Articles and Chapters

Many algorithms can find optimal bipartitions for various objectives including minimizing the maximum cluster diameter ("min-diameter"); these algorithms are often applied iteratively in top-down fashion to derive a partition P^k consisting of k clusters, with k > 2. Bottom-up agglomerative approaches are also commonly used to construct partitions, and we discuss these in terms of a worst-case performance for metric data sets. Our main contribution derives from a new restricted partition formulation that requires each cluster to be an interval of a given ordering of the objects being clustered. Dynamic programming can optimally split such an ordering into partition P^k for a large class of objectives that includes min-diameter. We explore a variety of ordering heuristics and show that our algorithm, when combined with an appropriate ordering heuristic, outperforms traditional algorithms on both random and non-random data sets.

Angle, J. 1996. How the Gamma Law of Income Distribution Appears Invariant Under Aggregation. Journal of Mathematical Sociology. 21(4): 325-58.

The Gamma Law of Income Distribution appears to be a scientific law because the gamma pdf 1) fits the grange of shapes seen in income distributions, 2) is parsimonious, 3) appears to be scale invariant, i.e., to show invariance under population aggregation, and 4) the gamma pdf's shape parameter provides a convenient descriptor of the range of shapes seen in income distributions, allowing the apparent invariance between education and the shape of the income distribution to be simply described. The Gamma Law of Income Distribution cannot, however, be a scientific law because it is not scale invariant. An unconditional distribution of income is a mixture, i.e., the weighted sum, of variously shaped income distributions. People at different education levels have differently shaped income distributions. These distributions are well fitted by gamma pdfs making the correspondent unconditional distribution a gamma shape mixture. A gamma shape mixture is not in general a gamma pdf. Aggregating the income distributions of population segments together can give rise to gamma shape mixtures. Thus the Gamma Law is not scale invariant. However, under certain conditions a gamma shape mixture can be hard to distinguish from GAM (*, l), the gamma pdf whose shape parameter is *, the weighted average of the _i's, the shape parameters of the component gamma pdfs of the mixture. GAM (*, l) has the same mean as the shape mixture. These conditions allow the Gamma Law of Income Distribution to appear to be scale invariant. These conditions occur in geographically defined populations in the contemporary U.S. They are 1) the distribution of income conditioned on education is itself gamma distributed, 2) is invariant under aggregation, 3) most of the population has attained an education whose corresponding income distribution is fitted by GAM (*, l) where _i > 1, 4) there is a close relationship between the shape of the income distribution and education, and 5) the distribution of people over education is approximately symmetric, unimodal, and peaked at its mode. The Gamma Law of (unconditional) Income Distribution appears to work because a Gamma Law of Income Conditioned on Education exists.

Banks, D. L., and K. M. Carley. 1996. Models for Network Evolution. Journal of Mathematical Sociology. 21(1-2): 173-96.

This paper describes mathematical models for network evolution when ties (edges) are directed and the node set is fixed. Each of these models implies a specific type of departure from the standard null binomial model. We provide statistical tests that, in keeping with these models, are sensitive to particular types of departures from the null. Each model (and associated test) discussed follows directly from one or more socio-cognitive theories about how individuals alter the colleagues with whom they are likely to interact. The models include triad completion models, metric models, and the constructural model. We find that many of these models, in their basic form, tend asymptotically towards an equilibrium distribution centered at the completely connected network (i.e., all individuals are equally likely to interact with all other individuals); a fact that can inhibit the development of satisfactory tests.

Barkey, K., and R. Van Rossem. 1997. Networks of Contention: Villages and Regional Structure in the Seventeenth-Century Ottoman Empire. American Journal of Sociology. March 102(5): 1345-82.

This article analyzes peasant contention in 17th-century Ottoman villages. The authors argue that peasant contention results from the position of the village in the regional structure, with village-level organization providing the means for contention The article uses court records to reconstruct the formal and informal networks within and across villages in western Anatolia. Under conditions of state and market expansion, those villages in intermediate positions in the regional structure tend to experience the vagaries of these changes more than central or isolated villages Those intermediate villages are also most prone to contention Cooperative village organization is also found to promote contention.

Barnes, G. R., P. B. Cerrito, and I. Levi. 1996. Algebraic Structure of the Interaction Semigroup as Related to the Homogeneity of Network. Journal of Mathematical Sociology. 21(4): 295-323.

This paper addresses the development of a semigroup model of social networks. Data matrices which represent the perceived relationships between members of a social network are use dot construct a (possibly infinite data semigroup of derived relations defined by (real) matrix multiplication. This complex structure is analyzed by forming interaction semigroups. These semigroups are homomorphic images of the data semigroup. The corresponding congruences are generated by identifying products of finite order which are highly positively correlated. Several methods of generating the interaction semigroups are examined and are shown to generate nonhomomorphic semigroups. For each congruence, an associated triple of numbers can be defined which may serve as an indicator of the validity and/or a measure of the stability of the semigroup model. A series of hypothetical examples is developed to study how the algebraic properties of interaction semigroups reflect and uncover properties of associated networks. Specifically, relationships between homogeneity of a network and the algebraic structure of the corresponding interaction semigroup are addressed. The applicability of the above techniques to blockmodels is demonstrated.

Batchelder, W. H., E. Kumbasar, and J. P. Boyd. 1997. Consensus Analysis of Three-Way Social Network Data. Journal of Mathematical Sociology. 22(1): 29-58.

Three-way social network data occurs when every actor in a social network generates a digraph of the entire network. This paper presents a statistical model based on cultural consensus analysis for aggregating these separate digraphs into a single consensus digraph. In addition, the model allows estimation of separate hit and false alarm rates for each actor that can vary within each actor in different regions of the digraph. Several standard signal detection models are used to interpret the hit and false alarm parameters in terms of knowledge and response bias. A published three-way data set by Kumbasar, Romney, and Batchelder (American Journal of Sociology, 1994) is analyzed, and the model reveals that both response bias and knowledge decrease with distance from ego.

