The effects of clustered data on standard error estimates in covariance structure analysis : a field data application

Purpose – The purpose of this article is to present an empirical analysis of complex sample data with regard to the biasing effect of non-independence of observations on standard error parameter estimates. Using field data structured in the form of repeated measurements it is to be shown, in a two-factor confirmatory factor analysis model, how the bias in SE can be derived when the non-independence is ignored.

Design/methodology/approach – Three estimation procedures are compared: normal asymptotic theory (maximum likelihood); non-parametric standard error estimation (naïve bootstrap); and sandwich (robust covariance matrix) estimation (pseudo-maximum likelihood).

Findings – The study reveals that, when using either normal asymptotic theory or non-parametric standard error estimation, the SE bias produced by the non-independence of observations can be noteworthy.

Research limitations/implications – Considering the methodological constraints in employing field data, the three analyses examined must be interpreted independently and as a result taxonomic generalisations are limited. However, the study still provides “case study” evidence suggesting the existence of the relationship between non-independence of observations and standard error bias estimates.

Originality/value – Given the increasing popularity of structural equation models in the social sciences and in particular in the marketing discipline, the paper provides a theoretical and practical insight into how to treat repeated measures and clustered data in general, adding to previous methodological research. Some conclusions and suggestions for researchers who make use of partial least squares modelling are also drawn.

The reliability of performance of healthy adults on three visual perception tests

Problems with visual perceptual skills have been shown to have a negative impact on the daily living skills of individuals and are, therefore, commonly assessed by occupational therapists. The purpose of this study was to examine two types of reliability (internal consistency and test-retest reliability) of three newly revised or developed adult visual perception tests. The participants were 50 healthy adults, aged 18 to 55 years, from Melbourne, Victoria, Australia. The participants completed the Developmental Test of Visual Perception - Adolescent and Adult (DTVP-A), the Motor-Free Visual Perception Test - Third Edition (MVPT-3) and the Test of Visual Perceptual Skills (non-motor) - Third Edition (TVPS-3). Internal consistency was examined using Cronbach's alpha calculations and test-retest reliability was analysed using Spearman rho non-parametric correlation coefficients.

The results indicated that the DTVP-A, the MVPT-3 and the TVPS-3 had total scale internal consistency correlation scores of 0.60 or higher (0.60, 0.69 and 0.63 respectively). The majority of the subscales of each test had lower correlation coefficients than the overall scores (ranging from 0.22 to 0.49). For the DTVP-A, MVPT-3 and TVPS-3 total scale scores, the test-retest reliability correlation coefficients were statistically significant (rho = 0.46, p<0.05; rho = 0.62, p<0.01; and rho = 0.59, p<0.01, respectively). Overall, the three visual perceptual tests exhibited low to moderate levels of internal consistency and test-retest reliability.

Productivity growth and efficiency change in Malaysian manufacturing : recent evidence from disaggregated data

This paper estimates productivity growth in Malaysian manufacturing over the period 1983-1999. Malmquist productivity Indices (MPIs) have been computed using non parametric Data Envelopment Analysis (DEA) type linear programming, which show productivity growth sourced from efficiency change and growth in technology. Unlike previous studies, this study identifies the Malaysian manufacturing industries at the five digit breakdown of Malaysian Standard Industrial Classification (MSIC) thereby revealing more industry specific efficiency and technical growth patterns. Results indicate that two third of the industries (76 out of total 114 categories) experienced average annual
productivity improvement ranging from 0.1% to 7.8% over the sampled period. Average annual technical progress was recorded by 95 industry categories while technical efficiency improvement was achieved by 53 industries. Overall yearly average indicated relatively low productivity growth from the mid 1990’s onwards caused by either efficiency decline or technical regress. Summary results for industries reveal that some of the high rates of productivity growth have been recorded in glass and glass products (7.3%), Petroleum and coal (7.2%), industrial chemicals (4.9%) contributed from both efficiency improvement and technical progress ranging from 0.8% to 5.4% and from 1.7% to 4.1%, respectively. These results are expected to have some implications for ongoing and future strategic policy reform in Malaysian manufacturing generating a more sustainable growth for specific industry categories.

Conducting event studies with Asia-Pacific security market data

We investigate the effectiveness of several well-known parametric and non-parametric event study test statistics with security price data from the major Asia-Pacific security markets. Extensive Monte Carlo simulation experiments with actual daily security returns data reveal that the parametric test statistics are prone to misspecification with Asia-Pacific returns data. Two non-parametric tests, a rank test [Corrado and Zivney (Corrado, C.J., Zivney, T.L., 1992, The specification and power of the sign test in event study hypothesis tests using daily stock returns, Journal of Financial and Quantitative Analysis 27(3), 465-478)] and a sign test [Cowan (Cowan, A.R., 1992, Non-parametric event study tests, Review of Quantitative Finance and Accounting 1(4), 343–358)] were the best performers overall with market model excess returns computed using an equal weight index.

