover that the spiral sequence model, while not part of anyone's conscious process at the simulation, can still fit my narrative summary fairly well. Naturally, I am a very biased observer, but I think there is more going on here. The size of the assembled group was under twenty. The meeting was clearly a "full-scale assembly," as illustrated by my observations above, and not a "partial network assembly" or network session on the Garrison model. Given the size of the group, I think an experienced team could manage it easily by being conscious of the need to shift between the problem-solving and and emotional sharing modes of group process. Once the size of the meeting starts to creep above thirty, however, things get a lot more complicated. Side conversations get going, and the meeting sometimes seems more like a three-ring circus than a town meeting. The team gets caught up in a very powerful induction process, more powerful, I think, the larger the assembled group. In such circumstances, I have often found that my teammates and I wouldn't know what the hell to do without having the more complex six-stage spiral sequence model as a map to guide us.
In practical terms, I think that the compleat team for a full-scale network assembly should be careful to convene a balanced network, and be sure to be familiar with both the Mount Tom and the Speck and Attneave models for the group process. Whenever possible, they should use the Mount Tom model, always ready to shift their frame of reference to the six-stage model when the need arises.
I am very grateful to the Norwich Network Therapy Project and to Larry Ruhf for this opportunity to confirm with my experience the essential continuity between Speck and Attneave's pioneering work and Mount Tom's current work in the practice of full-scale network assembly.
- Salzinger, S., Antrobus, J., Hammer, M. (Eds.). (1988). Social networks of children, adolescents, and college students. Hillsdale, N. J.: Lawrence Erlbaum Associates.
This is a collection of papers originally presented at the February, 1985, Conference on Social Connections from Crib to College, convened at the City College of New York. They are research reports, for the most part correlational studies, with the exception of two studies of Haitian immigrants to the U.S. and of Yoruba children in Nigeria, which are more ethnographic and use descriptive statistics. The studies were aimed toward the scientific understanding of social networks throughout the life cycle. As a clinician, I often had difficulty pulling out information which could be useful for my work. I admire the quality of the scientific work; the authors showed strong imagination in posing questions suited to the relatively primitive state of research methods for the study of social networks, and in devising better methods. The volume showed a high standard of care in the clarification, qualification, and interpretation of research findings.
The network characteristics measured in most of the studies reflected only the most basic concepts from social network theory, e.g., size, composition, number and frequency of contacts, type of activity. One study looked at clusters, and two others at density. There were no measurements of directionality, identification of nexus nodes, or differentiation of network members by degree of importance, strength of relationship (either by intensity of feeling, by durability of ties, or by supported vs. unsupported ties), or patterns of ties among network members. How big is a personal network? It depends on how you ask the question. Some studies excluded family or household members; some included them. Some restricted the definition of network to self-reported close friends. Less obvious differences in method of observation affected the size of the measured network; without one standard operational definition (and some way of matching the samples studied for population characteristics) it isn't possible from these studies to draw any conclusions about changes in network size across the life cycle. None of the network sizes reported were anywhere near as large as the networks assembled in an old-fashioned ful-scale network assembly, nor as large as those conceptualized in the the models of Barnes, Boissevain, or Mitchell.
I did encounter a couple of useful terms for network characteristics in Ladd et al.'s paper. Extensivity is "the degree to which contacts occur in large as opposed to small groups". Dispersion refers to the geographic location of network members. I was also struck by a relatively simple measure of density used by Antrobus, Dobbelaer, and Salzinger in a study of the close friendship networks of college students. It appeared a straightforward way to get data from more than one informant about the structural properties of a personal network. After Ego named a set of close friends, each of those close friends was asked to name his/her own close friends (second-order network of Ego). A simple count of the number of coincidences between Ego's choices and the choices made by Ego's friends provided a measure of the interconnections in these small networks.
I have struggled for a while over the question of how to review this work. I spent some time preparing notes for a detailed review, citing each particular study by authors and title, and critiquing methods and results before discussing the implications of each set of findings. I have had to conclude that such a project would be too daunting for the reader of a newsletter aimed primarily at clinical practitioners. My apologies to the book's editors and contributors, and to the NETLETTER reader, therefore, for the absence of specific citations, and for a simplification of the studies' findings certainly not intended by their authors. I have tried to pull out what would be interesting or useful for the clinician. The reader should use the information cautiously, primarily as a source of hypotheses in his/her clinical work. Those wanting to make more of it are directed to the original source.