Bearman, P. 1997. Generalized Exchange. American Journal of Sociology. March 102(5): 1383-415.

Generalized exchange, in which sections of a tribe exchange women in a cycle and thus guarantee social solidarity, was induced from models of the norms governing classificatory kinship systems. A block model analysis of one aboriginal tribe yields sections that serve as marriage classes in a generalized exchange system, though the norms that govern kinship would fail to manifest, if followed, a cycle for exchange. Generalized exchange systems emerge from inequalities exogenous to the kinship system, specifically gerontocracy. Models of norms are weak predictors of actual exchange structures. Models of relations yield insight into the etiology of systems that build social solidarity from social exchange.

Beggs, J. J., J. S. Hurlbert, and V. A. Haines. 1996. Community Attachment in a Rural Setting: A Refinement and Empirical Test of the Systemic Model. Rural Sociology. 61: 407-26.

As they examine the complex issues currently facing rural America, rural sociologists draw increasingly on studies of community attachment. Because this research tradition has established the superiority of the systemic model, recent studies in rural and urban settings have focused on the conceptualization and operationalization of its components. We introduce four operational refinements to this model, and we test our refined model with data from one geographic area in southwest Louisiana. We find that, although our operational refinements improve our understanding of community attachment, additional refinements are necessary. We conclude by exploring the implications of community attachment studies for attempts to revitalize community in rural settings.

Bonacich, P., and E. J. Bienenstock. 1997. Latent Classes in Exchange Networks: Sets of Positions with Common Interests. Journal of Mathematical Sociology. 22(1): 1-28.

While it is known that positions in exchange networks are interdependent, little work has been done to develop a systematic procedure to determine the type and pattern of these interdependencies. Wording from an insight by Schubik (1984), this paper proposes a canonical form for such interdependence. The model does not depend on the existence of pre-existing categories of complementary positions, such as buyers and sellers or men and women. The model uses only network position to create classes of positions with linked fates. Simulation and experimental results are presented that support this schema. These findings lay the groundwork for the study of collusion in exchange networks: members of latent classes, who share a common fate, would be likely to form coalitions when free communication exists.

Braun, N. 1996. Why Fix It When It Ain't Broken? A Reply to Henning's Critique. Journal of Mathematical Sociology. 21(4): 379-88.

This rejoinder discusses Henning's arguments against my model for exchange systems with restricted market access. it also presents an alternative rational choice model that combines Henning's central demands with my original modeling idea (according to which restricted access has effects before the optimization). This alternative approach yields similar conclusions as my earlier analysis. A comparison with Henning's results (which refer to a scenario in which access restrictions explicitly enter the optimization program) suggests that the respective model implications depend crucially on the decision where and how restricted access is introduced into a model. From this perspective, Henning's negative assessment of my work is questionable because, in the absence of empirical evidence, he has no better rationale for modeling restricted access than I do.

Bulder, Bert, F. Leeuw, and H. Flap. 1996. Networks and Evaluating Public-Sector Reforms. Evaluation. 3(July):261-276.

Surveys of client satisfaction and network analysis of the Dutch Ministry of the Interior and a semiindependent executive agency in the Netherlands were used to investigate relationships between social capital and government reforms. Results indicated that social networks greatly contribute to the productivity of individual employees as well as organizations. It is argued that well-intentioned reorganizations or reforms can turn social capital into "sour" capital, leading to a deterioration in efficiency, effectiveness, and customer satisfaction.

Cheng, R., and G. W. Milligan. 1996. Measuring the Influence of Individual Data Points in a Cluster Analysis. Journal of Classification. 13: 315-35.

The problems of measuring the impact of individual data points in a cluster analysis is examined. The purpose is to identify those data points that have an influence on the resulting cluster partitions. Influence of a single data point is considered present when different cluster partitions result from the removal of the element from the data set. The Hubert and Arabie (1985) corrected Rand index was used to provide numerical measures of influence of a data point. Simulated data sets consisting of a variety of cluster structures and error conditions were generated to validate the influence measures. The results showed that the measure of internal influence was 100% accurate in identifying those data elements exhibiting an influential effect. The nature of the influence, whether beneficial or detrimental to the clustering, can be evaluated with the use of the gamma and point-biserial statistics.

Church, A., and Reid Peter. 1996. Urban Power, International Networks and Competition: The Example of Cross-Border Cooperation. Urban Studies. October 33: 1297-318.

This paper reviews the military sociology and popular literature on soldier isolation during force-projection operations and the (tele)communication resources available and used to lessen feelings of isolation. During World War 11 and the Cold War era (Korea and Vietnam Wars), soldiers were isolated. Communication with significant others and the larger society was primarily one-way (e.g., via letter mail, Victory-mail, radio, movies, audio tapes, and tape delayed television). During the late-Cold War (e.g., invasion of Grenada, peacekeeping missions, and the invasion of Panama) and more recently during the post-Cold War era (e.g., the Persian Gulf War, Los Angeles riots, and the Somalia relief effort), the diffusion of one-way, and revolution in twoway communication media f e.g., telephones, video tapes, voice-mail, live television, and e-mail) resulted in a continued decrease in isolation for soldiers. Notwithstanding, there is a consistent and salient need among soldiers and families to continue to lessen isolation and separation. Although there has been a revolution in types of communication media (CM) available during force-projections, use across military operations has been evolutionary. In other words, high-tech has not displaced lowtech, but supplemented it. Recognizing that new and old CM are an ever-evolving and important element of the post-Cold War soldiers' experience, the authors stress the importance of social science research in ascertaining the social implications of regulating CM in military and other social institutions.

de Leeuw, J., and P. J. F. Groenen. 1997. Inverse Multidimensional Scaling. Journal of Classification. 14: 3-21.

For metric multidimensional scaling much attention is given to algorithms for computing the configuration for fixed dissimilarities. Here we study the inverse problem: what is the set of dissimilarity matrices that yield a given configuration as a stationary point? Characterizations of this set are given for stationary points, local minima, and for full-dimensional scaling. A method for computing the inverse map for stationary points is presented along with several examples.

