Comparing mean ranks for repeated measures data

Rank tests are considered that compare t treatments in repeated measures designs. A statistic is given that contains as special cases several that have been proposed for this problem, including one that corresponds to the randomized block ANOVA statistic applied to the rank transformed data. Another statistic is proposed, having a null distribution holding under more general conditions, that is the rank transform of the Hotelling statistic for repeated measures. A statistic of this type is also given for data that are ordered categorical rather than fully rankedo Unlike the Friedman statistic, the statistics discussed in this article utilize a single ranking of the entire sample. Power calculations for an underlying normal distribution indicate that the rank transformed ANOVA test can be substantially more powerful than the Friedman test.     

Abstracts

Objective: To ascertain if two commonly used health status measures could be successfully self-administered to a group of older men recruited from the community to volunteer for a clinical research project and to correlate responses to the health status questionnaire to testosterone levels. Methods: In a cross-sectional survey 93 men, age 70 or greater, responded to a random mailing to a suburban community. Sixty-eight men were eligible and 47 participated in the study. Two health status questionnaires, the Sickness Impact Profile (SIP) and the Short-form 36 (SF-36), were mailed to participants with instructions to complete the surveys on the same day and return them by mail. Blood samples were obtained for bioavailable and total testosterone measurements. Pearson correlation coefficient analysis was used to compare the responses to the SF-36 and SIP and testosterone levels. Results: Of the eligible volunteers 69% participated and were able to self-administer the SIP and the SF-36. Summation scores for the surveys were above those reported in similar age populations. Perfect scores were reported frequently in all subsets of the SIP; perfect scores were reported in 43-92% of men on various subscales. The responses to the SF-36 were also frequently perfect although perfect scores ranged from 2 to 82% on various subsets. There was moderate correlation between the results from the two surveys. There was no correlation between age, SIP or SF-36 and testosterone level. Conclusions: Healthy older men volunteering for clinical research were able to self-administer both the SIP and SF3-6. Responses to both surveys were skewed toward perfect scores; the SIP failed to reveal many men with functional deficits. Bioavailable testosterone levels did not predict perceived health in this group of men.     

“A stone wall out of the earth that thundering cannon cannot destroy"? Bastion and Moat at the Castle,Cape Town

Although the urgency with which the Dutch East India Company administration at the Cape built defensive works was a direct result of their fear of attack from both land and sea, there was an additional imperative, for the Castle at Cape Town stood as a symbol of Dutch colonial aspirations. In this paper, we compare written and material texts of one part of the Castle's architecture to show firstly, how the discordances between such sources can reveal an underside of early colonization and secondly, how the Castle moat, in the seventeenth century a stamp of aspirant power, was by the mid‐nineteenth century a dump and a public nuisance.     

Battling the Plantation Mentality: Memphis and the Black Freedom Struggle

Theodore Evergates, Feudal Society in the Bailliage of Troyes under the Counts of Champagne, 1152–1284 (1976), xii+274 (Johns Hopkins University Press, Baltimore and London, £11.20).

Pierre Guichard, Structures sociales ‘orientales’ et ‘occidentales’ dans l'Espagne musulmane (1977), 426 (Editions Mouton et cie, Paris and La Haye, 176F).

J. S. Cockburn (ed.), Crime in England 1550–1800 (1977), xiv+334 (Methuen, £10.00, paperback £6.95).

Jack Goody, Joan Thirsk and E. P. Thompson (eds), Family and Inheritance. Rural Society in Western Europe, 1200–1800 (1976), vi+421 (Cambridge University Press, Cambridge, £8.80); Peter Laslett, Family Life and Illicit Lore in Earlier Generations. Essays in Historical Sociology (1977), vii+270 (Cambridge University Press, Cambridge, £15.00, paperback £4.95).

David C. Itzkowitz, Peculiar Privilege: A Social History of English Foxhunting, 1753–1885 (1977), 248 (The Harvester Press, Hassocks, £10.50).

Alfred F. Young (ed.), The American Revolution: Explorations in the History of American Radicalism (1976), xv+481 (Northern Illinois University Press, DeKalb, Illinois, paperback, n.p.).

Clive Bush, The Dream of Reason: American Consciousness and Cultural Achievement from Independence to the Civil War (1977), xi+397 (Edward Arnold, £14.95).

Albert Soboul, Problèmes paysans de la Révolution, 1789–1848 (1976), 446 (François Maspéro, Paris, 50F.).

James H. Bater, St Petersburg. Industrialization and Change (1976), xxiii+474 (Edward Arnold, £14.95).

Nicholas V. Riasanovsky, A Parting of the Ways: Government and the Educated Public in Russia 1801–1851 (1977), viii+323 (Clarendon Press, Oxford, £12.00).

J. Michael Phayer, Sexual Liberation and Religion in Nineteenth‐Century Europe (1977), 176 (Croom Helm, £7.50).

John Woodward and David Richards (eds.), Health Care and Popular Medicine in Nineteenth‐Century England (1977), 195 (Croom Helm, £7.95).

Jörg Calliess, Militär in der Krise. Die bayerische Armee in der Revolution 184849 (1976), vi+224 (Harald Boldt, Boppard am Rhein, DM 40).

Trygve Tholfsen, Working Class Radicalism in Mid‐Victorian England (1976), 332 (Croom Helm, £9.95).

Daniel Field, The End of Serfdom: Nobility and Bureaucracy in Russia, 1855–1861 (1976), xiii+472 (Harvard University Press, Cambridge, Mass, and London, £11.90).     

Die Auswirkungen von Neid auf individuelle Leistungen: Ergebnisse einer Panelanalyse

Relative income differences are likely to lead to envy within a reference group. Envy in turn influences social behavior and individual performance. While positional concerns are apparent in daily life, empirical evidence is rare in the economic literature. This paper investigates the impact of the relative income position on individuals’ performance or productivity. As ?performance“ is difficult to measure we turn to soccer players whose performance has been well documented. The broad sample covers eight seasons of the German premier soccer league (Bundesliga) between 1995 and 2004, and includes 1040 players, a salary proxy and several performance variables. The results show that player performance is strongly affected by relative income position. A disadvantage in the relative income position reduces productivity. The larger the income differences within a team, the stronger are the effects of positional concern. Team composition also significantly affects behaviour.