Salzinger and Hampson, and Hampson in a second study, examined the networks of toddlers, together with the networks of their mothers. Their primary concern was the relationship between network characteristics and language development. Current work on language acquisition has departed from the older models of an invariant developmental sequence to consideration of different courses of development. Some children are stronger in the development of referential or "personal" speech, which emphasizes naming and describing objects and events. Such speech tends to be more comprehensible and clearly articulated than another main form of speech, "expressive" or socially directed speech, which tends to be more for expressing needs or controlling other people. "Personal" speech contains a high proportion of nouns, names for things. "Expressive" speech is often in the form of short phrases, at first undifferentiated, e.g., "Gimme", "Pickmeup". In the most general terms, the more a toddler's network is shared with other toddlers, the more her/his speech will tend to the expressive; the more adults in the network, the more personal the speech. The greater the number of adults in the network, and the more the contacts with adults, the more comprehensible the child's speech; more children in the network and more contacts with children correlate with more socially functional speech. These trends also apply to the proportion, as well as the number, of contacts with adults or children. Children with fewer adults in their networks, or fewer contacts with adults, or smaller proportions of adults in their networks, ask fewer questions about objects.
Ladd et al. studied older and younger groups of preschoolers, specifically studying the nonschool networks of these subjects. They came up with the rather puzzling finding that the number of separate playgroups [clusters?] in a child's network correlated negatively with nursery school adjustment for younger children (23-40 month) but positively for older children (41-55 month). Children both of whose parents were employed had larger peer networks than did other children, probably reflecting their time spent in daycare. I was more interested in some of the studies Ladd et al. reviewed, particularly in the insight they gave into the problem of social marginality for preschoolers. Marginal or rejected children's playground behavior reveals tendencies to play with younger children, with unpopular children, or in smaller groups. Children who started the school year pursuing more extensive peer contacts at the beginning of the school year were more likely to be rated either as liked or as disliked by their peers later in the year. Well-liked kids reduced the range of their peer contacts over the year, reflecting selectiveness in the choice of companions and stability of these relationships; disliked kids extended the range of contacts over the year, reflecting difficulty forming and maintaining specific relationships. Popular children had denser play groups, i.e., they were more likely to name each other as friends. For boys, there is a positive correlation between social maturity and tendency to play in large group settings; for girls, social maturity correlates positively with tendency to play in dyadic or small-group settings.
Feiring and Lewis compared 3 and 6 year old boys and girls from middle and upper-middle class families. Children from both age groups had more females than males in their networks, although boys had a higher proportion of male network members than did females. The proportion of contacts with same- sex peers increased with age. By the age of three, all children had a higher proportion of nonkin than kin in their networks. Three year old girls had a higher proportion of kin than did three year old boys. The proportion of kin decreased with age for girls, not for boys. Middle-class children had a higher proportion of family than nonfamily contacts; this difference from upper-middle-class children decreased with age. Both middle- and upper-middle-class groups showed in increase with age in proportion of daily contacts with peers; the middle-class children started out at age 3 with a relatively lower proportion of daily contacts than the upper middle class children.
Cochran and Reilly interviewed the mothers of six year olds, comparing single- and two-parent African-American and white families. They compared social class as well; my reading of their method suggests that the class variable may have been confounded by the single- versus two-parent family variable. In addition to comparing the networks of these different groups, this study looked at the relationship between network characteristics and school performance.
Mother's education was the single strongest predictor of number and supportiveness of social ties. The authors raised the question that mother's education might also have been an important source of bias in a study based on mother's report of network characteristics.
Overall, the best predictor of good school performance was the number of relatives involved in task-oriented activities with the child. For single-parent white families, the greater the number of kin involved in outings, the better the child's school performance. This latter finding makes sense in light of some comparisons between subgroups. The networks of children from black single-parent families were dominated by kin, with the highest number of kin children among all four groups, despite the fact that single-parent black families had the smallest networks. This group had the smallest proportion of nonkin, and reported the fewest neighbors in their networks. In contrast to most of the two-parent black families who lived in fairly pleasant neighborhoods with single-family homes, the single-parent black families tended to live in crowded, unsafe neighborhoods. A larger number of cousins appeared to compensate for a relative dearth of neighborhood playmates. The proportion of nonkin was higher in the networks of single-parent white families than in the networks of single-parent black families. White single- parent families had more neighborhood ties than black single-parent families. Single-parent white families had networks which were 20% smaller than two- parent white families; this difference resulted primarily from a smaller number of kin. The difference by race in the proportion of kin and nonkin in the networks of children from single-parent families appears to reflect the subcultural differences between white and African-American attitudes toward single parenthood; black relatives tending to be more accepting and involved with such households, and white single parents more likely to be extruded from their extended families, forcing them to rely more on supportive ties with nonkin. In this context, the better school performance by children of single-parent white families who go on outings with their kinfolk may reflect the better integration of these households into their mothers' extended families.
