Sektion Wissenssoziologie

Ohne Zusammenfassung unter Mithilfe von Jule Lorenzen, Sabine Petschke, Lisa-Marian Schmidt und René Tuma     

On Integer Programming Approaches for Morphological Template Decomposition Problems in Computer Vision

In morphological image processing and analysis, a template or structuringelement is applied to an image. Often savings in computation time and abetter fit to the given computer architecture can be achieved by using thetechnique of template decomposition. Researchers have written a multitude ofpapers on finding such decompositions for special classes of templates.Justifying recent integer programming approaches to the morphologicaltemplate decomposition problem in its general form, this paper proves theNP-completeness of this problem.     

THE DIFFUSION OF MEDICAL TECHNOLOGY: SOCIAL NETWORK ANALYSIS AND POLICY RESEARCH

The purpose of this article is to demonstrate how social network analysis can be used to provide information for policy decisions pertaining to physicians' adoption and utilization of new medical technology. Interviews and hospital records were used to obtain data on refenals, consultations, discussions, and on-call coverage; utilization of a computer-based hospital information system; and personal and practice characteristics from 24 physicians who belong to a private group practice. The results of a block-model analysis suggest that the physician's relative position in the network is an important determinant of his/her participation in the diffusion process. A number of policy implications related to the introduction of new medical technology into practice settings are discussed.     

Principal component and correspondence analysis of compositional data: some similarities

Principal component and correspondence analysis can both be used as exploratory methods for representing multivariate data in two dimensions. Circumstances under which the, possibly inappropriate, application of principal components to untransformed compositional data approximates to a correspondence analysis of the raw data are noted. Aitchison (1986) has proposed a method for the principal component analysis of compositional data involving transformation of the raw data. It is shown how this can be approximated by a correspondence analysis of appropriately transformed data. The latter approach may be preferable when there are zeroes in the data.     

Who Cares If Police Become Violent? Explaining Approval of Police Use of Force Using a National Sample

Understanding police violence is important in part because police violence, if generally tolerated, could transform a society based on law into one governed by political and personal whim. This research asks whether the conflict perspective can explain which groups are more likely to approve of police use of force, and whether several dimensions of power are at issue or just a few. Previous studies have found that race is the most important factor in determining approval of police use of force. Gender, age, class, and other variables have shown mixed results. Using a national sample (1998 General Social Survey), this paper contributes a more comprehensive examination of this issue to the small body of literature in this area. We look at approval of police use of force in five scenarios. Our findings show that minorities and women are more likely to disapprove of police use of force. Supporting a narrow application of conflict theory, other dimensions of power appear not to affect attitudes toward police hitting citizens. Context of violence matters too. When situations are not actively threatening for officers, by near consensus, violence is unacceptable. Disagreement comes from less clear-cut circumstances.     

Regularities and irregularities in a random 0,1 sequence

A few properties of the random 0,1 sequences are studied. These properties can be utilized to test the randomness of a 0,1 sequence produced by a computer (random generator). This test seems to be appropriate for discovering the periodicity (or the length of the period) of a pseudorandom sequence. A possible application of the presented results in regression analysis is also discussed.     

Revisiting the Moderating Effect of Social Support on Strain: A Gendered Test

Agnew's General Strain Theory (GST) has both revived and advanced strain theory. It has recently been applied to topics ranging from gender differences in delinquency to differences in community crime rates. Tests of GST are now focusing on the intermediate variables that fall between GST and delinquency, such as anger and social support, but these tests have produced varying results. This paper adds to GST literature by empirically analyzing whether social support moderates individuals' responses to strain and whether this moderating effect varies by gender. Testing is conducted using data from wave six of the National Youth Survey (NYS). Results provide support for GST and suggest that there are gender differences in the types and levels of strain experienced during the late teens. Further, there is evidence to support the hypothesis that social support does moderate females' responses to certain types of strain.     

Minimax-variance L- and R-estimators of location

We consider the problem of minimax-variance, robust estimation of a location parameter, through the use of L- and R-estimators. We derive an easily checked necessary condition for L-estimation to be minimax, and a related sufficient condition for R-estimation to be minimax. Those cases in the literature in which L-estimation is known not to be minimax, and those in which R-estimation is minimax, are derived as consequences of these conditions. New classes of examples are given in each case. As well, we answer a question of Scholz (1974), who showed essentially that the asymptotic variance of an R-estimator never exceeds that of an L-estimator, if both are efficient at the same strongly unimodal distribution. Scholz raised the question of whether or not the assumption of strong unimodality could be dropped. We answer this question in the negative, theoretically and by examples. In the examples, the minimax property fails both for L-estimation and for R-estimation, but the variance of the L-estimator, as the distribution of the observation varies over the given neighbourhood, remains unbounded. That of the R-estimator is unbounded.