Simulation study of confidence regions for canonical variate analysis

Canonical variate analysis often involves the construction of confidence regions round points representing group means in a 2-dimensional plot. Traditionally circles have always been constructed, but some authors have recently advocated ellipses as being more appropriate. This paper describes a Monte Carlo study investigating the effect of a range of factors on the inclusion rates of true population means within both types of region for normal data. The traditional circles do not perform too badly within a restricted range, but they are nearly always under-included. The ellipses usually have higher inclusion rates, and so are often closer to the nominal rate, but are sometimes over-included.     

HIGH SCHOOLS AND CRIME: A REPLICATION

Using Cleveland data, we replicate Roncek and LoBosco's study of the effect of proximity to San Diego's high schools on crime in their surroundings. We also examine a major alternative hypothesis whether having other non-residential land uses in the schools' surroundings accounts for crime in these areas. Our findings closely parallel theirs. Proximity to public high schools only increases crime on city blocks which are immediately adjacent to the schools. Second, the size of a school's enrollment is not important for explaining crime in its surroundings. The analysis of non-residential effects supports Roncek and LoBosco's claim that crime effects are due solely to the presence of public high schools, although other land uses affect crime near private high schools. While there are differences in the exact magnitudes of the effects, the substance of their findings for San Diego generalizes well to a very different city. More generally, we argue that Wirth's early arguments about the effect of urbanism on impeding social control can be extended to the residential areas within the city.     

POLITICAL ECONOMY AND THE SOCIAL MARGINAL COST OF PUBLIC FUNDS: THE CASE OF THE MELTZER‐RICHARD ECONOMY

In previous studies on the social marginal cost of public funds (SMCF), the existing tax system has been assumed to be either arbitrary or optimal. This note explores another possibility: the existing tax system itself represents a political equilibrium. Our exploration proceeds in Meltzer and Richard’s (1981) political economy of redistributive taxation. An interesting feature of our finding is that the degree of income inequality as measured by the ratio of mean to median income can play an important role in estimating the SMCF and judging whether the level of redistribution is excessive or inadequate. (JEL D61, D72, H21)     

Children's Vocational Development: A Research Rationale

Vocational development research and interventions have focused primarily on adolescents and young adults. The lack of attention to career development antecedents in children has led to a serious neglect of this period of life when the foundation is laid for career choices and outcomes in later life. A harmful by‐product is the frequent preclusion of gender‐atypical occupational pathways by boys and especially by girls. To address this situation, the authors recommend identifying a core set of constructs that describe children's vocational development and developing sound instruments to measure them, leading to a longitudinal study ranging from childhood to early adulthood.     

Family,marital, and occupational satisfaction for recovering pathological gamblers

This paper describes the efforts of a private gambling treatment program to conduct a retrospective study of pathological gamblers treated in a publicly supported residential program. The Johns Hopkins Center for Pathological Gambling treated approximately 100 pathological gamblers nationwide from 1979 to 1983 in its residential component. Twenty-six gamblers were assessed regarding their recovery status with specific focus on family, marital status and occupational satisfaction. It was hypothesized that the more satisfied patients would display lower frequencies of gambling and that gambling frequency was dependent upon the level of satisfaction. Thirty-one percent of these gamblers reported abstinence. Regression analyses indicated that the recovery of pathological gamblers as measured by frequency of gambling is significantly influenced by job satisfaction, and moderately influenced by marital status and family life satisfaction. Consequently, these factors should be addressed by mental health professionals in research and treatment of pathological gambling.     

MORE EVIDENCE ON THE MONEY-OUTPUT RELATIONSHIP

Recent studies have found that money loses its explanatory power over output if the 2980s are included in the sample. Interest rates, not money, appear to predict output. Using annual data for 1915–1993 and quarterly data for 1960–1993, we demonstrate that the supposed breakdown in the money-output relationship stems from the type of stationary assumption imposed on the data. Assuming difference-stationary produces results found in the literature. Assuming trend-stationary produces results indicating that money and output remain statistically related. Moreover, the change in the stationarity assumption greatly affects the quantitative importance cf interest rates in explaining output.     

Dimensions of Community Attachment and Their Relationship to Well‐Being in the Amenity‐Rich Rural West*

Abstract This research had two primary objectives: 1) to broaden the sociological construct of community attachment to incorporate both social and natural environment dimensions of attachment, and 2) to examine how variations in attachment relate to two dimensions of well‐being in natural amenity‐rich rural communities. The specific dimensions of well‐being measured are two important factors identified in previous research—collective action and perceptions of open communication. Factor analysis of fourteen measures of attachment indicated social attachment and attachment to the natural environment are distinct dimensions of the broader concept of community attachment. Participation in collective action and perceptions of open communication within a respondent's community explained only a small portion of the variance in levels of both social and natural environment attachment. Religious affiliation and length of residence were strongly associated with social attachment, supporting findings from previous empirical work. However, length of residence and religious affiliation were not statistically associated with levels of attachment to the natural environment, further reinforcing the distinction between the two dimensions of attachment.