There were some interesting contrasts between the two-parent and single-parent African- American family networks; contrasts which bear on the model of "fictive kinship" which Carol Stack employed in All our kin (1974). In Stack's study of northern urban families, she described a pattern of informal adoption of nonkin into extended family roles. In Cochran and Reilly's study, two-parent black family networks were larger, and had a substantially larger proportion of nonkin, than did the single-parent black families. The single-parent black family networks in fact had the highest proportion of kin among all four groups studied. In Cochran and Reilly's study, the two-parent black families were the African-American families which more closely fit the fictive kinship model. Children of two-parent black families were more likely than any of the other children to engage in task-oriented and outing activity with nonrelatives rather than with kin. "Our impression," stated the authors, "is that some of the adult friends in the networks of these children functioned like relatives in their interactions with them, and may in certain situations have taken the place of relatives for the children and their parents."
The study was conducted in the Syracuse, New York area, which had recently implemented a busing plan to achieve racial integration. Several findings illuminated the phenomena of racism and the effects of social policy efforts to attack racism. White children reported more school peers than black children, and black parents had fewer contacts with school than did black parents. Blacks were more likely to report whites in their networks than were whites to report blacks in theirs. Single-parent black families had fewer cross-race ties than two-parent black families. This is consistent with the larger proportion of nonkin ties in the two-parent black families, and with the apparent restrictions of social contact outside of extended family observed among the single-parent black families, most of whom lived in unsafe neighborhoods. Single-parent white families had more cross-race ties than two-parent white families. The single- parent white families were more likely to maintain social ties outside of their extended families, from whom they may have been extruded. It is possible also that these families, experiencing social stigmatization, would be less likely to avoid contact with blacks, for whom white racism produces a similar experience of perjorative social judgment by others.
McHale and Gamble studied the social networks of the mothers and older siblings of mentally retarded children ages 3 to 11 years; their older siblings were from 8 to 14 years old. These children's networks were not dissimilar from the networks of children with normal younger siblings. The children reported more peer support from their networks than did their mothers, who relied on their networks more than their children did for feelings of self-worth. The older siblings reported more contact with their families, and more support from their families (particularly grandparents, parents, and other siblings) than did the control group of children with normal siblings, but this increased family support was not at the expense of non-family contacts. Although the mothers of handicapped children tended to have more contacts with parents, in-laws and professionals than did mothers of normal children, they were similar to the control group in getting more contact and support from friends than from family. Particularly for the siblings of a handicapped family member, this study challenged the stereotype of family isolation associated with a family member's handicap. My own experience with the families of young children with autism and brain damage tells me that there may be a "threshold of burden," beyond which a family begins to lose many members of its support network.
The foregoing studies, the editors of this volume note, describe the networks of people who, as young children, have relatively limited control over their social networks. Many of their relationships are mediated by family, school, and other community institutions. As the person enters adolescence, s/he has increased capacity to shape and modify her/his own personal network.
Blyth and Traeger studied relationships between adolescent self-esteem and perceived relationships with parents and peers. They administered questionnaires to 1,617 white middle-class seventh through tenth grade respondents, who were living in intact, nuclear family households. The major dependent variable, global self-esteem, was measured by a modified version of the Rosenberg Self-Esteem scale. This scale was chosen partially because it did not include social relational items, thus avoiding a confounding of the dependent variable with the network variables. Network variables included the size of the set of "important" relationships reported by the subjects, frequency of contact and perceived intimacy with members of this set. Most remarkable about this massive study is the very small magnitude of the effects observed. For both males and females, only 6% of the variance in the data was explained by the factors studied. For males, frequency of contact with a close friend and intimacy with parents counted most for self-esteem; for females, intimacy with parents (particularly mother) counted most. The somewhat stronger findings for males may reflect the greater variability in this gender; girls tended to measure consistently higher on the relational dimensions. In trying to account for their relatively weak findings, the authors suggested that, by choosing a measure of self-esteem with no relational items, they were perhaps not getting a strong enough measure of that variable. My impression is that the network variables themselves were too simple. The study essentially looked only at size, frequency of contact, and intensity (perceived intimacy). The need for simplicity is understandable for such a massive study, but a smaller study might have been more productive by examining the pattern of interconnections in the respondents' networks, the number and size of clusters, and the positions of these clusters in the social ecology of extended family, neighborhood, and school system.