Doreian, P., R. Kapuscinzki, D. Krackhardt, and J. Szczypula. 1996. A Brief History of Balance Through Time. Journal of Mathematical Sociology. 21(1-2): 113-31.

We present methods for establishing the amount of reciprocity, transitivity and group balance (a generalization of structural balance) in sociometric structures. These methods are applied to the second time series of sociometric data provided by the Newcomb (1961) study. The amount of reciprocity was above chance levels at the outset and showed no systematic variation thereafter. Transitivity has a very different time scale. it climbed steadily through the first nine weeks and remained stable thereafter. While consistent with chance at the beginning of the study period, it grew to above chance levels at week 3. Group imbalance declined throughout the entire period. The reasons for these different time scales are discussed.

Everett, M. G., and S. P. Borgatti. 1996. Exact Colorations of Graphs and Digraphs. Social Networks. 18: 319-31.

A coloration is an exact regular coloration if whenever two vertices are colored the same they have identically colored neighborhoods. For example, if one of the two vertices that are colored the same is connected to three yellow vertices, two white and red, then the other vertex is as well. Exact regular colorations have been discussed informally in the social network literature. However they have been part of the mathematical literature for sometime, though in a different format. We explore this concept in terms of social networks and illustrate some important results taken from the mathematical literature. In addition we show how the concept can be extended to ecological and perfect colorations, and discuss how the CATREGE algorithm can be extended to find the maximal exact regular coloration of a graph.

Feld, S. L. 1997. Structural Embeddedness and Stability of Interpersonal Relations. Social Networks. 19: 91-95.

The amount of structural embeddedness of a tie between two individuals is defined as the extent of overlap of social relations between those two individuals, and presumably reflects the extent of shared foci of activity that bring these individuals together with the same others. It is suggested that, in comparison with other strengths of ties (e.g. strengths of feelings, and frequencies of interactions), structural embeddedness: (I) is less under individuals' control, and therefore (2) tends to be more stable. Analyses of data on the relationships among male students at one small liberal arts college in November and April of their first year in college support the expectation that structural embeddedness is relatively stable, and tends to have some effect on other strengths of ties as well as the continuity of relationships.

Flache, A., and M. W. Macy. 1996. The Weakness of Strong Ties: Collective Action Failure in a Highly Cohesive Group. Journal of Mathematical Sociology. 21(1-2): 3-28.

Following Homans, exchange theorists have modeled informal social control as an exchange of peer approval for compliance with group obligations. The exchange model predicts higher compliance in cohesive networks with strong social ties. However, previous specifications failed to incorporate bilateral exchange of approval. Computer simulations using a Bush-Mosteller stochastic learning model show that bilateral exchanges evolve more readily than multilateral, causing social control to flow into the maintenance of interpersonal relationships at the expense of compliance with group obligations, a structural form of the "second-order free-rider problem."

Frank, Kenneth A. 1996. Mapping Interactions within and between Cohesive Subgroups. Social Networks. 18(2): 93-119.

The structure of interactions and the pattern of influence in an organization can be characterized in terms of a map of interactions within and between cohesive subgroups. I extend the work of Festinger, Schachter and Back (Social Problems in Informal Groups, 1950 Stanford University Press) who constructed a map based on the patterns of communication within and between apartment courts. In order to generalize Festinger et al.'s approach, I substitute a posteriori subgroups for Festinger et al.'s apartment courts, and I replace the distances of a physical geography with those of a metric multidimensional scaling. I apply the technique to data indicating professional discussions among teachers in a high school. After confirming that discussions are concentrated within a posteriori subgroups at a level that is unlikely to have occurred by chance alone, I construct a map of discussions within and between the cohesive subgroups. The map allows me to characterize the process of influence at the teacher and school levels through which the school responds to external conditions, and I argue that a map based on blocks of structurally similar actors does not sustain a comparable characterization.

Frey, C. U., and C. Rothlisberger. 1996. Social Support in Healthy Adolescents. Journal of Youth and Adolescence. 25(1): 17-31.

Social support is examined in a representative sample of 141 healthy adolescents. By means of a revised version of the Mannheim Interview on Social Support, the number, type, perceived adequacy (satisfaction), and quality (importance) of the social relationships available were assessed. While peers were found to provide prime supportive functions in day-to-day matters, the social support provided by parents has a stress-buffering effect in emergency situations. The role of other family members is discussed Differences in gender and education are moderate. The data suggests the adequacy of social support and social integration, contrary to the traditional view of adolescence as a time of crisis and conflict.

Gigone, D., and R. Hastie. 1997. The Impact of Information on Small Group Choice. Journal of Personal and Social Psychology. 72(1): 132-40.

The impact of information on an aggregate decision is directly related to the number of members of the aggregate who know the information prior to the group decision. This experiment obtains this common knowledge effect in small group choice. Group discussion affects group choice in some cases, signaled by group choices that "overturn" the choice of a majority of the group members. In those overturned majorities, moreover, the group tended to choose the correct option, leading group choices to be more accurate than member choices. Consistent with results comparing individual judgment and choice, groups pool information less thoroughly and rely on prominent items of information more heavily when choosing as compared with estimating.

Gil-Mendieta, J., and S. Schmidt. 1996. The Political Network in Mexico. Social Networks. 18: 355-81.

The most important political processes in Mexican politics including presidential succession since the 1920s have been conducted within a network whose political rationale has been political stability. All presidential elections have been won by a single political party. We analyzed the role of the network and presidential successions measuring significant relationships using the system UCINET IV. We contrasted computer distributions with historical facts. Applying the structural block model algorithm we found two well differentiated sub-networks, one representing a military-based group and the second a financial-based group. Measuring the network's centrality is one of the main objectives of network analysis for understanding concentrations of power and the distribution of influence in the political system. In this article we evaluate the maximum node and clique network index value concentration for the core of the Mexican network of power. Centrality and power indexes in the network are presented and their results are discussed in connection with cohesiveness.

