Sampling and Estimation in Hidden Populations Using Respondent‐Driven Sampling

Standard statistical methods often provide no way to make accurate estimates about the characteristics of hidden populations such as injection drug users, the homeless, and artists. In this paper, we further develop a sampling and estimation technique called respondent‐driven sampling, which allows researchers to make asymptotically unbiased estimates about these hidden populations. The sample is selected with a snowball‐type design that can be done more cheaply, quickly, and easily than other methods currently in use. Further, we can show that under certain specified (and quite general) conditions, our estimates for the percentage of the population with a specific trait are asymptotically unbiased. We further show that these estimates are asymptotically unbiased no matter how the seeds are selected. We conclude with a comparison of respondent‐driven samples of jazz musicians in New York and San Francisco, with corresponding institutional samples of jazz musicians from these cities. The results show that some standard methods for studying hidden populations can produce misleading results.     

Exponential Family Models for sampled and census Network data

Much progress has been made on the development of statistical methods for network analysis in the past ten years, building on the general class of exponential family random graph (ERG) network models first introduced by Holland and Leinhardt (1981) . Recent examples include models for Markov graphs, "p*" models, and actor‐oriented models. For empirical application, these ERG models take a logistic form, and require the equivalent of a network census: data on all dyads within the network. In a largely separate stream of research, conditional log‐linear (CLL) models have been adapted for analyzing locally sampled ("egocentric") network data. While the general relation between log‐linear and logistic models is well known and has been exploited in the case of a priori blockmodels for networks, the relation for the CLL models is different due to the treatment of absent ties. For fully saturated tie independence models, CLL and ERG are equivalent and related via Bayes' rule. For other tie independence models, the two do not yield equivalent predicted values, but we show that in practice the differences are unlikely to be large. The alternate conditioning in the two models sheds light on the relationship between local and complete network data, and the role that models can play in bridging the gap between them.     

Improved Regression Estimation of A Multivariate Relationship with Population Data on the Bivariate Relationship

Regression coefficients specify the partial effect of a regressor on the dependent variable. Sometimes the bivariate or limited multivariate relationship of that regressor variable with the dependent variable is known from population‐level data. We show here that such population‐level data can be used to reduce variance and bias about estimates of those regression coefficients from sample survey data. The method of constrained MLE is used to achieve these improvements. Its statistical properties are first described. The method constrains the weighted sum of all the covariate‐specific associations (partial effects) of the regressors on the dependent variable to equal the overall association of one or more regressors, where the latter is known exactly from the population data. We refer to those regressors whose bivariate or limited multivariate relationships with the dependent variable are constrained by population data as being "directly constrained." Our study investigates the improvements in the estimation of directly constrained variables as well as the improvements in the estimation of other regressor variables that may be correlated with the directly constrained variables, and thus "indirectly constrained" by the population data. The example application is to the marital fertility of black versus white women. The difference between white and black women's rates of marital fertility, available from population‐level data, gives the overall association of race with fertility. We show that the constrained MLE technique both provides a far more powerful statistical test of the partial effect of being black and purges the test of a bias that would otherwise distort the estimated magnitude of this effect. We find only trivial reductions, however, in the standard errors of the parameters for indirectly constrained regressors.     

Rapport‐building activities in corner shop interactions

This paper examines a range of interactional activities participants carry out in the course of corner shop transactions in Quito, Ecuador. These activities include: phatic communication exchanges as traditionally conceived, that is, conventionalized forms such as how‐are‐you and health inquiries; individualized exchanges in the form of conversational work around a range of topics (e.g. politics, health, school); and creative language play activities (e.g. wordplay and linguistic play with names). These activities are described as reflecting participants' orientation to the maintenance of positive rapport or friendly relations ( Aston 1988a ) with the effect that the service transaction becomes a pleasant and even an entertaining encounter. Such orientation is described here in relation to the familiarity existing between shopkeepers and customers as a result of frequent contact in the context of the barrio (residential neighbourhoods) in Quito. As such the study lends support, from a different socio‐cultural perspective, to recent work in the area that highlights the centrality of phatic communication in task‐oriented interactions in English (cf. Coupland 2000a , 2000b , 2003 ), and brings to the fore a wider range of activities which appear to be employed for rapport‐building purposes in service encounters.     

Problems with the 'language‐as‐resource' discourse in the promotion of heritage languages in the U.S.A.