Vondra and Garbarino's study of social influences on adolescent behavioral problems also proved disappointing. Measuring network size, percentage of kin, social competence, behavioral problems, and social functioning in 10 to 16 year olds, they hoped to find support for a model in which family functioning affects social competence, which in turn affects mental health. They found no relationship between their family cohesion model and any measure of social functioning. Social competence did not predict behavioral problems. Social functioning was predictable on the basis of perceived parental support. They took their finding that acting-out behavior was related to parental support to be weak evidence at best for a social network model in which lack of parental support leads to a stronger peer orientation, which in turn produces less acceptance of social norms and more antisocial behavior. There was a scattering of interesting discrete findings. For younger (10-13) but not for older (14-16) children, parental support correlated with network size. Only the younger girls had more kin than nonkin in their networks; for both boys and girls, the proportion of kin in the network dropped with age. There was not much increase in size of network with age. For the older group only, the larger the number of kin in the network, the fewer behavioral problems (the proportion of kin still remaining smaller than the proportion of nonkin). In critiquing their effort, the authors advocated taking a more sophisticated view of social networks, looking at content as well as structure (e.g., individual differences in utilization of social support, norms within the subjects' reference group regarding social support).
Antrobus, Dobbelaer, and Salzinger interviewed City College of New York freshmen and their close friends, at the beginning of the school year and four months later. They were looking for relationships between network characteristics and academic success. Students' networks tended to change over time from homogeneity to greater racial/ethnic integration. With the exception of non-Hispanic white males, there was no relationship between close friend network size and academic performance. For white males, the larger the on-campus network, the lower the grade point average and the higher the dropout rate. This was the study, mentioned earlier, which took a look at density by interviewing the network members named by the original informant. Each of these close friends was asked to name his/her close friends. Density was measured by "reciprocity," or the number of coincidences in naming close friends among each such set of network members. The naming reciprocation rate was twice as high for women as for men. There was no relationship between grade point average and reciprocation per se. Large, dense close friendship networks (minimum of four persons and five common naming choices), were found primarily among African American students. There was a modest tendency for density to be associated with grade point average; the authors suggested that this finding may reflect the students' connections with pre-existing networks. The strongest meaningful finding was the positive correlation between the informant's academic performance and the academic performance of his/her close friends. The authors cautioned that the number of friends beyond the "close" range may be important in influencing academic performance; the mean number of on-campus friends was 2.3; mean number of off-campus friends was 2.7. They also suggested that it may take more than four months for a student to develop an effective on-campus network.
Culbert, Good, and Lachenmeyer studied network characteristics and social support in the first and second year at Fairleigh Dickinson University in northern New Jersey. They used the same Friendly Network Questionnaire as did Antrobus et al. Among first-year students, they found no relationships between grade point average and number of on- or off-campus friends, nor between GPA and type and number of activities. Using a measure of significant life events as indicators of stress, they found a negative correlation between the number of events and GPA. They took this finding to mean that the first- year students had not developed their networks sufficiently for them to act as support systems capable of buffering stress. This conclusion appeared borne out by findings on the second-year students, for whom there was no correlation between life events and GPA. The second-year students counted more friends than first-year students, with more friends on campus (although the proportion of off-campus friends remained larger for second- as well as for first-year students). For second-year students, the number of activities with on-campus friends correlated positively with GPA. For all students, there was a tendency for GPA to rise with the number of hours worked part-time. As in the Antrobus et al. study, the GPA of the informants correlated with the GPA of their close friends. There was no correlation between close relationships with faculty members and GPA; there was a negative correlation between close relationships with faculty and number of off-campus friends. Second-year students who reported more activities with fewer friends produced higher GPAs (a finding reminiscent of Ladd et al.'s finding, reported above, that more popular preschoolers reduced the range of their peer contacts over the year). Students who dropped out had the most on-campus friends, worked the fewest hours in part-time jobs, and had the lowest grades. Students who transferred out (mostly to state schools) had the highest first-semester GPAs, worked the most hours in part-time jobs, and had the fewest on-campus friends. The authors explained the transfer finding on the basis of "financial need," suggesting that the transfer students sustained a "high cost in work hours...[with]...low benefits in on-campus networks." As a former resident of northern New Jersey who remembers Fairleigh Dickinson as a very low-prestige college (it was universally nicknamed "Fairly Ridiculous"), I can come up with another explanation. The transfer students seemed the brighter and more industrious, and retained their primary reference group among their off-campus friends, thus facilitating their move to more prestigious (albeit public) colleges.
