On the effect of item nonresponse on the estimation of a two–panel–waves wage equation

Summary: In this paper, we present results of the estimation of a two–panel–waves wage equation based on completely observed units and on a multiply imputed data set. In addition to the survey information, reliable income data is available from the register. These external data are used to assess the reliability of wage regressions that suffer from item nonresponse. The findings reveal marked differences between the complete case analyses and both versions of multiple imputation analyses. We argue that the results based on the multiply imputed data sets are more reliable than those based on the complete case analysis.* We would like to thank Statistics Finland for providing the data. We are also very grateful to Susanna Sandström and Marjo Pyy–Martikainen for their helpful advice using the Finnish data. Helpful comments from Joachim Winter and participants of the Workshop on Item Nonresponse and Data Quality in Large Social Surveys, Basel, October, 2003, on an earlier version of the paper are greatfully acknowledged. Further, we would like to thank three anonymous referees and the editor for helpful comments and suggestions.     

家园之思--论贾平凹90年代之后的乡村叙事

<土门>、<高老庄>、<怀念狼>是贾平凹20世纪90年代后系列长篇小说创作中的三部乡村叙事.三部作品在言说作家的"失家之痛"的基础上,进而对家园之失有所思.作者对乡村的观照眼光也明显有所改变,包含着更为深沉的理性内涵.而这也决定着他新的乡村发现.总体来看,这些长篇乡村叙事中所寄寓的家园之思,体现为无望的抗争、微茫的建构和怀念和谐等三个层面.     

Statistical issues in phase III human immunodeficiency virus vaccine trial design

Several scientific questions are of interest in phase III trials of prophylactic human immunodeficiency virus (HIV) vaccines. In this paper we focus on some issues related to evaluating the direct protective effects of a vaccine in reducing susceptibility, VE_S, and its effect on reducing infectiousness, VE_I. An estimation of VE_I generally requires information on contacts between the infective and susceptible individuals. By augmenting the primary participants of an HIV vaccine trial by their steady sexual partners, information can be collected that allows an estimation of VE_I as well as VE_S. Exposure to infection information, however, may be expensive and difficult to collect. A vaccine trial design can include a small validation set with good exposure to infection data to correct bias in a larger, simpler main study with only coarse exposure data. The large main study increases the efficiency of the small validation set. More research into the combination of different levels of information in vaccine trial design will yield more efficient and less biased estimates of the efficacy measures of interest.     

基于认知视角的马航失联事件危机响应研究

有别于以往危机响应以传统管理理论为依托,本文从认知视角对公众的危机认知作用机理进行了探讨,并结合危机情境沟通理论对危机响应的一般模式进行了重塑。以此为基础,通过对马航失联事件的案例探讨验证了公众认知在危机响应中的重要地位,并指出马航在危机响应中存在的问题及应对策略。     

Analysing survey data with incomplete responses by using a method based on empirical likelihood

Summary.  In many surveys, missing response is a common problem. As an example, Zahner, Jacobs, Freeman and Trainor analysed data from a study of child psychopathology in the State of Connecticut, USA. In that study, the response variable, psychopathology, was inferred from questions that were addressed to teachers of the children and was subject to a high level of missingness. However, the missing responses were supplemented by surrogate information that was provided by the parents and/or the primary care providers of the children. In such a situation, it is conceivable that the supplemental information can be used to recover some of the information that has been lost in the cases with missing response. This paper considers a method using empirical likelihood. Empirical likelihood is well known in providing nonparametric inference. But its application has largely been confined to complete-data situations. The method proposed exploits the semiparametric nature of empirical likelihood. The method gives consistent estimates if the cases with non-missing responses form a random sample of the population. In large samples, the method behaves similarly to a regression estimate that is applied to estimating equations. The method is easy to implement with standard statistical packages. In a small sample study, the method was found to give favourable results, when compared with existing methods.     