In the United States, language 'rights' have been tethered to ethnic or racial entitlements as a means to redress historical patterns of discrimination and exclusion. The perception that language 'rights' are about the redress of past wrongs has had negative effects on efforts to gain broad public support for the teaching and maintenance of languages other than English. The language‐as‐resource orientation ( Ruiz 1984 ) is considered as an alternative to a language rights approach. However, analysis of texts produced by advocates of the heritage language movement reveals the shortcomings of the language‐as‐resource metaphor in advancing broad‐based support for the teaching, maintenance, and use of minority languages in the U.S. While efforts to promote heritage language education as a national strategic priority may result in short‐term governmental support, wider and more sustained popular support for such programs will require significant modifications in the underlying values and ideologies about the status and role of languages other than English in education and public life.     

Gambling on Electronic Gaming Machines in Germany: An Age-Period-Cohort Analysis

To provide insights into etiological factors of gambling at the population level, it is critical to document the separate contributions of biological aging, period influences, and birth cohorts on observed temporal trends in gambling. This study investigated age, period, and cohort effects on prevalence rates of electronic gaming machine (EGM) gambling in Germany. We used data from a series of repeated cross-sectional surveys from the Federal Centre for Health Education covering the period 2007–2015. A total of 53,005 participants were surveyed about their past-year participation in EGMs. Using the intrinsic estimator, we disentangled the separate effects of age (16–17 to 64–65 years), period (2007, 2009, 2011, 2013, 2015), and birth cohort (1941–42 to 1997–98). Age effects were highest for those aged 18–33 (p < 0.001). Period effects were highest in 2013 and lowest in 2015 (p < 0.005). A significant increasing birth cohort effect was evident in cohorts born between 1985 and 1998 (p < 0.005). Implications of these results are discussed with regard to the vulnerability of younger ages to gambling, period changes in the regulatory framework of Germany, and differential cohort vulnerability of the Millennial generation due to early-life video gaming experiences.     

Overcoming the Reference Category Problem in the Presentation of Statistical Models

Effects of categorical variables in statistical models typically are reported in terms of comparison either with a reference category or with a suitably defined "mean effect," for reasons of parameter identification. A conventional presentation of estimates and standard errors, but without the full variance–covariance matrix, does not allow subsequent readers either to make inference on a comparison of interest that is not presented or to compare or combine results from different studies where the same variables but different reference levels are used. It is shown how an alternative presentation, in terms of "quasi standard errors," overcomes this problem in an economical and intuitive way. A primary application is the reporting of effects of categorical predictors, often called factors, in linear and generalized linear models, hazard models, multinomial–response models, generalized additive models, etc. Other applications include the comparison of coefficients between related regression equations—for example, log–odds ratios in a multinomial logit model—and the presentation of multipliers or "scores" in models with multiplicative interaction structure.     

Globalisation, the nation‐state and global society

This article examines the notion that globalisation has resulted in the decline of the nation‐state. It argues that during most of its history globalisation was in fact associated with the rise of the nation‐state through the construction of overseas empires that extended and strengthened it. The development of global political organization has reinforced the nation‐state through the growth of international organizations based on the principle of national sovereignty. It is transnational flows, networks, and organizations that have challenged the authority of the nation‐state. These do not, however, float in a global limbo but are always rooted in and dependent upon some nation‐state or other. Transnational movements can embarrass nation‐states but have little real impact upon them and challenge them far less than did the nationally based labour movements that have been undermined by globalisation. Some see globalisation as leading to a global society transcending national units but this view overlooks the continued importance of national institutions and international relationships, and the emergence of strong regional organizations, in a society that is essentially multi‐level rather than global in character.     

International Terrorism and the World‐System

Theories of international terrorism are reviewed. It then is noted that waves of terrorism appear in semiperipheral zones of the world‐system during pulsations of globalization when the dominant state is in decline. Finally, how these and other factors might combine to suggest a model of terrorism's role in the cyclical undulations of the world‐system is suggested .     

Fixed–Effects Negative Binomial Regression Models

This paper demonstrates that the conditional negative binomial model for panel data, proposed by Hausman, Hall, and Griliches (1984), is not a true fixed–effects method. This method—which has been implemented in both Stata and LIMDEP—does not in fact control for all stable covariates. Three alternative methods are explored. A negative multinomial model yields the same estimator as the conditional Poisson estimator and hence does not provide any additional leverage for dealing with overdispersion. On the other hand, a simulation study yields good results from applying an unconditional negative binomial regression estimator with dummy variables to represent the fixed effects. There is no evidence for any incidental parameters bias in the coefficients, and downward bias in the standard error estimates can be easily and effectively corrected using the deviance statistic. Finally, an approximate conditional method is found to perform at about the same level as the unconditional estimator.     

