Stephen Haslett

SF-36 health survey reliability, validity and norms for New Zealand.

Objective: To assess the acceptability, reliability and validity of the SF‐36 health survey in the New Zealand population and provide key population norms.

A challenge to the cross-cultural validity of the SF-36 health survey: factor structure in Maori, Pacific and New Zealand European ethnic groups.

This paper reports on a principal component factor analysis of the SF-36 health status questionnaire in the three major ethnic groups in New Zealand (New Zealand Europeans, Maori and Pacific). The SF-36 is hypothesised to have a two-dimensional structure with distinct (weakly correlated) mental and physical health components, and support for this structural model has generally been found cross-nationally. However, in Maori and Pacific models of health mental and physical dimensions are not generally seen as separable, or independently functioning. This raises the possibility that the questionnaires hypothesised structural model would not be supported among Maori and Pacific ethnic groups. This study evaluated that possibility. The results of the analysis showed a similar factor structure among New Zealand Europeans, and younger Maori (<45 years) to that reported by Ware et al. for Western European countries. Among Pacific people and older Maori (45 years and over), however, the factor structure did not clearly differentiate physical and mental health components. Implications are discussed both specific to the SF-36 (and in particular the use of principal component summary scores), and more generally for the cross-cultural validity of self-reported health status measures.

Updating linear models with dependent errors to include additional data and/or parameters

Abstract Although updating equations for general linear models exist, they have been restricted to formulations with uncorrelated error structure, or to models involving fixed parameters of fixed length with correlated error structure. In this paper updating equations are developed for general linear models with correlated errors when the parameter vector is supplemented or evolves as further data are added. These equations form part of a general review in which updating formulae for minimum variance, linear, unbiased parameter estimates and their variance, applicable where a general linear model is altered by the inclusion of additional parameters and/or data, and where the variance matrices for the model error and the parameter vector are known and nonsingular, are outlined. When variance matrices are estimated rather than known, the equations derived give estimated generalized least squares estimators. The formulae are developed for both fixed and random parameter vectors. For models in which additional data and parameters are added simultaneously, it is shown that a generalization of the Kalman filter is possible that allows for arbitrary forms of model error autocorrelation, given known nonsingular error variance.

When and Why Talking Can Make Writing Harder

This chapter reports on a quasi-experimental study investigating the effects of talk on the writing of argument texts by second language students in a New Zealand high school. Results are compared on variables related to three conventional text quality measures (communicative quality, ideas and organisation, and grammatical accuracy and complexity), and also on variables specifically related to the rhetorical organisation of argument texts (frequency of claims, grounds and elaboration of grounds, as identified by Toulmin, Rieke, and Janki, 1984). The results of the study indicate that the opportunity to work with a peer before and during writing had limited and specific effects on the texts the students wrote. Working in a solitary way resulted in significantly higher mean scores for linguistic accuracy and complexity (a general text quality measure). Significant and positive effects for opportunity to talk with a peer were seen in the quantity of grounds-related propositions (an argument text measure), but only when students wrote texts that appeared to require more domain-specific knowledge to support the claims made in their argument texts. This chapter seeks to explain these effects with reference to the notion of attentional resources and aspects of noticing. A number of task and learner factors are explored to explain the way in which attentional resources are divested and the degree to which learners attend to or notice aspects of the language and content experienced in the interaction.

Associations between dietary patterns, socio-demographic factors and anthropometric measurements in adult New Zealanders: an analysis of data from the 2008/09 New Zealand Adult Nutrition Survey

