The Signed Mantel test to cope with autocorrelation in comparative analyses

In biology, medicine and anthropology, scientists try to reveal general patterns when comparing different sampling units such as biological taxa, diseases or cultures. A problem of such comparative data is that standard statistical procedures are often inappropriate due to possible autocorrelation within the data. Widespread causes of autocorrelation are a shared geography or phylogeny of the sampling units. To cope with possible autocorrelations within comparative data, we suggest a new kind of the Mantel test. The Signed Mantel test evaluates the relationship between two or more distance matrices and allows trait variables facultatively to be represented as signed distances (calculated as signed differences or quotients). Considering the sign of distances takes into account the direction of an effect found in the data. Since different metrics exist to calculate the distance between two sampling units from the raw data and because the test results often depend on the kind of metric used, we suggest validating analysis by comparing the structures of the raw and the distance data. We offer a computer program that is able to construct both signed and absolute distance matrices, to perform both customary and Signed Mantel tests, and to explore raw and distance data visually.     

Modelling method effects as individual causal effects

Summary.  Method effects often occur when different methods are used for measuring the same construct. We present a new approach for modelling this kind of phenomenon, consisting of a definition of method effects and a first model, the method effect model , that can be used for data analysis. This model may be applied to multitrait–multimethod data or to longitudinal data where the same construct is measured with at least two methods at all occasions. In this new approach, the definition of the method effects is based on the theory of individual causal effects by Neyman and Rubin. Method effects are accordingly conceptualized as the individual effects of applying measurement method j instead of k . They are modelled as latent difference scores in structural equation models. A reference method needs to be chosen against which all other methods are compared. The model fit is invariant to the choice of the reference method. The model allows the estimation of the average of the individual method effects, their variance, their correlation with the traits (and other latent variables) and the correlation of different method effects among each other. Furthermore, since the definition of the method effects is in line with the theory of causality, the method effects may (under certain conditions) be interpreted as causal effects of the method. The method effect model is compared with traditional multitrait–multimethod models. An example illustrates the application of the model to longitudinal data analysing the effect of negatively (such as 'feel bad') as compared with positively formulated items (such as 'feel good') measuring mood states.     

A parsimonious multivariate poisson model for market basket analysis

Intensified research on multivariate Poisson models offers new opportunities for the analysis of purchase quantities in market basket data. The investigation of positive or negative correlations in quantity decisions among product categories facilitates a deeper understanding of consumer purchase behavior. The applied multivariate log-normal Poisson model introduces interdependencies between categories with multivariate normal-distributed latent effects by means of a covariance matrix. As the size of this covariance matrix depends on the number of categories in the model, its ation may become tedious. Furthermore, we assume that quantity decisions do not interact for all pairs of categories. That is why we propose to use covariance selection to derive a parsimonious representation of the correlation structure. For two market basket data sets, we show that the vast majority of off-diagonal elements in the covariance matrix are irrelevant. For a data set with product categories, the model with a partly restricted covariance matrix achieves a better fit to the holdout data than the model with full covariance matrix. For a data set with subcategories of the broader category beverage, the proposed model with restricted covariance outperforms the model with full covariance matrix even on the calibration data. We conclude that interactions of quantity decisions are overall the exception, even for complements-in-use.     

Falling through the social safety net? Analysing non-take-up of minimum income benefit and monetary social assistance in Austria

Non-take-up of means tested benefits is a widespread phenomenon in European welfare states. The paper assesses whether the reform that replaced the monetary social assistance benefit by the minimum income benefit in Austria in 2010/11 has succeeded in increasing take up rates. We use EU-SILC register data together with the tax-benefit microsimulation model EUROMOD/SORESI. The results show that the reform led to a significant decrease of non-take-up from 53 to 30% in terms of the number of households and from 51 to 30% in terms of expenditure. Following the three-t's (threshold, trigger, and trade-off) introduced by Van Oorschot, estimates of a two-stage Heckman selection model as well as expert interviews indicate that the taken measures include both threshold and trade-off characteristics. Elements such as the higher degree of anonymity within the claiming process, the provision of health insurance, binding minimum standards, the limitation of the maintenance obligations, new regulations related to the liquidation of wealth, as well as the general coverage of the benefit reform in the media and in public discussions led to an improved access to the benefit.     

