Imputation of MĀori Descent for Electoral Calculations in New Zealand

The New Zealand Government Statistician decided that, for electoral purposes, Statistics New Zealand should impute Māori–descent status for individuals not responding Yes or No to theMāori–descent question in the 1996 Census of Population and Dwellings. Imputation provides a sounder basis for calculating electoral populations than the approach used in 1994, when all who had not answered clearly Yes or No in the 1991 Census were effectively allocated to non–Māori descent. For the purposes of imputation, the key variables related to the Māori–descent variable were identified using a statistical technique called CHAID (Chisquared Automatic Interaction Detector). Subgroups were created by cross–classification across five variables—island, iwi, Māori ethnic group, Māori–descent composition of the rest of the household, and age group. Within each subgroup, the proportion who responded Yes or No for Māori descent was used to allocate the remainder to Yes or No. The imputation increased the proportion allocated to Māori descent from 16.0% to 17.4% of the total population. However, the proportion of the population imputed to Māori descent was smaller than the proportion who specified Māori descent originally.     

Modelling non-response in the National Child Development Study

Summary.  There is widespread concern that the cumulative effects of the non-response that is bound to affect any long-running longitudinal study will lead to mistaken inferences about change. We focus on the National Child Development Study and show how non-response has accumulated over time. We distinguish between attrition and wave non-response and show how these two kinds of non-response can be related to a set of explanatory variables. We model the discrete time hazard of non-response and also fit a set of multinomial logistic regressions to the probabilities of different kinds of non-response at a particular sweep. We find that the best predictors of non-response at any sweep are generally variables that are measured at the previous sweep but, although non-response is systematic, much of the variation in it remains unexplained by our models. We consider the implications of our results for both design and analysis.     

Using number of failed contact attempts to adjust for non-ignorable non-response

Summary.  We present a general method of adjustment for non-ignorable non-response in studies where one or more further attempts are made to contact initial non-responders. A logistic regression model relates the probability of response at each contact attempt to covariates and outcomes of interest. We assume that the effect of these covariates and outcomes on the probability of response is the same at all contact attempts. Knowledge of the number of contact attempts enables estimation of the model by using only information from the respondents and the number of non-responders. Three approaches for fitting the response models and estimating parameters of substantive interest and their standard errors are compared: a modified conditional likelihood method in which the fitted inverse probabilities of response are used in weighted analyses for the outcomes of interest, an EM procedure with the Louis formula and a Bayesian approach using Markov chain Monte Carlo methods. We further propose the creation of several sets of weights to incorporate uncertainty in the probability weights in subsequent analyses. Our methods are applied as a sensitivity analysis to a postal survey of symptoms in Persian Gulf War veterans and other servicemen.     

Inequality and happiness: Insights from Latin America

Inequality is a contentious topic in economics, and its effect on individual welfare remains an open question. We address this question from the perspective of a novel approach in economics – the study of happiness. In this discussion, we draw from our research on the topic, which is based on new empirical evidence from Latin America. We find several differences from studies conducted in the United States and Europe, especially regarding the role of perceptions of mobility and status. We find that inequality has negative effects on happiness in Latin America, where it seems to be a signal of persistent unfairness. Our research also examines the effects of several variables, including wealth, status, and reference group size, on the link between inequality and happiness, with the presumption that these variables can help us identify the channels through which inequality operates as a signaling mechanism. This article is based on a longer research paper [14], which is under review for publication.     

Reciprocal relations: The recognition and co-construction of caring with Alzheimer's disease

The literature on caring for people with Alzheimer's disease predominantly underscores the role of the care giver, largely ignoring the care receiver. Care is viewed as a service to mind someone rather than a compassionate act of shared relationship. Alzheimer's disease is portrayed as negatively influencing reciprocity in caring relationships, precipitating their gradual dissolution. We report findings from an ethnographic study that followed nearly 100 dyads of people with Alzheimer's disease and those who cared for them for more than 1 year. ‘Give and take’ in a relationship appears richer and more ambiguous than previously supposed; despite one party's cognitive losses, shared meanings can still be experienced by both people. In an in-depth examination of the narratives of co-carers, gathered during home visits, relationships were seen to be dynamic co-constructions built upon everyday events, interactions, environments and disease progression. Importantly, we show that people with Alzheimer's disease express insight and awareness despite decline.     

The Impact of Parental Attachment on Adolescent Perception of the School Environment and School Connectedness

The extent to which students feel accepted, valued, respected and included in the school has recently surfaced as one of the most important predictors of adolescent mental health (particularly depressive symptoms). The school environment is an established predictor of school connectedness, but we set out to examine whether parental attachment predicts both adolescents' perception of the school environment and school connectedness. A study of 171 high school students from years 8 to 12 showed that parent attachment strongly predicted both. We also confirmed that the relationship between parent attachment and school connectedness is not a direct one but that parent attachment influences individual differences in the way adolescents perceive the school environment, which in turn influences school connectedness. This finding shows how multiple systems might be interlinked in influencing wellbeing in adolescents, and confirms the importance of intervening at the double level of both the family and the school system.     

Receptivity to Change in a General Medical Practice

In this paper, Pettigrew, Ferlie and McKee's concepts of receptive and non receptive contexts for change are examined using data from an evaluation study of a general medical practice implementing a Personal Medical Services Pilot between 1998 and 2001. Four questions concerning the applicability, associatedness, temporality and dynamism of the eight factors Pettigrew et al. identified as constitutive of receptivity are used in the analysis, which extends the original model through an operationalization yielding 21 ‘focal questions’. Although the process of change at the fieldwork site was stalled, the receptivity model is shown to have been a useful tool in identifying the factors blocking change. The analysis shows that when the eight factors are conceptualized in terms of the strength, direction and continuity of their influence, the receptivity metaphor provides a distinctive tool for the analysis of change. The paper concludes that further refinement of the model could provide a basis for the derivation of change management typologies.