Gutwirth-Winston studied the networks of American-born children of Haitian immigrants. Her methods included reviews of the ethnographic literature, interviews of patients at a New Jersey prenatal clinic, and some participant observation. Much of the report is in the form of a "case description" formed out of a composite of a number of families she observed. The Haitian-American child is "never just a child of one mother and father, but, rather, the child of an entire set of kin." He/she may spend no more time with mother than with others in the extended family. A typical childhood will be spend in a series of households, which themselves are constantly changing composition. Children with such a pattern of life require socialization to a high level of social competence, as they must be able to adapt easily to a wide range of situations, surroundings, and relationships. Gutwirth-Winston raises the question as to whether or not this life pattern reflects a cultural history of poverty originating in Haiti, with its associated high rates of mortality of family members, and patterns of migration between rural and urban environments, which predate immigration to the United States for many generations. An alternative explanation would be that this family culture reflects Haitians' West African origins. The question appears another version of the "class versus race" controversy in understanding African American experience.
The volume ends with Hammer and Sutton's study of the social world of the Yoruba child, an excellent companion to the Gutwirth-Winston study. Their study shows a remarkable similarity to the study of Haitians, in that children live in a succession of households over childhood, in a family context which extends far beyond that of the European/American nuclear (or even close extended) family. The Yoruba child's social experience, with a large number of adults and children, inside and outside one's own household, including both kin and nonkin, demands social competence. Socialization emphasizes values of adaptation, cooperation, and integration. A child living with his/her parents usually has children from the extended family living in that home, and some of the child's siblings are often living in other households. As time passes, the child moves into other households, usually with parents' siblings. Migration is often for purposes of schooling, apprenticeship, work, or obligation to a relative. The ties made over the course of time are maintained regardless of one's place of residence. Hammer and Sutton describe such a system as "socializing children into a wide network of reciprocating kin and kinlike ties in which children acquire obligations to adults who care for them, while gaining a wider span of general and specialized culturally valued knowledge from their experiences." Eventually, individuals leave extended family households to set up their own marital households, which become their own centers for exchanging residence with extended family. The older one gets, the larger and more varied one's household. Yoruba society is quite diverse in education, class, and privilege; the structure of these extended family networks cuts across these dimensions. Hammer and Sutton's study provides strong support, in my opinion, for the argument that Haitian and other African-American patterns of social exchange ( as in Carol Stack's All Our Kin) represent continuity with African cultural tradition, and not just adaptation to poverty and oppression in the Western Hemisphere.
I hope that I have stimulated some readers to get their hands on this book. My next project for extensive review is Deborah Belle (Ed.)(1989). Children's social networks and social supports. (New York: John Wiley and Sons).
-From Anne Coppard (September, 1989) "...The COTA Social Network Therapy Team is now a multi-disciplinary one drawing from education, social work, and occupational therapy backgrounds. The mandate of the program has been expanded to include those clients with a diagnosis of manic depressive psychosis, schizo-affective disorder and major depression as well as those with chronic schizophrenia. "The agency works on an individual referral basis and is continuing to seek new ways to serve our clientele. A recent initiative has been of benefit to the Social Network Therapy Program. For those severely isolated, socially impaired clients who have little or no network, we first introduce a "Community Rehabilitation Worker," a registered nursing assistant who performs primary level case management functions. They relate to the client on a one to one basis developing trust and linking to appropriate resources in the community. These workers are supervised by SNT therapists. This two- pronged approach of the Social Network Therapists working with the network and CRW being an advocate for the client, seems an efficient motor for changes to take place in the network. "Program activities have included a number of half day seminars and a full day workshop at the annual conference of British Occupational Therapists in Glasgow. Presentations have also been made at the Canadian Occupational Therapists' annual conference. Dr. Ben Gottlieb and Ms. Susan James, a former member of the SNT team at COTA gave a full day pre-conference workshop on Social Network Therapy. "Dr. Don Wasylenski is pursuing his analysis of the data gathered at the beginning of COTA's Social Network Therapy Program. At the same time he is a major mover in the reorganization of the funding and delivery of mental health services in Ontario... ."
-Again from Anne Coppard (February, 1990)... "There does appear to be an increasing interest in this model of practice off the Continental United States. For example, I did a one day workshop to Occupational Therapists attending the British Association of Occupational Therapists Annual Meeting last June. Following this three Community Mental Health Workers came from England to COTA's two day training workshop in Social Network Therapy early this year. Workers who are part of the self help movement are also finding Social Network Therapy blends in their philosophy. For your interest we were very pleased to have two ORTHO Social Network Therapy Panel members make presentations at our workshop. Larry Ruhf did a hands on session on Crisis Intervention and Donald Lugtig presented the Neighborhood Parenting Support Network Project in Winnipeg. "We are looking forward to the visit of two professors from the University of the Netherlands who are on a Study Tour seeking hands on knowledge of how we are approaching the use of Social Network Therapy concepts. A meeting has been arranged with Ben Gottlieb, Don Wasylenski and COTA's Social Network Therapy Team at the end of March... ."