USE OF EXPERT RATINGS AS SAMPLING STRATA FOR A MORE COST-EFFECTIVE PROBABILITY SAMPLE OF A RARE POPULATION

We consider situations in which externally observable characteristics allow experts to quickly categorize individual households as likely or unlikely to contain a member of a rare target population. This classification can form the basis of disproportionate stratified sampling such that households classified as "unlikely" are sampled at a lower rate than those classified as "likely," thereby reducing screening costs. Design weights account for this approach and allow unbiased estimates for the target population.We demonstrate that with sensitivity and specificity of expert classification at least 70%, and ideally at least 80%, our approach can economically increase effective sample size for a rare population. We develop heuristics for implementing this approach and demonstrate that sensitivity drives design effects and screening costs whereas specificity only drives the latter. We demonstrate that the potential gains from this approach increase as the target population becomes rarer. We further show that for most applications, unlikely strata should be sampled at 1/6 to ? the rate of likely strata.This approach was applied to a survey of Cambodian immigrants in which the 82% of households rated "unlikely" were sampled at ? the rate as "likely" households, reducing screening from 9.4 to 4.0 approaches per complete. Sensitivity and specificity were 86% and 91% respectively. Weighted estimation had a design effect of 1.26 so screening costs per effective sample size were reduced 47%. We also note that in this instance, expert classification appeared to be uncorrelated with survey outcomes of interest among eligibles.     

Outsiders and Intrapreneurs: The Institutional Embeddedness of Social Entrepreneurship in Germany

The social entrepreneurship discourse in Germany has become more prominent at a time when the deeply rooted corporatist traditions of social provision have come under pressure for marketization. This article examines the potential role of “social entrepreneurs” in the institutionally established German welfare state. The article analyzes the opportunities and constraints that new players face. Drawing on survey data and case studies in the areas of elderly care and advancement of children with immigrant background, the analysis retraces the structure and diffusion of social entrepreneurial projects. It concludes that the simple transfer of the social entrepreneurship model is unlikely. The analysis suggests that successful social ventures in Germany adapt the notion of social entrepreneurship to prevalent institutional realities. In the context of more encompassing social services, dense decentralized networks, and different cultures of philanthropism, new players have a complementary role that stimulates rather than dominates the process of social innovation.     

Engaging diverse experts in a global environmental assessment: participation in the first work programme of IPBES and opportunities for improvement

The need for interdisciplinary expert groups from different regions of the world to be involved in the fields of sustainability science and environmental change research is increasingly recognised. The Intergovernmental Platform on Biodiversity and Ecosystem Services (IBPES) was established in 2012 as a science-policy interface and has gone beyond previous initiatives in its articulation of a clear commitment to inter- and transdisciplinary approaches that mandate a diversity of genders, disciplines and regional backgrounds within its expert groups. The first IPBES work programme, carried out between 2014 and 2018, has been supported by 17 expert groups, comprising over 1000 experts, who have been selected from over 2000 government and stakeholder nominations through formal procedures. In this paper, we present and critique the framework through which IPBES identifies and selects experts to participate in its processes. In addition, we synthesise and carry out a quantitative analysis on the expert nomination and selection data relating to the first assessment activities of IPBES. Identifying that the balance of regions, genders, disciplines and knowledge systems represented within these expert groups is still disproportionally dominated by male natural scientists from the Global North, the paper makes recommendations of how to better engage knowledge holders from different disciplines and diverse knowledge systems in future iterations of the IPBES work programme.     

Die Umzugsentscheidung von Paaren als Verhandlungsproblem

Recently bargaining models became a popular theoretical concept within family research. Especially compared to household economic theory, bargaining models are less abstract and provide a tool for analysis of conflicting interests and their consequences in family relationships. However, although there are many different topics which can be tackled with this kind of theory, there is still a lack of direct empirical tests concerning this concept. The aim of this article is to provide such a test by using a special type of a quasi-experimental research design. With a so-called factorial design we vary systematically incentives for a household move in order to model different structural conditions of a relationship like earning possibilities or chances for a career of a partner. This allows us to observe varying power allocations within the relationship. The empirical results support the hypotheses derived from the bargaining model: the respondent anticipates a loss of his or her bargaining power as well as an increasing power of the partner. Consequently they answer such a change with a reduced tendency for a household move and a higher perceived potential for conflicts within the relationship. However, due to our limited knowledge about the validity of the quasi-experimental design these results need to be validated by future research.