Der Einsatz von Missing Data Techniken in der Arbeitsmarktforschung des IAB

Zusammenfassung Dieser Beitrag gibt einen überblick über aktuelle Forschungsprojekte des Instituts für Arbeitsmarkt- und Berufsforschung (IAB), die sich alle auf unterschiedliche Weise mit der Behebung von Problemen mit Missing Data befassen. Hierzu geh?ren einerseits Projekte, die Stichproben verwenden wie das IAB-Betriebspanel, die IAB-Besch?ftigtenstichprobe oder die Erhebung der offenen Stellen, die entweder wie alle Stichprobenerhebungen durch Antwortausf?lle beeintr?chtigt sind oder aber zensierte Beobachtungen aufweisen. Auf der anderen Seite z?hlen hierzu Forschungsprojekte, die versuchen, unterschiedliche Datengrundlagen vereinbar zu machen, die auf verschiedenen regionalen oder klassifikatorischen Abgrenzungen beruhen. Au?erdem wird das Projekt TrEffeR des IAB und der Bundesagentur für Arbeit (BA) vorgestellt, das der oft ausgesprochenen Forderung gerecht wird, die Ma?nahmen der aktiven Arbeitsmarktpolitik auf breiter Basis zu evaluieren, und diese Erkenntnisse für das operative Gesch?ft der BA nutzbar macht.     

A local-influence-based diagnostic approach to a speeded item response theory model

Summary.  An item response theory model for dealing with omitted responses in a test is proposed. In this model formulation, non-response does not only depend on an examinee's ability and on item difficulty, but additionally also on 'test speededness'. Using a local-influence-based diagnostic approach, the sensitivity of the model regarding assumptions concerning the drop-out mechanism is explored. The methodology proposed is applied to the Chilean Sistema de Medición de la Calidad de la Educación mathematics test case-study.     

On the adequacy of variational lower bound functions for likelihood-based inference in Markovian models with missing values

Summary. Variational methods have been proposed for obtaining deterministic lower bounds for log-likelihoods within missing data problems, but with little formal justification or investigation of the worth of the lower bound surfaces as tools for inference. We provide, within a general Markovian context, sufficient conditions under which estimators from the variational approximations are asymptotically equivalent to maximum likelihood estimators, and we show empirically, for the simple example of a first-order autoregressive model with missing values, that the lower bound surface can be very similar in shape to the true log-likelihood in non-asymptotic situations.     

A comparison of multiple imputation and doubly robust estimation for analyses with missing data

Summary.  Multiple imputation is now a well-established technique for analysing data sets where some units have incomplete observations. Provided that the imputation model is correct, the resulting estimates are consistent. An alternative, weighting by the inverse probability of observing complete data on a unit, is conceptually simple and involves fewer modelling assumptions, but it is known to be both inefficient (relative to a fully parametric approach) and sensitive to the choice of weighting model. Over the last decade, there has been a considerable body of theoretical work to improve the performance of inverse probability weighting, leading to the development of 'doubly robust' or 'doubly protected' estimators. We present an intuitive review of these developments and contrast these estimators with multiple imputation from both a theoretical and a practical viewpoint.     

A Bayesian hierarchical model for categorical longitudinal data from a social survey of immigrants

Summary.  The paper investigates a Bayesian hierarchical model for the analysis of categorical longitudinal data from a large social survey of immigrants to Australia. Data for each subject are observed on three separate occasions, or waves, of the survey. One of the features of the data set is that observations for some variables are missing for at least one wave. A model for the employment status of immigrants is developed by introducing, at the first stage of a hierarchical model, a multinomial model for the response and then subsequent terms are introduced to explain wave and subject effects. To estimate the model, we use the Gibbs sampler, which allows missing data for both the response and the explanatory variables to be imputed at each iteration of the algorithm, given some appropriate prior distributions. After accounting for significant covariate effects in the model, results show that the relative probability of remaining unemployed diminished with time following arrival in Australia.