Hall, E. J., and E. M. Cummings. 1997. The Effects of Marital and Parent-Child Conflicts on Other Family Members: Grandmothers and Grown Children. Family Relations. 46(2): 135-44.

Family conflict may have pervasive effects, but little is known of effects on family members outside of the immediate family environment of parents and minor children. Analogue methods were used to explore relations between specific conflict contexts and specific relationships outside of the immediate family. Grandmothers (N = 32) and grown children (young adult women, N = 28) responded to videotaped portrayals of conflict scenarios represented as occurring in their adult children's, or parents', families, respectively. While both groups reported experiencing negative reactions, grandmothers and grown children reported significantly different patterns of appraisal, emotional response, dispositions to intervene, expectancies, and predictions. These results suggest that the meaning and implications of family conflict are quite different depending upon an individual's role within the family.

Henning, C. H. C. A. 1996. A Critical Comment on Braun's "Restricted Access in Exchange Systems." Journal of Mathematical Sociology. 21(4): 369-77.

In general this comment tackles the problems and difficulties potentially combined with an application of formal economic models and constructs, such as the Walras equilibrium and microeconomic demand theory, to pure sociological contexts. In particular, this is done by analyzing a further attempt, as recently suggested by Braun (1993 and 1994), to extend the well-known Coleman Model by incorporating the embeddedness of social transactions in incomplete social network structures. "Pars pro toto" it is proved that Braun's conceptualization contains some weakness which imply that fundamental conclusions drawn in his article have to be revised.

Ishii-Kuntz. 1997. Intergenerational Relationships Among Chinese, Japanese, and Korean Americans. Family Relations. 46(1): 23-32.

Previous studies that have focused on Asian American intergenerational relationships used the cultural concept of filial obligation to explain an adult child's commitment to his/her elderly parents. Using data gathered from 628 Chinese, Japanese, and Korean Americans and their elderly parents, this study found that such financial and structural factors as an adult child's income and his/her parent's need for assistance significantly influenced the frequency of an adult child's support for his/her parents. The effects of filial obligation on an adult child's provision of support for his/her parents change depending on several of these financial and structural factors. The magnitude of these relationships varies, however, among three groups of Asian Americans. It is therefore necessary to examine interaction effects of cultural, financial, and structural factors on Asian American intergenerational relationships as well as the diversity of these relationships within the Asian American population.

Jackson, M. O., and A. Wolinsky. 1996. A strategic model of social and economic networks. Journal of Economic Theory. 71: 44-74.

We study the stability and efficiency of social and economic networks, when self-interested individuals can form or sever links. First, for two stylized models, we characterize the stable and efficient networks. There does not always exist a stable network that is efficient. Next, we show that this tension persists generally: to assure that there exists a stable network that is efficient, one is forced to allocate resources to nodes that are not responsible for any of the production. We characterize one such allocation rule: the equal split rule, and another rule that arises naturally from bargaining of the players.

Jehn, K. A., and P. P. Shah. 1997. Interpersonal Relationships and Task Performance: An Examination of Mediating Processes in Friendship and Acquaintance Groups. Journal of Personal and Social Psychology. 72(4): 775-90.

This study used multiple methods to examine group processes (information sharing, morale building, planning, critical evaluation, commitment, monitoring, and cooperation) that mediate the effect of relationship level on group performance. The study uses a 2 by 2 experimental design, crossing relationship (friendship vs. acquaintance) as a between-subjects variable and task type (decision making vs. motor) as a within-subject variable. Fifty-three 3-person groups participated in the study, and data from 4 types of measurement were used to analyze the mediating processes between relationship level and task performance. Friendship groups performed significantly better than acquaintance groups on both decision-making and motor tasks because of a greater degree of group commitment and cooperation. Critical evaluation and task monitoring also significantly increased decision-making performance, whereas positive communication mediated the relationship between friendship and motor task performance.

Kirke, D. M. 1996. Social Networks. 18: 333-46.

This paper outlines a method for collecting accurate 'peer' data by using two name generator and 13 name interpreter questions to collect 'friend' and 'pal' data from teenagers. The 'peer' data are then used to identify the complete network and to delineate the naturally existing peer networks in it. The findings suggest that the concept 'friend' should not be avoided when collecting 'friend' data.

Korenman, S., and S. C. Turner. 1996. Employment Contacts and Minority-White Wage Differences. Industrial Relations. 35(1): 106-22.

We study effects of job contacts on wages in inner-city Boston in 1989 and in the 1982 NLSY. Race/Hispanicity in wages are not explained by an absence of contacts among minority youths. Rather, in the Boston data, lower wages of black youths are explained by lower "returns" to their contacts. In the NLSY there is little evidence of lower return to black youths' contacts, but there is evidence of lower returns to Hispanic youths' contacts.

Larrick, R. P., and S. Blount. 1997. The Claiming Effect: Why Players are More Generous in Social Dilemmas Than in Ultimatum Games. Journal of Personal and Social Psychology.: 810-26.

The term procedural frames is introduced and defined as different representations of structurally equivalent allocation processes. Study 1 compared 2 well-known games, sequential social dilemmas and ultimatum bargaining, that share the same structure: Player 1 creates an allocation of a resource and Player 2 decides whether to allow it or deny it. Study 1 found that Player 1 made more favorable allocations and Player 2 accepted more unfavorable allocations in a social dilemma frame than in an equivalent ultimatum bargaining frame. Study 2 revealed the critical determinant was whether Player 2 had to respond to an allocation by accepting or rejecting it (as in the ultimatum game) or by making a claim (as in the social dilemma). Two additional studies explored how these actions are perceived. The inconsistency of behavior across procedural frames raises methodological concerns but illuminates construal processes that guide allocation.

Leenders, R. T. A. J. 1996. Evolution of Friendship and Best Friendship Choices. Journal of Mathematical Sociology. 21(1-2): 133-48.

