Using geocoded survey data to improve the accuracy of multilevel small area synthetic estimates

This paper examines the secondary data requirements for multilevel small area synthetic estimation (ML-SASE). This research method uses secondary survey data sets as source data for statistical models. The parameters of these models are used to generate data for small areas. The paper assesses the impact of knowing the geographical location of survey respondents on the accuracy of estimates, moving beyond debating the generic merits of geocoded social survey datasets to examine quantitatively the hypothesis that knowing the approximate location of respondents can improve the accuracy of the resultant estimates. Four sets of synthetic estimates are generated to predict expected levels of limiting long term illnesses using different levels of knowledge about respondent location. The estimates were compared to comprehensive census data on limiting long term illness (LLTI). Estimates based on fully geocoded data were more accurate than estimates based on data that did not include geocodes.     

To test or not to test: do workplace drug testing programs discourage employee drug use?

Workplace drug testing programs are often met with intense criticism. Despite resistance among labor and consumer groups and a lack of rigorous empirical evidence regarding effectiveness, drug testing programs have remained popular with employers throughout the 1990s and into the current century. The present study analyzed nationally representative data on over 15,000 US households to determine whether various types of workplace drug testing programs influenced the probability of drug use by workers. The study estimated several empirical specifications using both univariate and bivariate probit techniques. The specification tests favored the bivariate probit model over the univariate probit model. Estimated marginal effects of drug testing on any drug use were negative, significant, and relatively large, indicating that drug testing programs are achieving one of the desired effects. The results were similar when any drug use was replaced with chronic drug use in the models. These results have important policy implications regarding the effectiveness and economic viability of workplace anti-drug programs.     

Are there sex differences in the utilization of educational capital among college-educated workers?

This paper introduces the concept of educational utilization as an overlooked part of the education-to-work transition and a potential mechanism by which occupational sex segregation is generated among the college-educated labor force. The paper begins with a critical discussion of the operationalization approaches that have been used in prior research that implicitly measures educational utilization. Multiple empirical measure of the concept are then developed using data from the O*NET and the National Surveys of College Graduates. The explanatory power of each measure is assessed using conditional logit models of occupational attainment. A combined measure is then used to assess sex differences in educational utilization using data from the 1993 and 2003 National Surveys of College Graduates for 2 cohorts of college graduates—those who earned their baccalaureate or post-baccalaureate degrees and entered the labor market in the years 1985–1993 and 1995–2003. The analysis identifies sex differences in educational utilization that vary across field, degree level and cohort and concludes with an examination of the implications of sex differences in educational utilization for occupational segregation.     

Qualitative variance,entropy, and correlation ratios for nominal dependent variables

In this paper we derive a measure of dispersion for a nominal variable having k ? 2 categories and compare it with ordinary quantitative variance and with entropy. We then develop two qualitative analogs to the R² statistic, one based on qualitative variance, the other on entropy. For concreteness, data from the study of The American Soldier by S. A. Stouffér, E. A. Suchman, L. C. Devinney, S. A. Star, and R. M. Williams, Jr. (1949, Princeton Univ. Press, Princeton, N.J.) are used to illustrate the computations and compare these two correlation ratios for a variety of logit models. It is shown that both measures are reasonable measures of the proportion of variation of one or more qualitative dependent variables explainable by any mixture of qualitative and quantitative predictor variables. They can be used with logit, probit, dummy variable regression models, or probability models of some other type.     

Model fertility schedules revisited: The log-multiplicative model approach

This paper reconsiders Coale and Trussell's (1974) specification of model fertility schedules by age. It formally presents model fertility schedules within the framework of categorical data analysis. Specifically, births are assumed to follow an independent Poisson distribution for each age interval of each population. Identification and estimation problems are discussed. It shows that the Coale-Trussell specification corresponds to Goodman's (1979) log-multiplicative model. Following Goodman's algorithm, the paper simultaneously estimates Coale and Trussell's υ (age), m, and M through an iterative maximum likelihood procedure. This is demonstrated with the same data that were used in Coale and Trussell's article. The new estimates are superior to those of Coale and Trussell according to an array of conventional goodness-of-fit criteria.     

Recursive models for categorical data

Standard methods for recursive models with continuous endogenous variables are extended to models with categorical endogenous variables. The concept of a reduced-form equation is generalized in a natural way to cover nonlinear regression functions and, in particular, models with categorical endogenous variables. Maximum-likelihood estimation and asymptotic chi-square tests are described. Two numerical examples are presented: a linear recursive two-equation model for all-categorical data, and a combined linear and logit three-equation recursive model with both categorical and continuous endogenous variables. Limitations of the present work and directions for further extension are noted.     

