Reducing bias in ecological studies: an evaluation of different methodologies

Summary.  Statistical methods of ecological analysis that attempt to reduce ecological bias are empirically evaluated to determine in which circumstances each method might be practicable. The method that is most successful at reducing ecological bias is stratified ecological regression. It allows individual level covariate information to be incorporated into a stratified ecological analysis, as well as the combination of disease and risk factor information from two separate data sources, e.g. outcomes from a cancer registry and risk factor information from the census sample of anonymized records data set. The aggregated individual level model compares favourably with this model but has convergence problems. In addition, it is shown that the large areas that are covered by local authority districts seem to reduce between-area variability and may therefore not be as informative as conducting a ward level analysis. This has policy implications because access to ward level data is restricted.     

The Value of Hate in the Countertransference

The author looks at the experience of hate in the countertransference and develops the idea that hate can be of particular value to the treatment process. He begins with an exploration of the therapist’s needs and discusses how unmet self-regulatory needs sometimes lead to feelings of frustration and rage toward the patient. The author explores these ideas as they are connected to Winnicott’s (1949, 1971) work on aggression and Jessica Benjamin’s (1992) work on recognition and destruction. The author espouses a difference between rage and hate, where rage is seen as indicative of unmet needs, and hate is seen as a force that can emancipate the therapist from the encumbrance of unmet needs. Case examples are provided to show the utility of this perspective. Laurence Green, M.S.W., Psy.D. is on the faculty and a member of the Institute of Contemporary Psychoanalysis. He is Co-chair of the History Committee of the International Federation of Psychoanalytic Education. He is in private practice in Los Angeles, California working with children, adolescents and adults.     

Attitudes Toward Control in Uncontrollable Situations: The Multidimensional Impact of Health Locus of Control on the Well-Being of Mothers of Children with Disabilities

This study examines the additive and interactive impact of four dimensions of health locus of control (internality, desire for information, powerful others, and chance) on two measures of well-being (perceived social support and subjective caregiver burden) among 81 mothers of children with disabilities. Narrative accounts and interactive interviews with such mothers are used to interpret and contextualize the quantitative results. Findings clearly support the multidimensionality of health locus of control. Among these mothers, internality is uncorrelated with both of the dimensions of externality often treated as its polar opposite. It is also uncorrelated with desire for information—often used as its indicator. Complex individual, additive and interactive affects on the measures of well-being were found for dimensions of health locus of control. Desire for information is negatively correlated with subjective burden. Mothers who do not want information about health feel greater burden of care. Internality and chance interact in their relationships to both aspects of well-being. Mothers who believe that both their own actions and chance affect health outcomes report the highest levels of social support. Where internality is low, the belief in chance increases perceived social support. Conversely, where chance is low, internality increases support. For highly internal mothers, belief in chance serves to reduce perceived subjective burden. Mothers high in chance but low in internality, on the other hand, feel more burdened by their parenting task. Implications for research and policy are discussed.     

Boundary detection through dynamic polygons

A method for the Bayesian restoration of noisy binary images portraying an object with constant grey level on a background is presented. The restoration, performed by fitting a polygon with any number of sides to the object's outline, is driven by a new probabilistic model for the generation of polygons in a compact subset of R² , which is used as a prior distribution for the polygon. Some measurability issues raised by the correct specification of the model are addressed. The simulation from the prior and the calculation of the a posteriori mean of grey levels are carried out through reversible jump Markov chain Monte Carlo computation, whose implementation and convergence properties are also discussed. One example of restoration of a synthetic image is presented and compared with existing pixel-based methods.     

SOCIAL ISOLATION OF THE URBAN POOR:

We focus on the effects of race, class, and neighborhood on social isolation. Using data from households in Atlanta, Georgia, we compare poor and nonpoor African Americans to nonpoor whites on two types of social ties and the social resources inherent in those ties. We find that poverty has an important influence on the social resources available to African Americans in and outside of their household. Poor blacks are less likely than other blacks and nonpoor whites to live with another adult, to have even one person outside the household with whom they discuss important matters (a discussion partner), or to have a college-educated person in their discussion network. Higher neighborhood poverty reduces the size of the discussion network for whites and blacks and affects the probabilities of having any kind of social contacts. Important for the social isolation thesis is our finding that among African Americans, living in a very poor neighborhood increases social isolation and reduces access to social resources via one's network of close ties.     