It has been recognized in the literature that the mechanisms driving friendship choices differ when different settings are considered. At the same time, it is likely that different types of friendships are governed by different mechanisms. Employing longitudinal sociometric data from classrooms in elementary schools, it is examined whether gender similarity, reciprocity, and proximity (joint membership of study groups) have similar effects on 'friendship' and 'best friendship' choices. The results suggest that children use loose definitions of 'friendship', as opposed to their definition of 'best friendship'. The networks resulting from these different choices are found to evolve according to (partly) different mechanisms. This especially holds for the effect of gender similarity, which is profoundly predominant in the 'best friend' networks, but loses much of this importance when 'friends' are also considered. Also, 'best friend' choices are much more stable. Reciprocity of choices is found to primarily be a by-product of the preference to choose friends of the same gender, rather than being an important choice criterion of itself.

Leik, R. K., and M. A. Chalkley. 1997. On the Stability of Network Relations Under Stress. Social Networks. 19: 63-74.

This paper addresses possible sources of instability in network relations. Four distinct sources of observed low auto-correlation (unreliability of measurement, inherent instability, systemic change (from endogenous sources), and external change (from non-system sources)) arc discussed and a preliminary mathematical model is developed. Illustrations are provided from current family research and suggestions arc made for sorting out possible sources of instability. Even without a complete endogenous model we demonstrate differential impacts of the distinct sources of variability.

Li, R. M., and S. F. Newcomer. 1996. The Exclusion of Never-Married Women from Chinese Fertility Surveys. Studies in Family Planning. 27(3): 148-54.

Never-married women have been regularly excluded in official national surveys of fertility in China, even in light of evidence of increasing premarital sexual activity. Likewise, never-married women in the United States were consistently excluded from national fertility surveys prior to 19820 because of the perceived sensitivity of questions about contraceptive use and sexual activities. Data on sexual and fertility behavior from all women of reproductive age, regardless of marital status, can provide direct measures of sexual activity and unintended pregnancies, as well as facilitate modeling of social networks underlying the sexual transmission of diseases. China's need for such information, however sensitive, will become more difficult to ignore given increasing pressures to attend to the health need of their never-married but sexually active population.

Meulman, J. J. 1996. Fitting a Distance Model to Homogeneous Subsets of Variables: Points of View Analysis of Categorical Data. Journal of Classification. 13: 249-66.

An approach is presented for analyzing a heterogeneous set of categorical variables assumed to form a limited number of homogeneous subsets. The variables generate a particular set of proximities between the objects in the data matrix, and the objective to the analysis is to represent the objects in low-dimensional Euclidean spaces, where the distances approximate these proximities. A least squares loss function is minimized that involves three major components: a) the partitioning of the heterogeneous subsets; b) the optimal quantification of the categories of the variables, and c) the representation of the objects through multiple multidimensional scaling tasks performed simultaneously. An important aspect from an algorithmic point of view is in the use of majorization. The use of the procedure is demonstrated by a typical example of possible application, i.e., the analysis of categorical data obtained in a free-sort task. The results of points of view analysis are contrasted with a standard homogeneity analysis, and the stability is studied through a Jackknife analysis.

Morgan, D. L., M. B. Neal, and P. Carder. 1996. The Stability of Core and Peripheral Networks over Time. Social Networks. 19: 9-25.

This article uses data on 'network instability' to show how differences across multiple measurements on the membership of personal networks can yield important insights into the nature of these networks. The data come from a sample of 234 recent widows, aged 59-85, who completed seven interviews about their networks over a l-year period. We use these data to investigate the stability of both overall networks and individual ties, as well as the linkage between the stability of individual ties and the stability of the network's aggregate properties. We find that instability in these networks is best thought of in terms of an underlying 'core-periphery' structure, whereby some network members are likely to be named repeatedly (the core), while others are relatively unlikely to appear in any given elicitation of the network (the periphery). We explore the implications of this core-periphery structure for cross-sectional elicitations of personal networks.

Neto, O. A., and G. W. Cox. 1997. Electoral Institutions, Cleavage Structures, and the Number of Parties. American Journal of Political Science. January 41(1): 149-74.

Theory: A classic question in political science concerns what determines the number of parties that compete in a given polity. Broadly speaking, there are two approaches to answering this question, one that emphasizes the role of electoral laws in structuring coalitional incentives, and another that emphasizes the importance of preexisting social cleavages. In this paper, we view the number of parties as a product of the interaction between these two forces, following Powell (1982) and Ordeshook and Shvetsova (1994).

Hypothesis: The effective number of parties in a polity should be a multiplicative rather than an additive function of the permissiveness of the electoral system and the heterogeneity of the society.

Methods: Multiple regression on cross-sectional aggregate electoral statistics. Unlike previous studies, we (I) do not confine attention to developed democracies; (2) explicitly control for the influence of presidential elections, taking account of whether they are concurrent or nonconcurrent, and of the effective number of presidential candidates; and (3) also control for the presence and operation of upper tiers in legislative elections.

Results: The hypothesis is confirmed, both as regards the number of legislative parties and the number of presidential parties.

O'Toole, L. J. Jr. 1997. Implementing Public Innovations in Network Settings. Administration and Society. May 29(2): 115-38.

Practitioners and scholars have devoted considerable attention in recent years to initiating public innovations ­ to the relative neglect of how to ensure the implementation of such efforts. Executing innovations over the longer term, particularly in complex network settings, can be expected to be problematic. And yet networks are likely to be crucial institutional settings for the implementation of public innovations. The analytic approach of game theory, used heuristically, can identify a set of actions useful to public managers in enhancing prospects that sound innovations will succeed. The implications of this inquiry run counter to some of the themes used as mantras in the recent re-invention discussion and focus attention on the centrality of institutional infrastructure, trust, and obligation for innovative success into the future.