PurposeTo investigate associations between dietary patterns, socio-demographic factors and anthropometric measurements in adult New Zealanders.MethodsDietary patterns were identified using factor analysis in adults 15 years plus (n = 4657) using 24-h diet recall data from the 2008/09 New Zealand Adult Nutrition Survey. Multivariate regression was used to investigate associations between dietary patterns and age, gender and ethnicity. After controlling for demographic factors, associations between dietary patterns and food insecurity, deprivation, education, and smoking were investigated. Associations between dietary patterns and body mass index and waist circumference were examined adjusting for demographic factors, smoking and energy intake.ResultsTwo dietary patterns were identified. ‘Healthy’ was characterised by breakfast cereal, low fat milk, soy and rice milk, soup and stock, yoghurt, bananas, apples, other fruit and tea, and low intakes of pies and pastries, potato chips, white bread, takeaway foods, soft drinks, beer and wine. ‘Traditional’ was characterised by beef, starchy vegetables, green vegetables, carrots, tomatoes, savoury sauces, regular milk, cream, sugar, tea and coffee, and was low in takeaway foods. The ‘healthy’ pattern was positively associated with age, female gender, New Zealand European or other ethnicity, and a secondary school qualification, and inversely associated with smoking, food insecurity, area deprivation, BMI and waist circumference. The ‘traditional’ pattern was positively associated with age, male gender, smoking, food insecurity and inversely associated with a secondary school qualification.ConclusionsA ‘Healthy’ dietary pattern was associated with higher socio-economic status and reduced adiposity, while the ‘traditional’ pattern was associated with lower socio-economic status.

The Rise of the Left and Working-Class Voting Behavior in New Zealand: New Methods

The extent to which mainstream left-wing parties attracted working-class votes during the first half of the twentieth century is exceptionally difficult to establish and explain. All of the various methods applied to the subject, including ecological regression of aggregate data, have had their problems, especially the ecological fallacy. A novel solution to these problems, in the context of New Zealand, takes occupational and party voting data at street level as its observations for ten towns from 1911 to 1951, and correlates the data treating each town for each year as a case. The working-class component in the total vote for the Labour Party varied surprisingly by town and followed unexpected trends.

Compositional data in neuroscience: If you’ve got it, log it!

BACKGROUND Compositional data sum to a constant value, for example, 100%. In neuroscience, such data are common, for example, when estimating the percentage of time spent for a behavioural response in a limited choice situation or a neurochemical within brain tissue. Compositional data have a distinct structure which complicates analysis and makes inappropriate standard statistical analyses such as general linear model analyses and principal components or factor analysis (whether Q-mode or R-mode), as a result of the correlation of the components, the dependence of the pairwise covariance on which other components are included in the analysis, and the bounded nature of the data. NEW METHOD This problem has been recognised in disciplines such as geology and zoology for decades, where log ratio methods have been successfully applied. The isometric log ratio (ilr) method has some particular advantages. COMPARISON WITH EXISTING METHOD Classical statistical methods such as t-tests, ANOVAs, and multivariate analyses are invalid when applied to compositional data. CONCLUSIONS The compositional data analysis methods developed by statisticians and used by geologists and zoologists should be considered for compositional data analysis in neuroscience.

Illustration of regression towards the means

This article, presents a procedure for generating a sequence of data sets which will yield exactly the same fitted simple linear regression equation y = a + bx. Unless rescaled, the generated data sets will have progressively smaller variability for the two variables, and the associated response and covariate will ‘regress’ towards their unconditional sample means.

Cleavage within the Working Class?: The Working-class Vote for the Labour Party in New Zealand, 1911-51

A convention in New Zealand historiography is that the electoral fortunes of the Labour Party in the first half of the twentieth century were tightly constrained by a particular structure of cleavage. Although by 1919 or 1922 Labour had won the support of the vast majority of urban working-class voters in the leading towns, it could not win office with their support alone. This was demonstrated in the 1930s when it gained electoral domination by winning over small farmers and the urban middle classes; and after the late 1930s when the support of non-manuals ebbed away and it lost office in 1949. This article tests the convention. It takes the ten largest provincial towns, determines the class composition of their streets in 11 general elections, estimates the distribution of the votes for each party in each street in each town for every year, and correlates the estimated percentages of Labour vote with percentage of working class. The results are the inverse of the trends claimed by the historiography. The article then examines the social geography of the towns and finds systematic evidence of an unexpected cleavage inside the working class: skilled workers had a much weaker tendency to vote Labour than the unskilled and no tendency at all to reside in the same residential areas as the unskilled.

A statistical test of single firm market power

A market structure-performance model is used to search for a structural break in the leading firm market share-profitability relationship. The model, incorporating both market power and efficiency variables to account for variations in price-cost margins, is tested using cross-section 1978/79 Census data for New Zealand manufacturing industries. A novel computational approach using backward and forward cusum and cusum of squared tests based on recursive residuals is used. A statistically significant breakpoint at a top firm market share of 16-17 per cent is found, with the model being much stronger for the sub-set of data above that point.