Adjusting for Time‐Varying Confounding in Survival Analysis

In this paper we illustrate how directly including endogenous time‐varying confounders in the model of the effect of an exposure on a response can lead to bias in discrete time survival analysis. An alternative to this method is Hernán, Brumback, and Robins'(1999) use of sample weights to adjust for endogenous time‐varying confounding. We discuss when this method can be used to provide unbiased estimators, and we illustrate the method by addressing a substantive research question using existing survey data. We also critically examine the robustness of the weighting method to violations of the underlying assumptions via a simulation analysis.     

Change in Gender‐Ideology, 1977–1996:The Contributions of Intracohort Change and Population Turnover

This paper addresses the processes underlying the dramatic shift in beliefs about women's work and family roles in the United States over the past two decades. Following Mason and Lu (1988) , we posited this shift to be a function of actual change in individual opinions, as well as changes in population membership that result from births and deaths. Using pooled cross‐sections from the General Social Surveys (1977 to 1996), we found that although demographic processes and microlevel attitude change are both important in understanding attitude trends, the contribution of cohort succession is substantially greater now than in the period that Mason and Lu examined. Multivariate analyses show that (a) the sex difference in attitudes is greater among recent cohorts, and (b) the strong association between education and attitudes that characterized earlier cohorts is significantly weaker among cohorts born after 1945.     

Self‐presentation in sociolinguistic interviews: Identities and language variation in Panambi, Brazil

This paper contributes to recent work examining the role of identity, and in particular the uses of language for self‐presentation and the expression of individual identity, through the analysis of two sociolinguistic interviews from a community of German origin in southern Brazil. Drawing from a quantitative study of variation in the southern state of Rio Grande do Sul, we conducted a detailed analysis of the interviews of two women of different ages and social backgrounds. We first describe the social, cultural and historical context of the interviews, and then discuss how our two speakers use their linguistic resources to express varied, and at times conflicting, aspects of their identities. More specifically, we show how our participants seem to maintain certain in‐group, German‐linked features, yet also use out‐group or Brazilian features in order to index both the (local) German and (regional) Brazilian aspects of their identities. Our data and analysis highlight how participants' identity and language use patterns can be better understood through close analysis of the content of their discourse.     

Dilemmas in planning English/vernacular relations in post‐colonial communities

In the context of globalization and post‐modern discourses, the debate about the relative status of local and dominant languages poses serious policy problems for post‐colonial communities. Critics of minority language rights (MLR) generally point out that engineering a language shift on behalf of a vernacular language – motivated by the preservationist interests, collective rights and sentimental associations of an ethnic group – is futile, as the economic and social mobilities of individuals are bound to work against this enterprise. Proponents of MLR have gone to the other extreme of essentializing the linguistic identity of minority communities, generalizing their language attitudes, and treating local language rights as non‐negotiable. This article addresses this debate in the context of the attempts to promote Tamil by the military leadership in the North and East of Sri Lanka. The paper brings together data gathered in sociolinguistic studies for four years in the Jaffna society in order to understand the reception of the language policy in everyday life. The leadership recognizes that language policy is a symbolic statement for political purposes and tolerates certain inconsistencies in policy and practice. While the community assures itself of ethnic pride and linguistic autonomy with the stated policies, it negotiates divergent interests in the gaps between the policy/practice divide. Scholars should recognize the agency of subaltern communities to negotiate language politics in creative and critical ways that transcend the limited constructs formulated to either cynically sweep aside or unduly romanticize language rights.     

How Much More XXX is Generation X Consuming? Evidence of Changing Attitudes and Behaviors Related to Pornography Since 1973

We use data from the General Social Survey (GSS) over a 40-year period (1973–2012) to evaluate changes in attitudes about pornography and pornography consumption among American young adults. One of the major challenges in making comparisons across birth generations is separating the effect of birth cohort from age and period effects. We use an intrinsic estimator to separately identify the effects of age, birth cohort, and time period using 40 years of repeated cross-section data. We find that, relative to the general population, young people's beliefs about whether pornography should be illegal have stayed relatively constant over this 40-year period and, if anything, have slightly increased. We also find that pornography consumption has been increasing across birth generations, though this increase has been smaller than would be inferred based on differences across generations at a single point in time, due to a strong age component in consumption patterns.     

A New Approach to Evaluating the Quality of Measurement Instruments: The Split‐Ballot MTMM Design

Two distinctly different quantitative approaches are used to evaluate measurement instruments: the split‐ballot experiment and the multitrait‐multimethod (MTMM) approach. The first approach is typically used to indicate whether variation in the method causes differences in the response distribution; the second approach evaluates the reliability and validity of different methods. The new approach, suggested in this paper, combines the more attractive features of both methods. The strength of the split‐ballot experiment is its use of independent random samples from the same population to provide information about differences in response distributions. This is also possible with the new approach, but this approach provides more detailed information about the reasons for the differences. The MTMM approach provides information about reliability and validity on the basis of repeated observation of the same traits using different methods. This information is also provided by the new design. The difference is that the new approach reduces the need for repeated observations of the same trait. Each sample is provided with a different combination of only two methods and the complete model with all methods is estimated as a multiple‐group model. This reduces the burden for respondents and also reduces memory and order effects. Alternative designs and estimation methods are discussed, their efficiency is analyzed, and illustrations are provided .     