Peterson, R. S. 1997. A Directive Leadership Style in Group Decision Making Can Be Both Virtue and Vice: Evidence From Elite and Experimental Groups. Journal of Personal and Social Psychology. 72(5): 1107-21.

The group dynamics Q-sort was used to investigate the effects of leader directiveness in group decision making. Past research on leadership style has consistently implicated directive leaders as a chief cause of defective process and poor outcomes in group decision making. Leader directiveness was decomposed into 2 components: (a) outcome directiveness (i.e., the degree to which a leader advocates a favored solution) and (b) process directiveness (i.e., the degree to which a leader regulates the process by which the group reaches a decision). Process directiveness emerged as a potent predictor of quality of group process and outcomes. Outcome directiveness was associated with a much smaller and less coherent array of group outcomes. These findings suggest that current prescriptive models of decision making overemphasize the potential harmful effects of outcome directiveness.

Reed, Janet and Eric Dubow. 1997. Cognitive and Behavioral Predictors of Communication in Mother-Adolescent Dyads. Journal of Marriage and the Family. 59(1): 91-103.

This study examines the relation between mothers' and adolescents' perceptions of one another's behavior and their actual communication behavior during a problem-solving task. Forty mother-adolescent dyads completed self-report and observational measures of conflict and negative beliefs regarding the other. Hierarchical multiple regression analyses indicated that adolescents' negative beliefs about their mothers significantly contributed unique variance in predicting adolescent negative communication, beyond the effects of maternal communication behavior. Thus, treatment of parent-adolescent conflict should address adolescents' and parents' perceptions of one another, as well as their problem-solving behavior.

Ridgeway, C. L., and J. W. Balkwell. 1997. Group Processes and the Diffusion of Status Beliefs. Social Psychology Quarterly. 60(1).

How are consensual beliefs about the status-value of individual characteristics created in a society? A recent theory posits that inequalities in the distribution of resources in a population are translated into greater or lesser levels of consensus via social interaction in small groups. According to this theory a macro-structural correlation between resources and a distinguishable individual difference variable constrains who interacts with whom and governs the group dynamics of these encounters. It engenders certain belief-acquisition processes that create and spread status beliefs about the variable eventually making them consensual. We constructed a formal model of this diffusion process that includes the group interaction effects posited by the theory also the effects of group size and the unmediated impact of macro-structural conditions. Calculations based on this nest integrated formulation support most of the original theoretical analysis. In addition simulation results suggest the likelihood that two- to four-person groups are especially important as creators and spreaders of status beliefs supporting in a slightly modified fashion the earlier claim that group processes have the power to translate macro-structural constraints on actors into macro-level outcomes. These simulations also clarify several contingencies and other implications of the theory not fully apparent in the original formulation.

Roberts, J. M. J. 1996. Alternative Approaches to Correspondence Analysis of Sociomatrices. Journal of Mathematical Sociology. 21(4): 359-68.

If the usual correspondence analysis is viewed as a decomposition of departures from the model of independence, there are problems in applications to sociomatrices with regard to the treatment of the diagonal and cell estimates under independence. Other related techniques from the literature are more appropriate for analysis of sociomatrices. The different approaches are used to analyze a familiar sociomatrix, and the results of the techniques are compared.

Rogerson, P. A. 1997. Estimating the Size of Social Networks. Geographical Analysis. 29(1): 50-63.

It is notoriously difficult to define and estimate the size of individuals' social networks. Methods are suggested for estimating the size of such networks, using known rates of mobility together with survey data on the number of acquaintances of individuals who have recently moved. Special attention is given to the likely correlation between mobility rates and network size. Estimates are found to be consistent with previous estimates in the literature. The relation of this approach to multiplicity sampling and to "the small world problem" is also discussed.

Ruan, D., L. C. Freeman, X. Dai, Y. Pan, and W. Zhang. 1997. On the Changing Structure of Social Networks in Urban China. Social Networks. 19: 75-89.

This study is a replication of a survey on personal networks conducted 7 years earlier in Tianjin, China. Comparing the results of the two surveys reveals a large amount of change. Tianjin residents now report having more ties to friends and to associates beyond work and family, and fewer workplace ties and far fewer family ties. Women have gained on men in the number of friends, and young people have fewer workplace ties. These changes at the micro-level are examined in the light of changes in the Chinese macro-social structure.

Skvoretz, J., K. Faust, and T. J. Fararo. 1996. Social Structure, Networks, and E-State Structuralism Models. Journal of Mathematical Sociology. 21(1-2): 57-76.

The method of E-state structuralism provides dynamic models for the evolution and development of networks in small groups. Our interest lies in the kind of social networks that these models produce. We ask the question of whether such models produce "interesting" structure from a network point-of-view, in particular, from the perspective of Holland and Leinhardt who argue that any network that can be modeled adequately using only properties of nodes and dyads has no social structure. We show that E-state structuralism models are models of social structure in this technical sense because they assume a bystander mechanism in the creation of ties.

Skvoretz, J., and P. Zhang. 1997. Actor's Responses to Outcomes in Exchange Networks: The Process of Power Development. Sociological Perspectives. 40(2): 183-97.

Leading theories of power in exchange networks make assumptions about actors' reactions to being included in or excluded from exchanges. These assumptions, that actors consistently included increase their demands on others and that actors consistently excluded decrease their demands, provide the behavioral mechanism by which structural differences in position convert into power advantages or disadvantages. We test these assumptions with data gathered by experiment from five different networks. We find that while actors generally respond as assumed, the parameters of response to inclusion and to exclusion are not symmetric, depend on the level of experience of the subjects, and interact both with type of network (strong power versus weak power) and with network position.

Snijders, T. A. B. 1996. Stochastic Actor-Oriented Models for Network Change. Journal of Mathematical Sociology. 21(1-2): 149-72.