-The American Group Psychotherapy Association's annual conference was in Boston in February this year. Last fall, I discovered on reading the preliminary program the following offering: "Leading from behind: Activating the network toward self-development." I sent a letter to the presenter, David Kipper, Ph.D., a psychodrama practitioner at the University of Chicago, to see whether or not I had stumbled on a network therapist. His response, "...My scheduled workshop concerns the use of confrontations with the network within a 'protective' treatment environment. It is based on a psychodramatic role playing approach and uses confederates who assume the role(s) of the original network members rather than engaging the actual players as in the assembly procedure. The prime focus is on the client's own perception of his/her network; a scope far more modest thatn the one you described where the actual network members have an opportunity to change/grow as well. "The network assembly approach has been, indeed, used in clinical role playing and psychodramatic treatments. This time, however, the workshop was not designed as such. I might consider it in the future. Knowing more about your approach will definitely be helpful.
"P.S. : The 'simulated funeral' scene, including enactments of its aftermath, for the treatment of suicidal patients was suggested - under certain conditions - by J.L. Moreno, long ago. It can also be used to 'bury a group' (so that a new one can emerge)... ." I attended the workshop on February 22. David proved to be a very thoughtful, engaging, and humorous person, a psychologist with appointments at Bar-Ilan University in Israel as well as at Chicago. In his conceptual presentation, he utilized the distinction between weak(unsupported) and strong(supported) ties in a typology of group therapies and group leader styles. He sorted groups into four categories distinquished by their positions on two bipolar continua, "Active versus Passive" and "Leading from Behind (Indirect) versus Leading from In Front(Direct)." Active groups led from In Front would be exemplified by behavior therapy groups, in which members help each other to learn specific ways of changing and managing their behavior. A support group would exemplify a group led from In Front by a leader taking a Passive stance, often delegating much of the authority for leading the group to its members. Kipper claimed that the ties between group members in the groups led from In Front are strong ties, i.e., bonds are formed among the members so that every member is connected with every other member in a dense network fabric. He described the psychoanalytic therapy group as a group led from Behind by a leader in a Passive stance; the psychodrama group is led from Behind by an Active leader. These groups led from behind, he claimed, show weak ties between the members, presumably because most of the action is in work with the significant emotional ties with family members and others who are not present in the room. I presume that group members relate to each other primarily through transference, and thus do not form a web of ties among themselves. David packed a lot into a short time into this workshop, which did not let him elaborate some of these points, thus my presumptions about some of the points he made. I find it interesting to try to fit the Kipper model to the "transitional network" group therapy which Jodie Kliman and I wrote about in Wolman and Stricker's (1983) Handbook of marital and family therapy. As I have practiced it, this type of group therapy, which integrates psychodynamic therapy with network construction, does at times require the group leader to move from "Behind" to "In Front" when activity between members outside the group requires coaching, mediation, or arbitration. I am sure that David would have expanded more on his use of network concepts had the time allowed. He is now on the mailing list (See below, DIRECTORY UPDATE). I would welcome his elaborating in some future correspondence. Much of the workshop was a demonstration of the psychodramatic technique, in which we worked on an issue of personal growth presented by a volunteer from the audience. Other participants got to play the parts of members of the volunteer's network, doubles(people taking the volunteer`s role, providing alternative perspectives), inner obstacles, "advisory committee", etc. It struck me that psychodrama could be very useful in the repertoire of a network therapist, both for those people who seem unable to convene their actual networks, and to prepare people for network assemblies by enlarging their repertoire of potential responses to confrontations with network members.
