Using the chi-square statistic to monitor compositional process data

SUMMARY We investigate the use of the chi-square control chart as a simple multivariate method for shopfloor monitoring of compositional process data. Although this chart is usually considered to be applicable only with multinomial process data, we show that it is also valid, in a certain asymptotic sense, for compositional data that arise from the Dirichlet distribution. For general compositional data, we show that the chi-square statistic can be used for process monitoring, provided that we make a simple adjustment to the degrees of freedom in the chi-square reference distribution. This method is illustrated and compared in four examples with the T 2 chart based on log-ratio transformation of the data.     

From disability to handicap: an inevitable response to social constraints?

La recherche sociologique et socio-psychologique sur la sclérose en plaques (SP) suggère que l'ajustement subséquent au diagnostic chez le patient se présente comme un choix entre deux stratégies: 'normalisation' ou 'retrait'. Mais en postulant un tel choix de stratégies, on ignore les facteurs sociaux qui influencent la façon dont les individus frappés d'une telle maladie chronique réagissent. En fait, la recherche indique que les individus atteints de SP choisissent des stratégies d'ajustement, mais ceci dans un contexte social et culturel qui limite sévèrement l'éventail de choix disponibles et qui affecte la probabilité de faire un choix plutôt qu'un autre. Les données présentées ici ont trait à quelques-uns de ces facteurs sociaux: la réponse de l'époux à la transformation de la relation conjugate, et l'attitude des médecins et des employeurs face à la situation d'emploi de la personne chez qui SP a été diagnostiquée.
Previous sociological and social-psychological research on the chronic illness multiple sclerosis (MS) and post-diagnosis adjustment to the illness has posited a dichotomous patient response: 'normalization' or 'withdrawal'. In this paper it is argued that the posited choice of strategies ignores the social factors that influence how individuals diagnosed with a chronic illness such as MS respond. In fact, the research findings indicate that people with MS choose adjustment strategies, but they do so in a social and cultural context that powerfully constrains both the range of choices available to them and the probability of making one choice rather than another. Data presented here pertain to the response of spouses to the marital relationship, and of doctors and employers to labor force participation of individuals diagnosed with MS.     

Gambling in revolutionary Paris — The Palais Royal: 1789–1838

By the revolution of 1789, the four-story, quadrangular Palais Royal in Paris had become the most glittering tourist center of Europe, with 180 shops and cafes in its ground floor arcades. By 1791, its basement and secondary story contained over 100 separate, illicit gambling operations featuring the most popular dice and card games. The mania for gambling had been transferred from defunct, monarchical Versailles to the thriving, bourgeois Palais Royal, where the five main gaming clubs throbbed from noon till midnight. During the Revolution, Prince Talleyrand won 30,000 francs at one club, and after Waterloo in 1815, Marshal Blucher lost 1,500,000 francs in one night at another. To bring the situation under control and raise taxes for the state, in 1806 Napoleon legalized the main clubs, which from 1819 to 1837 grossed an enormous 137 million francs. When the anti-gambling forces triumphed in 1837 and the clubs were closed down, the National Guard had to be called out to evict the mobs of gamblers who refused to leave the tables. Dramatic reports from Revolutionary police raids, and quotations from the memoirs of humorous French gamblers and shocked foreign visitors, provide anecdotal illustrations of the 49 years during which the Palais Royal was the most intriguing and picturesque gambling mecca of Europe—and probably of the world.A longer version of this article, with more complete bibliography and footnotes, can be found in W.R. Eadington and J.A. Cornelius (eds.),Gambling and Public Policy: International Perspectives, Institute for the Study of Gambling and Commercial Gaming, University of Nevada, Reno, 1991.     

