Median split, k-group split, and optimality in continuous populations

Continuous populations are grouped in many social, economic, medical, or technical fields of research. However, by grouping them, a lot of information provided by the continuous population is lost. Especially the median split, which is still adopted by many researchers, and its generalization to an equiprobable k-group split lead to a high efficiency loss. Here, this loss of information is investigated by analytical and numerical analyses for some typical symmetric and skew population distributions often found in applications. Various distribution parameters, numbers of groups, and split methods are taken from theoretical considerations and real data sets. Losses sometimes in excess of 50% can be reduced by optimal grouping.     

Der Hawthorne-Effekt – ein Effekt für viele Gelegenheiten?

It is shown that the concept of ?Hawthorne effect‘ was created many years after the Hawthorne studies in Western Electric Company. Presently, the ?Hawthorne effect‘ concept is increasingly used in social sciences and many other sciences. The concept is however used in different and occasionally digressive meanings. The question arises whether such a concept is useful at all. In addition, it is discussed why the rather imprecise reports of single Hawthorne field experiments have been spread as academic folklore, and why rather undefined concepts like the ?Hawthorne effect‘ are used so widely in psychology and social sciences.     

Einkommenseffekte beruflicher Weiterbildung

The claim for lifelong learning or vocational further training is most often set out without having clarified satisfactorily the individual gains which result from training activities. Based on human capital theory, which we extend by arguments from signaling- and game theory, the paper investigates for western and eastern Germany individual returns of training on income with longitudinal data (1996–1998) from the recently available “Mikrozensus-Panel”. A problem in identifying causal training effects is a selection bias caused by unobserved heterogeneity which – using conventional methods – leads to overestimation of the effects. Fixed- and Random-Effects-Models – which we discuss and confront empirically regarding the training issue – present a method for controlling the selection bias. The results show evidence for positive and significant training effects on income, although the actual amount of income growth is low. Whereas no differences by sex were found, separate analyses by age, skill levels and region indicate that not all of the subgroups benefit positively from further training activities. Therefore, the findings relativize the importance of social inequalities in the access to further training for the process of status attainment in general.     

Quadratic estimation in mixed linear models with two variance components

Explicit expressions for Bayes invariant quadratic estimates, biased and unbiased, are presented and proved to cover the entire class of admissible estimates in the considered classes. An unbalanced genetic model is studied for demonstration.     

Room assignment-rent division: A market approach

A group of friends consider renting a house but they shall first agree on how to allocate its rooms and share the rent. We propose an auction mechanism for room assignment-rent division problems which mimics the market mechanism. Our auction mechanism is efficient, envy-free, individually-rational and it yields a non-negative price to each room whenever that is possible with envy-freeness.We would like to thank seminar participants at Barcelona, Boston College, Duke, Koç, Málaga, MIT, Michigan, NYU, conference participants at the First Brazilian Workshop of Game Theory Society (Sao Paulo 2002), SED 2002-Conference on Economic Design (New York City 2002), The Sixth International Meeting of the Society for Social Choice and Welfare (Pasadena 2002), an anonymous referee for their comments and Ahmet Alkan, Selçuk Karabati, Bari Tan, nsan Tunali for insightful discussions. Sönmez gratefully acknowledges the research support of KoçBank via the KoçBank scholar program and Turkish Academy of Sciences in the framework of the Young Scientist Award Program via grant TS/TÜBA-GEBP/2002-1-19. Any errors are our own responsibility.     

Effects of multicomponent training on functional fitness in older adults

The purpose of this study was to evaluate the effects of a 9-week supervised multicomponent exercise program on functional fitness and body composition in independent older adults. Forty-two adults age 60-86 years were randomly assigned to an exercise or a control group and were evaluated before and after training. The training program consisted of 3 sessions of walking, strengthening, and flexibility exercises per week. The multicomponent training program resulted in significant (p < .005) improvements on the chair stand, arm curl, 6-min walk, and up-and-go tests. The findings of this study indicate that a 9-week training program increased upper and lower body strength, aerobic endurance, and agility/dynamic balance in older adults. The most affected components of functional fitness were lower body strength and aerobic endurance. There was no effect of the 9-week training on body composition.