Civil Society,Economic Distress,and Social Tolerance

The idea that civil society is declining has been much discussed recently, for example, by Fukuyama (Trust: The Social Virtues and the Creation of Prosperity. New York: Free Press, 1995) and Putnam (The American Prospect 13 (Spring):35–42, 19937semi; Nobel Symposium, Uppsala, Sweden, August 27–30, 1994; PS: Political Science and Politics, December: 664–683:. At the same time, another stream of research suggests that racial tolerance is increasing through time (Quillian, American Journal of Sociology 101 (3):816–860, 1996; Firebaugh and Davis, American Journal of Sociology 94 (2):251–272. This paper combines ideas from several intellectual traditions to suggest that economic conditions may affect civil society, which in turn may influence social tolerance. These ideas are explored, using multivariate analysis of pooled General Social Survey data from 1972 to 1994. The results suggest that the fabric of civil society and economic conditions may contribute somewhat to espoused tolerance. Greater economic security, together with the attitudes fostered by a vibrant civil society including greater trust and less anomia, appear to increase espoused social tolerance. Period effects and the effects of other demographic factors remain strongly related to expressions of tolerance. While alternative interpretations may be offered depending on whether the interaction terms, Year × Education or Cohort × Education, are included in the analysis, the high correlation between the two suggests that these interpretations cannot be seen as oppositional. Finally and most important, the analysis reveals many parallels between espoused racial and homosexual tolerance, suggesting a more generalizable model of social tolerance.     

Coping with Time‐Varying Demand When Setting Staffing Requirements for a Service System

We review queueing‐theory methods for setting staffing requirements in service systems where customer demand varies in a predictable pattern over the day. Analyzing these systems is not straightforward, because standard queueing theory focuses on the long‐run steady‐state behavior of stationary models. We show how to adapt stationary queueing models for use in nonstationary environments so that time‐dependent performance is captured and staffing requirements can be set. Relatively little modification of straightforward stationary analysis applies in systems where service times are short and the targeted quality of service is high. When service times are moderate and the targeted quality of service is still high, time‐lag refinements can improve traditional stationary independent period‐by‐period and peak‐hour approximations. Time‐varying infinite‐server models help develop refinements, because closed‐form expressions exist for their time‐dependent behavior. More difficult cases with very long service times and other complicated features, such as end‐of‐day effects, can often be treated by a modified‐offered‐load approximation, which is based on an associated infinite‐server model. Numerical algorithms and deterministic fluid models are useful when the system is overloaded for an extensive period of time. Our discussion focuses on telephone call centers, but applications to police patrol, banking, and hospital emergency rooms are also mentioned.     

What (a) Difference a Degree Makes: The Evaluation of the New Social Work Degree in England

After many years of debate in the UK about the need for a degree-levelqualification in social work, the arguments for a minimum degree-levelqualification were accepted. The requirements for the degreein England were developed drawing on work from a number of sources,including a benchmark statement for undergraduate degrees insocial work and focus groups with stakeholders. The new degreein England, launched in 2003, involves one extra year’sstudy; improvements in the qualifying standard for social work;and specific curriculum and entrance requirements. At the timeof launching the degree, the government department responsiblefor funding (Department of Health) commissioned a three-yearevaluation of the implementation of the new degree to establishwhether the new qualifying level leads to improvements in thequalified workforce. The aim of the evaluation is to describethe experiences of those undertaking the degree, collect theviews of the various stakeholders about the effectiveness ofthe degree and measure the impact of a degree-level qualificationon those entering the workforce. This article, written by theteam undertaking the evaluation of the England degree, exploresthe reasons for the methodological approach adopted and theissues that have arisen in setting up the research.     

Who is the victim anyway? The effects of bystander victim impact statements on mock juror sentencing decisions

Victim impact evidence was varied in a brief trial scenario given to 120 participants eligible for jury service. The scenario concerned the murder trial of a disgruntled employee accused of bombing his former workplace. Participants read either no victim impact evidence or one of three victim impact statements. For the victim impact statements, we varied the identity of the witness. The victim impact statement was given by either the wife of the victim, a coworker of the victim, or a firefighter called to the crime scene. Results revealed that only the victim impact evidence given by the coworker lead to harsher sentencing judgments. However, participants rated the suffering of the victim's wife as most severe, indicating that perceptions of suffering may not predict sentencing judgments in a straightforward manner. Implications for these findings for legal decisions such as Payne v. Tennessee (1991) are discussed.