Refine
Document Type
- Article (14)
- Other (4)
- Postprint (2)
- Part of a Book (1)
- Contribution to a Periodical (1)
- Preprint (1)
Language
- English (23) (remove)
Is part of the Bibliography
- yes (23)
Keywords
- COVID-19 (3)
- allocation policies (3)
- integration (3)
- language acquisition (3)
- language courses (3)
- refugees (3)
- rural (3)
- Eingaben (2)
- German Democratic Republic (GDR) (2)
- Participation (2)
Institute
Since COVID-19 became a pandemic, many studies are being conducted to get a better understanding of the disease itself and its spread. One crucial indicator is the prevalence of SARS-CoV-2 infections. Since this measure is an important foundation for political decisions, its estimate must be reliable and unbiased. This paper presents reasons for biases in prevalence estimates due to unit nonresponse in typical studies. Since it is difficult to avoid bias in situations with mostly unknown nonresponse mechanisms, we propose the maximum amount of bias as one measure to assess the uncertainty due to nonresponse. An interactive web application is presented that calculates the limits of such a conservative unit nonresponse confidence interval (CUNCI).
Leben in der ehemaligen DDR
(2019)
The long-standing approach of using probability samples in social science research has come under pressure through eroding survey response rates, advanced methodology, and easier access to large amounts of data. These factors, along with an increased awareness of the pitfalls of the nonequivalent comparison group design for the estimation of causal effects, have moved the attention of applied researchers away from issues of sampling and toward issues of identification. This article discusses the usability of samples with unknown selection probabilities for various research questions. In doing so, we review assumptions necessary for descriptive and causal inference and discuss research strategies developed to overcome sampling limitations.
A review of all research papers published in the European Sociological Review in 2016 and 2017 (N = 118) shows that only a minority of papers clearly define the parameter of interest and provide sufficient reasoning for the selected control variables of the statistical analysis. Thus, the vast majority of papers does not reach minimal standards for the selection of control variables. Consequently, a majority of papers interpret biased coefficients, or statistics without proper sociological meaning. We postulate that authors and reviewers should be more careful about control variable selection. We propose graphical causal models in the form of directed acyclic graphs as an example for a parsimonious and powerful means to that end.
The long term relationship between medicaid expansion and adult life-threatening chronic conditions
(2023)
We test whether the expansions of children's Medicaid eligibility in the 1980s–1990s resulted in long-term health benefits in terms of severe chronic conditions. Still relatively rare in the field, we use prospective individual-level panel data from the Panel Study of Income Dynamics (PSID) along with the higher quality income measures from the Cross-National Equivalent File (adjusting for taxes, transfers and household size). We observe severe chronic conditions (high blood pressure/heart disease, cancer, diabetes, or lung disease) at ages 30–56 (average age 43.1) for 4670 respondents who were also prospectively observed during childhood (i.e., at ages 0–17). Our analysis exploits within-region temporal variation in childhood Medicaid eligibility and adjusts for state- and individual-level controls. We uniquely concentrate attention on adjusting for childhood income. A standard deviation greater childhood Medicaid eligibility significantly reduces the probability of severe chronic conditions in adulthood by 0.05 to 0.12 (16%–37.5% reduction from mean 0.32). Across the range of observed childhood Medicaid eligibility, the probability is approximately cut in half. Greater childhood Medicaid eligibility also substantially reduces childhood income disparities in severe chronic conditions. At higher levels of childhood Medicaid eligibility, we find no significant childhood income disparities in adult severe chronic conditions.
Maintaining quality
(2015)
Survey Research Methods has slightly revised its publication policies. Firstly, starting with the publication of this Editorial, SRM will accept - under specified conditions - manuscripts that discuss experiments in non-probability samples for peer-review. Secondly, SRM will require authors to publish replication materials of their study as Online supplement to their article. Finally, Survey Research Methods will publish replication studies of articles published in the journal. This Editorial gives reasons for these changes.
This paper compares the usability of data stemming from probability sampling with data stemming from nonprobability sampling. It develops six research scenarios that differ in their research goals and assumptions about the data generating process. It is shown that inferences from data stemming from nonprobability sampling implies demanding assumptions on the homogeneity of the units being studied. Researchers who are not willing to pose these assumptions are generally better off using data from probability sampling, regardless of the amount of nonresponse. However, even in cases when data from probability sampling is clearly advertised, data stemming from nonprobability sampling may contribute to the cumulative scientific endeavour of pinpointing a plausible interval for the parameter of interest.