-Ross Speck put me in touch with Ted Long, who sent me a letter (July,1989) describing his work (His address is below, in DIRECTORY UPDATE). "...In answer to your inquiry, I have a grant to assist teens in long term foster care, in learning independent living skills. As a component of this grant, I have included Networking Therapy, two sessions for each child -- at the beginning and end. I am studying how they perceive their families, in light of Structural Family Theory. Olson's Circumplex Model (Faces III) is the instrument. It was modified by David Cole at Notre Dame, and I am using his revision. "As it turns, the 'therapy' aspect of networking is minimal for this program. Without an appropriate crisis, it appears that foster children have a rather ambivalent 'support network.' I will be able to know more about the network at the end of the program. At that time, I will have administered another instrument, and have been able to compare them. I have also included a control or 'blind' by having the caseworkers in the program also fill out Faces III. "My primary orientation to family networking is what I term 'Structural Networking Therapy.' It is a combination of Structural Family Therapy Theory, Networking Concepts, and Crisis Intervention theory. Mostly, I am using it with teenagers who are involved with Satanic Cults, a growing and serious problem here. I plan to expand it for work with skinheads, witches, and religious extremism as time goes on. I also use it with children who are nearing re-unification in foster care. "Carolyn Attneave has agreed to serve as a consultant to our growing, but still small program. (I have a video of an informal rap session she graciously provided to a few of my staff. It's available should you ever wish to view it). "I am currently a PhD student at Purdue; Charles Figley (of Post Traumatic Stress notoriety) has been my advisor. My work with Vietnam Vets and other victims has led me into networking. Support systems are of great importance, post-trauma. "There is really a lot to discuss concerning networking therapy. I hope to continue my work and begin writing for publication by Spring. It would be good to get together and compare notes. My interests in trauma and resultant impact on the family's development would benefit from critical discussion... ."
-From the Mount Tom team, October 16, 1989 (Kathryn Kaminsky, Elizabeth Hemly, and Larry Ruhf): "...Things have finally settled down here; our hours are set and our job descriptions are cleared so that we can now move away from the immediacy of the change back to how we can continue to incorporate the Network Concept into all our work. "...The Mount Tom Network Project's funds were eliminated July 1. This change was in the most part dictated by financial cutbacks and secondarily by changes in the [program] administration. This has meant a loss of autonomy for our team, and limitations on the amount of time we can spend on research and theory development. That's the downside. The upside is that we now have an opportunity, as the Clinical Team for Mount Tom's Emergency Services Program, to integrate a network approach directly within the front lines of crisis intervention. What this means in practice is encouraging front line crisis workers to include the client's network from the very beginning; in short, to 'think, network. This means doing a lot of consultation with staff, and running mini-networking meetings, as well as preserving our capacity to do large scale network meetings as a team. So it feels like we have weathered a storm and come through, with some loss of autonomy, but a greater sense of integration into a larger system. We continue to use the three stage model of convening, connecting, and shifting the locus of responsibility in most of our meetings and find it to be very adaptable in helping all kinds of networks to move along. All of these changes are challenging us to expand our view of network interventions and to embed it deeper in our work. Kathryn will address our evolution on the Ortho panel in April. Kathryn, Larry, and Julia [Halevy] are presenting at AAMFT in October....[Liz] is planning to go to Ortho, as is Larry, who will be participating on the panel in the Swedish team's presentation.
-From Ross and Joan Speck, December, 1989: "...Uri [Rueveni}, Mira, Joan, Ross, and Rick Bailey did a large assembly (70+) on a Messiah and his principal disciple (a young Jewish Yale graduate in history of religions). It's too early to say if the symbiosis has been broken...."
-In September, 1989, Don Lugtig and Don Fuchs of Winnipeg's Neighborhood Parenting Support Project sent along their Fifth Quarterly Report (April 1 - June 30, 1989), together with a monograph (Series #06931, Child and Family Services Research Group, School of Social Work, University of Manitoba), "The prevention of child maltreatment in high risk multi-ethnic and multi-cultural neighborhoods." Their project is designed to "demonstrate the use of social network intervention at the neighborhood level as a preventive component of a neighborhood child and family service agency." It involves careful study of several urban neighborhoods, and intervention designed to assist parents to have more supportive networks and to assist neighborhoods to be more supportive to parents. The selection of a neighborhood as the unit of intervention may appear to run counter to network analytic wisdom about personal networks as "personal communities" which transcend the boundaries of the immediate locality (cf. the work of Barry Wellman). Given the nature of parenting, which tends to keep the poor or working-class parent housebound or restricted to the local store, school, and parks, their neighborhood-based network approach makes sense. Using secondary sources, a key informant survey, ethnographic study, and information from local service agencies, they have identified two inner-city neighborhoods, within which they have further identified high- and low-risk "micro-neighborhoods." The neighborhoods are remarkably ethnically diverse, including native (Anglo) Canadian, Cree, Chinese, Cambodian, East Indian, Italian, Filipino, Portuguese, and Laotian households. The first stage of intervention, already finished at the time of the Fifth Quarterly Report, was a Neighborhood Parent Support Survey, designed not only to gain information about individual parents' support systems, but also to conduct education and to identify parents who would be interested in further participation in the Project. They have been gratified by a remarkably high rate of expressions of interest in Project participation, from respondents who were selected on the basis of random sampling procedures. Respondents were asked to identify people in their networks "1)...with whom they can share their deepest thoughts, feelings, and concerns about parenting (confidant support); 2)...who visits with them while they are caring for their children (socialization support); 3)...who praises them about the way they care for their children (esteem support); 4)...who gives them advice about parenting (cognitive guidance); 5)...who criticizes them about the way they care for their children (cognitive and behavioral feedback); 6)...who sets an example of the kind of parent they should be (role model support); 7)...who will care for their children when they have an emergency and who can be relied upon most times to do so (tangible assistance or reliable instrumental support); 8)...who can be reliably counted on most times to look after their children for up to a half day when they need a break or have an appointment (milieu reliability, tangible assistance support); and 9)...who lends or gives them toys, clothes or equipment or provides tangible assistance support."