A class of models is proposed for longitudinal network data. These models are along the lines of methodological individualism: actors use heuristics to try to achieve their individual goals, subject to constraints. The current network structure is among these constraints. The models are continuous time Markov chain models that can be implemented as simulation models. They incorporate random change in addition to the purposeful change that follows from the actors' pursuit of their goals, and include parameters that must be estimated from observed data. Statistical methods are proposed for estimating and testing these models. These methods can also be used for parameter estimation for other simulation models. The statistical procedures are based on the method of moments, and use computer simulation to estimate the theoretical moments. The Robbins-Monro process is used to deal with the stochastic nature of the estimated theoretical moments. An example is given for Newcomb's fraternity data, using a model that expresses reciprocity and balance.

Snijders, T. A. B., and K. Nowicki. 1997. Estimation and Prediction for Stochastic Blockmodels for Graphs with Latent Block Structure. Journal of Classification. 14: 75-100.

A statistical approach to a posteriori blockmodeling for graphs is proposed. The model assumes that the vertices of the graph are partitioned into two unknown blocks and that the probability of an edge between two vertices depends only on the blocks to which they belong. Statistical procedures are derived for estimating the probabilities of edges and for predicting the block structure from observations of the edge pattern only. ML estimators can be computed using the EM algorithm, but this strategy is practical only for small graphs. A Bayesian estimator, based on Gibbs sampling, is proposed. This estimator is practical also for large graphs. When ML estimators are used, the block structure can be predicted based on predictive likelihood. When Gibbs sampling is used, the block structure can be predicted from posterior predictive probabilities. A slide result is that when the number of vertices tends to infinity while the probabilities remain constant, the block structure can be recovered correctly with probability tending to 1.

Stokman, F. N., and E. P. H. Zeggelink. 1996. Is Politics Power or Policy Oriented? A Comparative Analysis of Dynamic Access Models in Policy Networks. Journal of Mathematical Sociology. 21(1-2): 77-111.

In policy networks actors use access relations to influence preferences of other actors. Establishment and shifts of access relations and their consequences for outcomes of decisions are the main focal points in this paper. Unlike most policy network studies, we therefore do not take the network an its relations as given and constant. Instead we device computer simulation models to account for the dynamics in policy networks. We compare different models and investigate the resulting network structures and predicted outcomes of decisions The choice among the alternative models is made by their correspondence with empirical network structures and actual outcomes of decisions.

In our models, we assume that all relevant actors aim at policy outcomes as close as possible to their own preferences. Policy outcomes are determined by the preferences of the final decision makers at the moment of the vote. In general, only a small fraction of the actors takes part in the final vote. Most actors have therefore to rely on access relations for directly or indirectly shaping the preference of the final decision makers. For this purpose actors make access requests to other actors. An access relation is assumed to be established if such a request is accepted by the other.

Access relations require time and resources. Actors are therefore assumed to be restricted in the number of access requests that they can make and the number of requests they can accept. Moreover, due to incomplete information and simultaneous actions by other actors, actors have to make simplifying assumptions in the selection of their "best" requests and learn by experience.

We devise two base models that correspond to two basic views on the nature of political processes. In the first view politics is seen as power driven. Corresponding to this view, actors aim at access relations with the most powerful actors in the field. They estimate their likelihood of success by comparing their own resources with those of the target actors. Power also determines the order in which actors accept requests. In the second view, policy matters and actors roughly estimate the effects access relations might have on the outcome of decisions. Actors select requests to "bolster" their own preference as much as possible.

We will show that these base models and some intermediate ones result in fundamentally different network structures and predicted outcomes. Moreover, we will show that the policy driven models do fundamentally better than the power driven models.

Suitor, J., and S. Keeton. 1997. Once a Friend, Always a Friend? Effects of Homophily on Women's Support Networks Across a Decade. Social Networks. 19: 51-62.

This study uses data collected on 42 women and 432 members of their social networks across a 10-year period beginning with the women's return to school in midlife. The paper addresses three questions: (1) To what extent did the same individuals continue to be active members of the women's social support networks across the decade; (2) to what extent did educational similarity explain which individuals continued to be active members of the women's networks; and (3) did educational similarity explain patterns of both support and socializing? The analyses revealed that between one-quarter and one-third of the associates named as sources of school/work support, general emotional support or socializing continued to be named 10 years after the first interview. The analyses also demonstrated that associates' educational attainment was important in explaining which individuals continued to serve as sources of support for school/work; however, educational attainment was not important in explaining which associates continued to serve as sources of either general emotional support or socializing.

Thye, S. R., M. J. Lovaglia, and B. Markovsky. 1997. Responses to Social Exchange and Social Exclusion in Networks. Social Forces. 75(3): 1031-47.

Various theoretical accounts of power in social exchange networks have emerged in recent years. We use a new experimental setting to test assumptions that appear to be implicit in all network exchange theories of power: Actor's in exchange networks increase their demands following social exchange and concede more resources when excluded. We also tested for the relative impact of inclusion versus exclusion and differential responses to multiple exchange partners give exchange with just one of them. We observed that the tendency to concede the following exclusion was significantly greater than the tendency to raise demands following an exchange. Furthermore, when an actor had multiple potential exchange partners, the clear tendency was to demand more from those with whom exchange just occurred, but not form those excluded. These findings are discussed in view of current theories of exchange and commitment.

Tourangeau, R., T. W. Smith, and K. A. Rasinski. 1997. Motivation to Report Sensitive Behaviors on Surveys: Evidence from a Bogus Pipeline Experiment. Journal of Applied Social Psychology. 27(3): 209-22.

We examined the effects of a procedure designed to increase motivation to give accurate reports of socially sensitive behaviors frequently collected in surveys. Respondents were assigned at random to a bogus pipeline condition, in which they were told that inaccurate answers could be detected by a physiological recording device, or to a control condition. Respondents in both conditions were interviewed with a questionnaire that contained 19 items ranging from smoking and exercise frequency to number of sexual partners and illicit drug use. While the bogus pipeline procedure generally increased the reporting of sensitive behavior, in one instance it reduced the proportion of respondents who reported a socially desirable behavior (exercising), indicating that respondents were motivated to respond accurately and not just to report more occurrences.