The making of family values: Developmental idealism in Gansu,China

This paper examines the role of developmental thinking in the making of family values. We analyze survey data collected from Gansu Province in China with regular and multilevel logit models. The results show that individuals’ endorsement of neolocal residence, self-choice marriage, gender egalitarianism, late marriage for women, and low fertility depends on the conjunction of preference for development and beliefs in its association with those family attributes, which we term developmental idealism associational evaluation. Furthermore, such impact of developmental thinking on family values holds robust in the presence of indigenous ideational forces, in this case Islamic religion. Although Islam influences family values in the opposite direction than developmental ideas do, the effect of Developmental Idealism associational evaluation does not differ significantly between Muslims and non-Muslims.     

Volunteer work, informal help, and care among the 50+ in Europe: Further evidence for ‘linked’ productive activities at older ages

Taking a cross-national perspective, we investigate linkages between volunteer work, informal help, and care among Europeans aged 50 or older. Based on 27,297 personal interviews from the 2004 Survey of Health, Ageing and Retirement in Europe, we estimate univariate and multivariate probit models, which allow us to analyze the interrelationship between those non-market productive activities. There is substantial variation in the participation in volunteering, helping, and caring between countries. Independent of the general level of activity in a country, we find evidence for a complementary and interdependent relationship between all three activities. Our findings not only suggest an important role of societal opportunity structures in elders’ productive engagement, but also support notions of the existence of a general motivation to be active.     

On population comparisons using factor indexes or latent variables

Population comparisons of multivariate models using factor indexes or latent factors are demonstrated to be unsound with current methods of comparative analysis unless one is willing to assume a priori that all of the variances of the latent factors in a model are identical across all populations analyzed. This is an assumption that is probably only rarely found in comparative research. Without this assumption, comparisons are unsound regardless of whether model estimations are made from regression analysis or maximum likelihood techniques. A general method is suggested which proposes to make comparative analyses of factors possible without the assumption of equal variances across populations.     

Revisiting the relationship between perceived discrimination and health: Evidence from sibling models with multiple health measures

Perceived discrimination (PD) is known to be significantly related to health outcomes. This link, however, warrants further scrutiny due to the possibility of unobserved family-level confounding. Using the Add Health and sibling fixed-effect approach, we examine whether the relationship between PD and health is confounded by family background characteristics such as genetics, family environment, and childhood social context (school and neighborhood effects). While the naive OLS models exhibit significant associations between PD and both physical and psychological health outcomes, our preferred sibling fixed-effect estimates reveal that the observed associations between PD and physical health outcomes are confounded by shared family background. In contrast, the observed associations for psychological health, self-reported health, and some of health behavior outcomes are robust to adjustment for sibling fixed-effects. Furthermore, we find similar overall patterns in the link between PD and health across races/ethnicities.     

商业银行风险承担行为的激励合约选择

商业银行的风险水平在一定程度上取决于管理者的风险承担行为,因此,恰当的管理者行为激励对于商业银行的风险控制至关重要.为克服已有研究缺乏对商业银行管理者激励合约设计的关注,以及正式激励合约的外在考核变量低效等问题,本文试图通过扩展已有理论模型,综合考虑商业银行管理者风险承担行为结果的真实信息和外在信息,从关系合约与正式合约选择的角度,建立一个更为全面和灵活的商业银行管理者风险承担行为的激励合约框架.     

Psychological ‘self–other overlap’ and support for slavery reparations

Neurobiological and social psychological studies suggest the existence of non-conscious ‘self–other overlap’ that can lead individuals to identify with other individuals or groups. This paper investigates whether this effect may help to explain support by some Whites for an otherwise unpopular policy on behalf of African Americans-slavery reparations. A representative telephone survey (n = 1200) serves as baseline of comparison for an online reaction time study with a non-representative sample (n = 1341) that measures ‘self–other overlap’ (interchangeably referred to as “implicit closeness” to Blacks). Partial proportional odds ordered logit analyses reveal implicit closeness to Blacks as the single most powerful predictor of support for reparations among White respondents. The magnitude of the absolute effect of implicit closeness exceeds that of traditional predictors such as racial resentment, ideology, and party identification. Methodological and political implications are discussed.     