PREDICTING MARITAL THERAPY DROPOUTS*

The purpose of this study was to predict therapy dropouts using data gathered at marital therpay intake. Intake data included standardized assessments and demographic information. The sample consisted of 474 couples who were seeking marital therapy and who attended at least one therapy session. Seventy-two couples met the dropout criteria, and a random sample of equal size was drawn from the rest of the couples for the analyses. Data from these two groups of couples were analyzed with multiple regression and discriminant analysis formulas. Three variables-having less than two children, having a male intake clinician, and a presenting problem relating only to one spouse-were significant predictors in accounting for couples who would drop out of therapy. In the best model, high phobic anxiety scores for husbands and a presenting problem relating to parenting, along with the other three variables, produced a model that accurately classified 82% of the couples who dropped out of therapy.     

The Navajo–U.S. Population Mortality Crossover since the Mid-20th Century

Mortality crossovers at older ages have been observed when comparing different populations, particularly disadvantaged populations with advantaged populations. A growing body of research indicates that mortality convergences to actual crossovers are real and not a result of overstating of age at the older ages. Only recently have the mortality experiences of Native Americans been compared with those of other Americans; specific Native American tribal populations have not been examined, however. Presented here are the mortality experiences of the Navajo and those of the total U.S. population and U.S. white population since the mid-20th century. Comparison provides further support to findings that convergences and crossovers actually occur between disadvantaged and advantaged populations.     

Decision making with stochastic dominance: An expository review

This paper surveys the use of stochastic dominance to decision making under uncertainty. The first part presents the relevant definitions and some properties of distributions satisfying one of the stochastic dominance conditions. These properties include restrictions on moments, an invariance property, and properties of random variables related by an exact formula.The second part contains some applications of the stochastic dominance method and especially the problem of selecting optimal portfolios. Most of the results in this section deal with conditions that make diversification an optimal strategy.     

NONVERBAL RECIPROCITY IN NONDISTRESSED MARITAL PARTNERS: AN EXAMINATION OF BASE RATE CHANGE*

The present study sought to incorporate recent suggestions for reciprocity research into an experimental design which could use behavioral base rate change as a method for determining nonverbal reciprocity in nondistressed couples. This study examined reciprocity by surreptitiously manipulating the touching behavior of one spouse, and monitoring responses of the other spouse. Forty couples were randomly assigned to either an experimental or control group. Couples were videotaped engaging in an assigned task which permitted interaction. During the interaction, positive touches, smiles, and headnods were monitored. Following instruction to do so, selected spouses in the experimental group increased their rate of touches. Results indicated that, in response, naive spouses subsequently increased their rate of touches. The results of this experiment provide supportive evidence for the lawful nature of reciprocity. The implications of these findings, and suggestions for subsequent reciprocity research, are forwarded.     

Rhythmic organization of social interaction and observer ratings of positive affect and involvement

Many investigators have speculated that a high degree of rhythmic patterning of social interaction and close coupling between the activity rhythms of partners are associated with positive affect, attachment, and interpersonal attraction. Others suggest that predictable or rhythmic patterning is an indication of inflexibility, and that close linkages between the physiological arousal of partners are associated with negative affect. In the present study, spectral and cross spectral analyses were applied to vocal activity and heart rates recorded during 12 conversations. Indexes of rhythmicity and strength of coupling were derived and used as predictors of observer ratings of pleasantness of affect and degree of involvement. There was a curvilinear relationship between affect and rhythmicity of vocal activity, such that moderately rhythmic social interactions were evaluated most positively. This relationship remained statistically significant even when other variables (such as mean and standard deviation of amount of vocal activity) were statistically controlled. Strength of coupling between partner vocal activity patterns and strength of coupling between the vocal activity and heart rate within speaker were not statistically significant predictors of ratings of pleasantness. Results suggest that there may be an optimum degree of rhythmicity in social interaction, with moderately rhythmic interactions evaluated most positively.We gratefully acknowledge support from the Central University Research Fund and from the Office of the Dean of Liberal Arts at the University of New Hampshire. Thanks to Kim Mooney, Charles Dufour, Mark Henn and Jay Goober for their contribution in developing the coding system.