Wang, Y., H. Yan, and C. Sriskandarajah. 1996. The Weighted Sum of Split and Diameter Clustering. Journal of Classification. 13: 231-48.

In this paper, we propose a bicriterion objective for clustering a given set of N entities, which minimizes [ad - (1 - a)s], where 0 a 1, and d and s are the diameter and the split of the clustering, respectively. When a = 1, the problem reduces to minimum diameter clustering, and when a = 0, maximum split clustering. We show that this objective provides an effective way to compromise between the tow often conflicting criteria. While the problem is NP-hard in general, a polynomial algorithm with the worst-case time complexity O(N²) is devised to solve bipartitions with respect to diameter and split, and it can be extended to yield an efficient divisive hierarchical scheme. An extension of the approach to the objective [a(d₁ + d₂) - 2(1 - a)s] is also proposed, where d₁ and d₂ are diameters of the two clusters of a bipartition.

Weesie, J., and W. Raub. 1996. Private Ordering: A Comparative Institutional Analysis of Hostage Games. Journal of Mathematical Sociology.

Hostage posting (in the sense of pledging a bond) is a commitment device that allows for cooperation of rational actors in economic and social relations with incentive problems, like in the Prisoner's Dilemma. This paper provides, first, an informal discussion of hostage posting as a mechanism of cooperation. We then analyzed noncooperative 2- and n-person games with complete information where players can post a hostage prior to their interaction. We compare rather general hostage 'institutions' that specify the conditions under which hostages are declared forfeited and, if forfeited, whether the hostages are transferred to another player or are lost. The problem of designing efficient hostage institutions is addressed and solved for 2-person settings. The minimal institutional requirement for individually rational hostage posting and subsequent cooperation is, roughly, that a player's hostage is forfeited if hostages have been posted by everyone and if the player deviates unilaterally from cooperation. Furthermore, the hostages posted have to be sufficiently valuable, i.e., match the players' costs of cooperation, and the transaction costs associated with hostage posting have to be sufficiently low.

Wellman, B. 1996. Are Personal Communities Local? A Dumptarian Reconsideration. Social Networks. 18: 347-54.

Are local ties important in personal community networks? Since local ties only make up a minority of people's active ties, network analysts have argued for decades that the neighborhood is not very important. Re-analysis of the Toronto data shows that when contacts become the unit of analysis instead of ties, the percentage of local relationships in active networks nearly doubles. Moreover, when we also take into account active contacts with coworkers, who like neighbors are physically proximate, we find that two-thirds of all contacts are 'local'. As Humpty-Dumpty has cogently reminded us, a network can be anything we want it to be. it depends on how we define it. When we change the definition, the conclusions change too.

Wellman, B., R. Yuk-lin, D. Tindall, and N. Nazer. 1997. A Decade of Network Change: Turnover, Persistence and Stability in Personal Communities. Social Networks. 19: 27-50.

We analyze changes in intimate ties in personal community networks. Our data come from interviews conducted a decade apart with 33 Torontonians. There is much turnover in these networks, with only 27% of intimate ties persisting. Durable ties tend to be with intimates who have provided social support, are infrequent telephone contact, or are kin. There was almost complete turnover in the networks of those respondents who got married during the decade. By contrast, the amount of turnover in networks is not associated with whether the respondents had children, moved to a different home, or started/stopped doing paid work during the decade.

Whitmeyer, J. M. 1997. The Power of the Middleman - A Theoretical Analysis. Journal of Mathematical Sociology. 22(1): 59-90.

In this article, I present a technique, multiple general equilibrium analysis (multiple GEA), based on Coleman's (1990) general equilibrium analysis (GEA), for determining the distribution of power among members of a positively connected network in which middlemen mediate the transfer of resources. In line with earlier simulation and experimental studies, multiple GEA indicates that middlemen actors gain considerable power and resources from their structural position. Results also show that, under the assumptions of multiple GEA, actors' relative power can vary considerably depending on the configuration of actor interests. Unlike previous uses of GEA to analyze such networks, the results follow without needing to use ad hoc factors, or to give middlemen actors initial control over any resource.

Yamaguchi, K. 1996. Power in Networks of Substitutable and Complementary Exchange Relations: A Rational-Choice Model and an Analysis of Power Centralization. American Sociological Review. 61(2): 308-32.

In this paper, I introduce a new measure of power in exchange networks under substitutable/complementary exchange relations. Although it is derived from a model based on a modification and extension of Coleman's model of collective action, the new measure reflects major characteristics of Emerson and Cook's power-dependence theory, in which power emerges as a result of exchange based on actors' benefit-maximizing actions under network constraints on exchange. In the new measure, an actor's power depends on the number of exchange partners interested in the actor's resource, the extend to which the actor's partners are not interested or are less interested in others' resources than in the actor's, the power of the actor's partners as interdependent correlates, and the consequences of shifts in demand made by the actor's partners under the substitutability/complementarity of exchange relations among their multiple partners. I also show high consistency between power distributions predicted by the new measure and corresponding experimental results by Cook et al. (1983), Yamagishi, Gilmore, and Cook (1988), and Skvoretz and Willer (1993). Finally, using simulated exchange network data, I derive an enriched set of hypotheses about the structural and relational determinants of power centralization under closely substitutable exchange relations.

Zeggelink, E. P. H., F. N. Stokman, and G. G. Van de Bunt. 1996. The Emergence of Groups in the Evolution of Friendship Networks. Journal of Mathematical Sociology. 21(1-2): 29-55.

Friendship networks usually show a certain degree of segmentation: subgroups of friends. The explanation of the emergence of such groups from initially dyadic par friendships is a difficult but important problem. In this paper we attempt to provide a first contribution to the explanation of subgroup formation in friendship networks by using the LS set as a definition for a friendship group. We construct a dynamic individual oriented model of friendship formation and provide preliminary simulation results that give an idea of how to continue the process of explaining group formation.