云计算环境中数据处理行为的法律规制研究

目前很多国家和地区的个人数据保护法都主张其具有域外效力,适用于跨境数据处理行为。而现行冲突规范中有关物权、知识产权的规则无法涵盖所有的数据处理行为,由于数据可能涉及国家安全,也无法适用合同法规则。在涉及数据处理行为的立法中明确其地域适用范围的,只有欧盟指令和即将生效的条例以及欧盟成员国的数据保护立法。欧盟立法主要采用“数据控制者的机构所在地”和“数据处理行为发生地”两种适用标准,而云计算技术的广泛应用给这两项标准带来了新的问题。“数据处理行为发生地”逐渐被淡化,相较之下,“数据控制者的机构所在地”成为更可靠的标准。我国的数据处理立法也应当确立其域外效力,规制跨境数据处理行为。     

Multivariate analysis and anthropological applications

A brief history of the use of statistical concepts in the field of anthropology is reviewed as is the history of the development of multivariate analysis. Those multivariate statistical techniques which are particularly of interest to anthropologists, that is, factor analysis, cluster analysis, and discriminant analysis, are explained. They are related to multivariate analysis theory in general, are compared among themselves, and are further treated in light of their individual characteristics. Anthropological applications of multivariate analysis are cited, and it is concluded that joint cooperation between anthropologists and statisticians in multivariate research would be beneficial to both.     

A global risk assessment model for civil wars

In this study, we construct a multivariate model that assesses the risk of an outbreak of civil war in a country over a period of 5 years into the future. In addition to structural factors of state weakness, which have dominated the literature on civil war onset, this model includes repression of basic human rights to personal integrity – an important harbinger of wars to come – as an aspect of state behavior. Our aim is not to explore the causal factors of civil war onset, but to build a model that includes indicators that correlate with civil war outbreak and may be used to predict it. Based on two versions of the model – logit and neural network – out-of-sample risk assessments for three different time periods are generated and compared to the historical record of civil war outbreak during those years. In addition, the model’s ability to produce in-sample risk assessments over a 5-year period is tested. Finally, we compute truly predictive civil war risk assessments for all countries for which data are available, for the years 2008–2012. The analyses show that with a relatively simple model and based on publicly available data sources, meaningful civil war risk assessments can be computed. The quality of the predictions exceeds that of prominent studies, in which the risk of interstate war is assessed.     

The analysis of ordinal dispersion: A comparison of Leik's “D” with a variance formula

The frequency of ordinal data and the importance of attitudinal consensus in social research motivated Leik [Leik, R. L. (1966), Pacific Sociological Review 9, 85–90] to formulate a simple, effective measure of ordinal dispersion, but he did not provide a method for its statistical analysis, and its operational interpretation for nonextremal values is unclear. Leik's D is an equal-interval, “folded” scoring vector inner dot-product, scalar transformation of the vector of ordered relative frequencies. The folding point is the median category. The metric analog of D is the difference between the random variable's expected values above and below the median. An equal-interval, pure variance measure of ordinal dispersion is posited and compared to D. They correlate closely but not perfectly, and D has a serious problem when an extremal category has more than half the total cases. Both formulas are easily generalized to non-equal-interval scorings. A general linear model for categorical data permits easy statistical comparisons of D, but not the variance. D tends to underestimate the degree of significant dispersion in comparison to the variance formula, it is suggested.     

Understanding selection bias,time-lags and measurement bias in secondary data sources: Putting the Encyclopedia of Associations database in broader context

Secondary data gathered for purposes other than research play an important role in the social sciences. A recent data release has made an important source of publicly available data on associational interests, the Encyclopedia of Associations (EA), readily accessible to scholars (www.policyagendas.org). In this paper we introduce these new data and systematically investigate issues of lag between events and subsequent reporting in the EA, as these have important but under-appreciated effects on time-series statistical models. We further analyze the accuracy and coverage of the database in numerous ways. Our study serves as a guide to potential users of this database, but we also reflect upon a number of issues that should concern all researchers who use secondary data such as newspaper records, IRS reports and FBI Uniform Crime Reports.     

基于Logit模型的上市公司财务预警实证分析

以2005-2007年中国上市公司为研究对象,采用Logit模型对上市公司财务预警进行实证分析,并对模型的预测效果进行了检验。结果表明logit模型可以对上市公司一年后是否出现财务危机进行预测。