Analysis of variance of paired data without repetition of measurement

Summary The objective of this analysis of variance of paired data is to estimate positive random error variances for each ofN=2 measurement methods. The two methods measure the same item only once without measurement repetition. The well-known unbiased Grubbs’ estimators are not suitable for practical purpose because they can become negative. With the help of Chebyshev’s inequality the probability was determined that Grubbs’ estimators become negative. Based on the Grubbs’ estimators new estimators were derived. The new estimators are indeed always positive, but they are biased. It is shown that the biases are small. In case the Grubbs’ estimators are positive a bias correction of the new estimators may be envisaged.

---

Zusammenfassung Das Ziel dieser Varianzanalyse von gepaarten Messungen ist die Sch?tzung zuf?lliger Messfehlervarianzen für jede derN=2 Messmethoden. Die beiden Messmethoden messen das gleiche Merkmal eines Elements nur einmal ohne Messwiederholung. Die bekannten unverzerrten Grubbs-Sch?tzer sind für die praktische Anwendung nicht geeignet, weil sie negativ werden k?nnten. Die Tschebyscheffsche Ungleichung wurde genutzt, um die Wahrscheinlichkeit zu ermitteln, dass Grubbs-Sch?tzer negativ werden. Basierend auf Grubbs-Sch?tzern wurden neue Sch?tzer hergeleitet. Diese neuen Sch?tzer sind zwar immer positiv, aber verzerrt. Es wird gezeigt, dass die Verzerrungen klein sind. Für den Fall, dass die Grubbs-Sch?tzer positiv ausfallen, k?nnte eine Korrektur der Verzerrung in Betracht gezogen werden.

     

What Makes an Entrepreneur and Does it Pay? Native Men,Turks, and Other Migrants in Germany

This paper focuses on the entrepreneurial endeavours of immigrants' and natives in Germany, concentrating on Turks, Germany's largest immigrant group and one under‐studied in the literature. Self‐employed Turks in Germany represent about 70 per cent of all Turkish entrepreneurs in the European Union. We use data from the German Socio‐economic Panel to study patterns of self‐employment. First, we identify the characteristics of the self‐employed individuals and understand their underlying drive into self‐employment. Next we investigate how immigrant entrepreneurs fare in the labour market and compare their earnings to those of the natives. It is important for decision makers to understand entrepreneurial patterns so that they can shape policy that better fosters entrepreneurial activities. This paper presents several findings that can inform better policymaking. First, our investigation indicates that education is not decisive in determining whether one will choose self‐employment over salaried work nor in explaining earnings. The estimated age‐earnings profiles are the same for natives and immigrants, while the proclivity to become self‐employed is concave with respect to age for both groups. Immigrants' start with a higher probability to work than natives but have a slower increase in the self‐employment probabilities thereafter. The earnings of self‐employed immigrants' are higher initially, but their earnings path crosses eventually that of the natives. Second, we find some suggestion of ethnic entrepreneurial spirit. Turks are 70 per cent more likely to be self‐employed than any other immigrant group, although they do not necessarily earn more. These patterns should be further explored.     

Eskalierendes Commitment gegenüber Mitarbeitern: Ein Rahmenmodell

Zusammenfassung:  Oft werden zus?tzliche Ressourcen in Mitarbeiter investiert, obwohl sie die in sie von ihren Vorgesetzten formulierten Erwartungen nicht erfüllen, und augenscheinlich die Trennung von ihnen die angemessene Entscheidung w?re. Ein solches Verhalten kann als „eskalierendes Commitment“ bezeichnet werden (Staw, 1997). In diesem Beitrag wird untersucht, welche Konsequenzen es hat, dieses „Festhalten“ an wenig geeigneten Mitarbeitern tats?chlich als „eskalierendes Commitment“ zu beschreiben. Hierzu werden zum einen Erkenntnisse aus der bisherigen Forschung zum eskalierenden Commitment herangezogen, zum anderen werden ausgehend von Kernmerkmalen einer Arbeitsbeziehung vier weitere Faktoren diskutiert, die eskalierendes Commitment gegenüber Mitarbeitern verst?rken k?nnen: (1) Sympathie, (2) Prognoseungewissheit bzgl. des vermuteten Potentials des Mitarbeiters, (3) dem Mitarbeiter unterstellte Motivationsprobleme und (4) M?glichkeiten zur Reinterpretation der Anforderungen an den Mitarbeiter. Abschlie?end werden Deeskalationsstrategien diskutiert

Determinants of voluntary assurance on sustainability reports: an empirical analysis

Driven by the ongoing discussion of corporate responsibility, growing numbers of companies have been publishing what have become known as sustainability reports. These reports have in part been subject to voluntary external assurance. Although the percentage of assured reports is significant, the market for this kind of assurance is still in an early stage of development. In this context the present article will ascertain the theoretical determining factors in the demand for voluntary external assurance and subject these in relation to the markets in Germany, the Netherlands and Great Britain to empirical analysis. In the context of the development of hypotheses shaped by agency theory, the constructs of agency costs and signalling can be distinguished. Variables measuring these constructs will be examined. It can be shown that the type and scope of the reporting (i.e., the choice of the Global Reporting Initiative ‘application level’), the existence of a sustainability department, and the size of the company are associated with the demand for voluntary assurance. Additional control variables are examined. Here it appears that the demand for external assurance is highest in Great Britain. Finally, research opportunities in this field will be indicated and recent normative developments briefly sketched.     