Ecological‐Evolutionary Theory: A Reanalysis and Reassessment of Lenski's Theory for the 21st Century

Gerhard Lenski's ecological‐evolutionary theory of human societies, originally presented and tested in Power and Privilege (1966) and Human Societies (1970), makes a number of general and specific predictions about the impact of subsistence technology on the fundamental features of societies, as well as identifying constraints that the techno‐economic heritage of currently industrializing societies continue to exercise on their development trajectories. This paper reviews the strategies adopted for presenting and for testing the theory, critically analyzes and extends some important results of its empirical tests, and explores issues confronting the future development and presentation of the theory.     

Extending Lenski's Schema to Hold Up Both Halves of the Sky—A Theory‐Guided Way of Conceptualizing Agrarian Societies that Illuminates a Puzzle about Gender Stratification

This paper suggests that Lenski's classification of agrarian societies into simple versus advanced, based on the use of iron in the latter, obscures important variations in the gender division of labor and the level of gender stratification. In particular, his categories lump the gender egalitarian irrigated rice societies of Southeast Asia with the great majority of agrarian societies, which are strongly patriarchal. Based on my general theory of gender stratification and experience coding and analyzing gender stratification in the ethnographic databases and fieldwork in 39 countries worldwide, I propose a three‐category alternative. First, agrarian societies are divided according to the technological criterion of irrigation into dry (rain‐fed) and wet (irrigated rice) categories. This distinguishes two gender divisions of labor: a male farming system in dry agrarian and an "everybody works" system in labor‐intensive rice cultivation, in which women are important in production. Second, irrigated rice societies are divided into patri‐oriented‐male advantage and those neutral to positive for women, based on the nature of the kinship system. This distinguishes the gender egalitarian Southeast Asian wet rice societies from the highly gender stratified majority of irrigated rice societies. Furthermore, these distinctions in gender equality are predicted by my gender stratification theory.     

Orientalist Knowledge and Social Theories: China and the European Conceptions of East‐West Differences from 1600 to 1900

This paper examines the long‐term development of Orientalism as an intellectual field, with the European learning of China between ca.1600 and ca.1900 as an exemplary case. My analysis will be aided by a theoretical framework based on a synthesis of the world‐system and network perspectives on long‐run intellectual change. Analyzing recurrent debates on China within European intellectual circles, I demonstrate that the Western conception of the East has been oscillating between universalism and particularism, and between naive idealization and racist bias. This oscillation is a function as much of the changing political economy of the capitalist world‐system as of the endogenous politics of the intellectual field. Despite their contrasting views, both admirers and despisers of the East viewed non‐Western civilizations as uniform wholes that had never changed. I argue that the fundamental fallacy of Orientalism lay, not in its presumptions about the ontological differences between East and West and the former's inferiority, as previous critics of Orientalism have supposed, but in its reductionism. Understanding non‐Western civilizations in their full dynamism and heterogeneity is a critical step toward the renewal of the twentieth‐century social theories that were built upon and impaired by the Orientalist knowledge accumulated in the previous centuries.     

Log-Multiplicative Association Models As Latent Variable Models For Nominal and/or Ordinal Data

Associations between multiple discrete measures are often due to collapsing over other variables. When the variables collapsed over are unobserved and continuous, log-multiplicative association models, including log-linear models with linear-by-linear interactions for ordinal categorical data and extensions of Goodman's (1979, 1985) RC(M) association model for multiple nominal and/or ordinal categorical variables, can be used to study the relationship between the observed discrete variables and the unobserved continuous ones, and to study the unobserved variables. The derivation and use of log-multiplicative association models as latent variable models for discrete variables are presented in this paper. The models are based on graphical models for discrete and continuous variables where the variables follow a conditional Gaussian distribution. The models have many desirable properties, including having schematic or graphical representations of the system of observed and unobserved variables, the log-multiplicative models can be read from the graphs, and estimates of the means, variances, and covariances of the latent variables given values on the observed variables are a function of the log-multiplicative model parameters. To illustrate some of the advantageous aspects of these models, two examples are presented. In one example, responses to items from the General Social Survey (Davis and Smith 1996) are modeled, and in the other example, panel data from two groups (Coleman 1964) are analyzed.