Network intervention strategies for parents and for neighborhoods include consulting, coaching, connecting, convening, and construction. Intervention with parents starts with the Identification stage, helping parents to recognize the potential for network building within their networks. In the Mapping stage, the parents help to construct a map revealing the strength of ties, frequency of contact, the basis of relationships(e.g., family, friend, neighbor, etc.), and the duration and intensity of the pattern of interconnections in the personal network. The Linking stage involves alteration of the personal network, including strengthening or loosening of ties, increase or decrease in network size, change in network composition and balance, or alteration of the size and density of clusters.
By the fall of 1989, the Project had already begun to locate natural helpers through information from parents at parent and child centers and from agency workers. I got a call from Don Lugtig recently; he reported that they have begun to collaborate with natural helpers who have volunteered to host small gatherings of neighbors in their homes.
At the close of their Quarterly Report, the Project staff report, "Acting on a suggestion by David Allen, we have initiated steps to form a national information network amongst persons and projects involved in social network intervention. We have contacted similar projects which use social network intervention to being information resource exchange." I hope to have more about this information resource exchange in the next NETLETTER!
-Thanks to Jodie Kliman for her editorial help with this and all issues of the NETLETTER!
-The alert NETLETTER reader has noticed that this issue, Volume IV, Number 2, was due out in the fall of 1989. I will do my best to produce two more issues this year, one in early summer which will include a report on the Ortho Conference, and another in the fall. Please send your correspondence, reviews, reports on your projects or conference presentations, short articles, etc. I hope that each NETLETTER will include more and more contributions from readers.
-Please note my address on the masthead of this issue. My last Directory listing gave my office address, as we were about to move at the time the last Directory was prepared. I fear that I may have lost some correspondence sent to my old (Saint Paul Street) address.
-I frequently get personal news from NETLETTER readers, either in correspondence or telephone or personal encounters. It has often not been clear whether or not the news was intended for NETLETTER circulation. My own Yankee/Scots-Irish ethnic bias has kept me from passing it along. If you do want me to pass along your personal news, please let me know.
-I continue to initiate new readers, acting on my own or on other readers' requests. Many of you have not yet paid the $5 yearly subscription or filled out a Directory listing. With the fall issue this year, I will be tightening up the subscription process. Everyone will receive a form requesting a Directory information update, and those who have not paid for three-year subscriptions will be billed for a subscription. If you want to receive NETLETTER in 1991, please be sure to subscribe.
-Ross Speck sends along word of the 40th Annual Meeting of the Canadian Psychiatric Association, in Toronto September 12, 13, and 14, 1990. Deadline for application is March 31; for applications write to Dr. Philip Beck, Chairman, Scientific Programme Committee, Institute of Community and Family Psychiatry, 4333 Cote Ste-Catherine, Montreal, Quebec, CANADA H3T 1E4.
-David Todd sent along some mail forwarded to him by the U.S. Post Office; it is apparently the contents of an envelope damaged in the mail. It came to him because one item was his address, in my handwriting, on a sheet of my stationery. Other contents include a couple of NETLETTER issues, one with Mansell Pattison's address highlighted in the Directory section, a couple of reprints of articles by Jodie Kliman and me, a reprint of Klovdahl's Note on images of networks from social networks, (1981), and some pencilled notes on some Hyatt hotel stationery. David Todd's accompanying note: "Our paths have not crossed for a long time- This is a bizarre way for it to happen now! This was sent to me in the enclosed envelope. As far as I can tell, it's yours, you were at Ortho and this folder got thrown in the mail by mistake!" It reminds of an early network analytic research method, "The small world problem." If this mail belongs to a reader, you can claim it from me.
-The last NETLETTER sent to Peter Farrel was returned. He is no longer at the address I have for him in Somerset, England. Can any reader help locate him?
USEFUL LINKS
This Site was Originally Created and Maintained by Ransom D. Trimble
and can now be viewed as a part of the Netletter archives.