Low species richness of non-biting midges (Diptera: Chironomidae) in Neotropical artificial urban water bodies

Chironomid assemblages of 22 artificial water bodies, mainly fountains, in two South American cities were surveyed. We found surprisingly low diversities, with a total of 11 taxa, averaging two taxa per site. The typical fountain assemblages mainly consisted of common species that have a wide distribution pattern and are tolerant to organic pollution. Also taxa independent of the natural aquatic sources, such as tap-water and semi-terrestrial species were represented. There was no significant difference between the taxa richness of the two S. American regions, however, the assemblage structures of the water bodies of Quito and Bogotá differed significantly. There was no correlation between sample similarities and distances among the sites either. The low number of taxa recorded can be attributed to the combination of naturally limited species pool and the heavy pollution in natural water bodies as colonization sources.     

A Typology of Consumers' Emotional Response Styles during Service Recovery Encounters*

Previous efforts investigating consumers' response styles during service failure and recovery encounters have focused on behavioural elements and, by and large, neglected the important issue of emotions elicited in this context. The present study uses experienced emotions during service recovery encounters as a way of classifying complainants and links the resultant typology to the relationship quality indicators of satisfaction, trust and commitment. By doing so, it serves as an empirically grounded basis for subsequently deriving strategies and tactics for successful recovery from service failures. The results reveal four distinct emotional response styles which are associated with perceptions of relationship quality following service recovery encounters. Managerial implications of the findings are discussed and future research directions identified.     

The Role of the Clinician in Natural Family Planning

Abstract

“The BCG Controversy: A Methodological and Statistical Reappraisal,” John D. Clemens, Jackie J. H. Choung, and Alvan R. Feinstein. Because of the conflicting results of eight major controlled trials, BCG vaccination against tuberculosis remains controversial despite more than 50 years of use. Suspecting a methodological source for the controversy, we reviewed the scientific and statistical quality of each trial. The analysis showed that (1) although biased allocation of the vaccine appeared an unlikely explanation for the disparate results, adequate demonstration of unbiased detection of tuberculosis was available only for the three trials reporting 75% or greater protective efficacy; and (2) in most trials reporting low efficacy, the results had wide confidence intervals that could not exclude high efficacy, but the trials reporting high efficacy all had narrow confidence intervals that excluded low efficacy. Because the trials with the best methodological quality and greatest statistical precision reported high efficacy, the evidence suggests that BCG can confer a high degree of protection against tuberculosis and that bias or inadequate statistical power may have contributed to the conflicting data. (Journal of the American Medical Association 1983;249:2362–2369.)

“Acquired Immunodeficiency Syndrome, Opportunistic Infections, and Malignancies in Male Homosexuals: A Hypothesis of Etiologic Factors in Pathogenesis,” Joseph Sonnabend, Steven S. Witkin, and David T. Purtilo. The acquired immunodeficiency syndrome (AIDS) occurs in a subgroup of male homosexuals having sexual contact with a large number of partners. Uncommonly, AIDS has also been diagnosed in Haitians, hemophiliacs, and intravenous drug users and their infants. Manifestations include autoimmune disturbances, opportunistic infections, Kaposi's sarcoma, chronic lymphadenomegaly, non-Hodgkin's lymphoma, or squamous cell carcinoma. The hypothesis receiving most consideration is that a yet-to-be-identifed virus causes AIDS. An alternative view is that repeated sexual involvement with multiple partners, in a subgroup of male homosexuals, exposes the men to the immunosuppressive impact of cytomegalovirus (CMV) and allogeneic semen. Antibody to asialo-Gm, and other antigens on sperm react with and impair lymphoid cells. We propose a biphasic process. First, a reversible acquisition phase of impaired T-cell immunoregulation permits reactivation of Epstein-Barr virus (EBV), and autoantibodies are produced by the activated B cells. If sexual activity continues at a high level, accumulating immune defects, including destruction of thymic epithelium, lead to a second, self-sustaining phase wherein cytotoxic lymphocytes fail to eliminate herpesvirus-infected cells. Evidence is mounting that Kaposi's sarcoma is caused by CMV and that EBV is responsible for the B-cell lymphomas in these patients. Multiple factors, rather than a novel virus, probably induce AIDS in male homosexuals. If this hypothesis is correct, then rational bases for prevention and intervention can be designed. (Journal of the American Medical Association 1983;249:2370–2374.)     

The two-sample <Emphasis Type="Italic">t</Emphasis> test: pre-testing its assumptions does not pay off

Traditionally, when applying the two-sample t test, some pre-testing occurs. That is, the theory-based assumptions of normal distributions as well as of homogeneity of the variances are often tested in applied sciences in advance of the tried-for t test. But this paper shows that such pre-testing leads to unknown final type-I- and type-II-risks if the respective statistical tests are performed using the same set of observations. In order to get an impression of the extension of the resulting misinterpreted risks, some theoretical deductions are given and, in particular, a systematic simulation study is done. As a result, we propose that it is preferable to apply no pre-tests for the t test and no t test at all, but instead to use the Welch-test as a standard test: its power comes close to that of the t test when the variances are homogeneous, and for unequal variances and skewness values |γ ₁| < 3, it keeps the so called 20% robustness whereas the t test as well as Wilcoxon’s U test cannot be recommended for most cases.