Publication Type: Paper/Poster Proposal
Abstract: In this paper we will try to get a better understanding of the question-answer process in structured interviews by studying the sequences of coded utterances of the interviewer and the respondent, and by trying to fit a model to these data. Ideally, structured interviews give rise to very short and simple sequences: correctly worded question, admissible answer, acknowledgement of perceipt of that answer (Maynard & Schaeffer, 2002). However, in practice, many of such sequences deviate from this simple paradigm. Sequences may deviate for a wide variety of reasons: both respondent and interviewer may behave inappropriately by misphrasing or misinterpreting questions, becoming engaged in social conversation, suggesting certain answers, etc. Such behavior contributes to measurement error. A better understanding of the mechanisms that generate non-paradigmatic sequences could lead to improved interviewing techniques, improved question wording and improved interviewer training. In the past decades, several attempts have been made to explain the frequent occurrence of non-paradigmatic sequences. These explanations (e.g. Suchman & Jordan, 1991 and Tourangeau et. al, 1988) have been formulated as quite general and qualitative theoretical frameworks, the tenability of which is hard to test empirically. In this paper we will make an attempt to formalize these cognitive frameworks in order to test their tenability and relate model-parameters to characteristics of questions, interviewers and respondents. The models discussed are so-called Interactive Hidden Markov Models. This class of models are used to mimic the interactive processes and allow for hidden Markov chains that are affected by observable utterances of both the respondent and the interviewer. An interpretation of the hidden states of the modelled process may shed new light on understanding the interaction between respondent and the interviewer.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: There is growing evidence that use of advance letters and answering machine messages in list sample telephone studies can improve cooperation rates and reduce initial refusals. However, there is less known about their effect in random-digit dial (RDD) surveys. The research presented here examines the results of two experiments to assess the effect on response rates and initial refusal rates of sending advance letters to and leaving scripted messages on the answering machines of BRFSS sample members. The data come from an experimental study conducted with 1,600 respondents (800 in each experiment – 400 treatment and 400 controls in each) in several states, following standard BRFSS protocol. Additionally, respondents were asked several questions about the advance letters and answering machine messages, such as did the selected respondent actually see the letter (or hear the message)? Did the letter (or message) influence their determination to participate in the survey? The analysis examines the impact of each of these approaches on response rates and initial refusal rates.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Most public old-age pension schemes around the world are based on the pay-as-you-go defined benefit (PAYGO DB) model. As these schemes have matured and some of the limitations of this model have become more salient, pension experts have begun considering alternative models. The notional defined contribution (NDC) model has emerged as one of the major new models. This paper has three goals: (1) to provide a brief history of pension policy in China, (2) to describe the core elements of the NDC model, and (3) to assess the relative merits of the NDC alternative as a possible option for China. For China the NDC's mechanisms for dealing with population aging and regional differences in wage levels would be definite strengths, particularly in light of the projected consequences of the nation's one child policy. The transparency of the NDC model would be advantageous given China's corruption problems and the de facto partial default in connection with the prior enterprise based pay-as-you-go defined benefit scheme. The NDC model provides an incentive for workers to remain in the labor force longer which would help alleviate the nation's projected dependency burden, but at the cost of exacerbating another problem, the high unemployment rates among young workers. The fear that the NDC model might contribute to higher unemployment rates among younger worker indirectly contributing to labor unrest and political instability could by itself make the model politically unacceptable to Chinese policy makers.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper discusses conditions favoring the use of force by young Russian-Germans from the former Soviet Union after their migration to Germany. The authors focus on the question of what kind of influences these young people‘s understanding of honor and the functions of the police has on their attitude toward force and their use of it. In addition to a presentation of the state of the art on prevalence as well as the motives of the use of force, they refer to their own empirical study of young Russian-Germans in the cities of Duisburg and Frankfurt/M., Germany. Under the conditions of great language deficits as well as bad educational and occupational opportunities, the concept of honor based upon a traditional ideal of masculinity turns out to be an effective means of social control which guides the actions of these young people to a considerable extent.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: Stress has been implicated in the development of substance dependence, possibly as a consequence of maladaptive coping behavior. This relationship was previously examined separately with respect to drugs and alcohol using a representative sample of nearly 1800 community residing young adults (most aged 19-21) in South Florida. The relationship between level of lifetime exposure to adverse experiences and the initial onset of both alcohol and drug dependence consisted of independent effects stress measured both distally and proximally. This pattern of relationship held in the context of controls for prior psychiatric disorders, AD/HD and conduct disorder, as well as gender, race/ethnicity, socioeconomic status and time. While stress was shown to be an important predictor of drug and alcohol dependence disorder, variation in stress exposure failed to explain social group differences in either type of dependence. The observed statistical patterns were the same for alcohol and drug dependence, but most of the affected individuals were different people in the two analyses. This paper explores whether finding the same pattern of relationship with respect to different substance dependencies is redundant, or whether the largely independent manifestation of alcohol and drug dependence in different stress-exposed individuals has sociologically significant meaning.