An illustration of variable precision rough set theory : The gender classification of the European barn swallow (Hirundo rustica)

This paper introduces a new technique in the investigation of object classification and illustrates the potential use of this technique for the analysis of a range of biological data, using avian morphometric data as an example. The nascent variable precision rough sets (VPRS) model is introduced and compared with the decision tree method ID3 (through a ‘leave n out’ approach), using the same dataset of morphometric measures of European barn swallows (Hirundo rustica) and assessing the accuracy of gender classification based on these measures. The results demonstrate that the VPRS model, allied with the use of a modern method of discretization of data, is comparable with the more traditional non-parametric ID3 decision tree method. We show that, particularly in small samples, the VPRS model can improve classification and to a lesser extent prediction aspects over ID3. Furthermore, through the ‘leave n out’ approach, some indication can be produced of the relative importance of the different morphometric measures used in this problem. In this case we suggest that VPRS has advantages over ID3, as it intelligently uses more of the morphometric data available for the data classification, whilst placing less emphasis on variables with low reliability. In biological terms, the results suggest that the gender of swallows can be determined with reasonable accuracy from morphometric data and highlight the most important variables in this process. We suggest that both analysis techniques are potentially useful for the analysis of a range of different types of biological datasets, and that VPRS in particular has potential for application to a range of biological circumstances.

Aqueous/micellar peroxyoxalate chemiluminescent detection for ion chromatography

This thesis covers the development of the traditionally fluorescent bis(8-quinolinol-5-sulfonic acid) magnesium (II) fluorophore as a chemiluminescent emitter. A brief description of luminescence spectroscopy and its application to analytical chemistry lays the foundation to the discussion of the results obtained herein. This includes the synthesis and identification of two so called ‘water soluble’ aryl oxamides 2,2’-oxalyl-bis(trifluoromethanesulfonyl) imino] ethylene-bis(N- methylpyridinium) trifluoromethane sulfonate (PETQ) and 2,2’-oxalyl-bis(trifluoromethanesulfonyl) imino]ethylene-bis(N-pyridinium) chloride (PETH), previously developed for the US navy as a possible emergency light source, yet the synthetic methodology were incomplete. The inconsistencies of the synthetic methods for PETQ and PETH were overcome with yields satisfactory for their preliminary analytical evaluation. The evaluation of these aryl oxamides, including 4,4’-oxalyI- bis[(trifluoromethanesulfonyl) imino]ethylene-bis(l-methyM-benzylpiperidinium) trifluoromethanesulfonate (BPTQ), 4,4’-oxalyl-bis [(trifluoromethylsulfonyl)imino] ethylene-bis(N-methylmorpholinium)trifluoromethanesulfonate (METQ) and the oxalate bis(2,4,6-trichlorophenyl) oxalate (TCPO) were performed with the peroxyoxalate chemiluminescent reaction using bis(8-quinolinol-5-sulfonic acid) magnesium (II) as the fluorophore. A univariate optimisation of this system resulted in 0,0082 mol 1-1 the detection limit of magnesium in the absence of cationic surfactants and 0.0041 mol 1-1 in their presence for the majority of these compounds. The oxamides were found to be insoluble in water with long ulrasonication periods required to dissolve the compound, with solvents such as acetonitrile preferred. The determination of other chemiluminescent metal-8HQS chelates to replace magnesium -8HQS in the peroxyoxalate were limited to Al (III), Cd (II), Ca (II), In (II) and Zn (II), unfortunately these metals all possessed poorer detection limits than those obtained using magnesium The base reaction conditions used for the flow injection system with chemiluminescent detection were transferred to an ion chromatographic configuration for the separation of magnesium from other cations on an exchange column. After a univariate and simplex optimisation of these conditions, the detection limit of magnesium was found to be 0.0411 mol 1-1 which was less than the limits that could be achieved with fluorescent detection, The further development of this reaction to incorporate the displacement of magnesium from Mg-EDTA by other metals that possessed a higher conditional stability constant than magnesium also proved to be problematic with interferences from not only EDTA but from the eluant (lactic acid) from the cation column. Using this system the detection limits of the displacing metals were found to be in the order of 10 mg 1-1 which was substantially less that what was observed when exactly the same configuration was used with fluorescent detection. The final component of the thesis entails the discussion of the background emission that results from the reaction of oxamides/oxalates with hydrogen peroxide. A detailed investigation into the reaction of TCPO and hydrogen peroxide in the presence of various additives, such as imidazole , heavy atoms and triethylamine illustrated the existence of a further intermediate in fee mechanism for this reaction. The species responsible for this emission was attributed to the degradation product 2,4,6-trichlorophenyi of TCPO, which was supported by the non-existent background present with the oxamides that do not contain this degradation product.

Correlates of screening mammography for Italian and Anglo-Australian women

The first aim of the research was to determine the applicability of certain variables from the Health Belief Model (HBM), the Theory of Reasoned Action (TRA), the risk dimensions from the Psychometric Paradigm, the Common-Sense Model of Illness Representations and the Locus of Control to Italian women’s beliefs and behaviours in relation to screening mammography. These models have predominantly been derived and evaluated with English-speaking persons. The study used quantitative and qualitative methods to enable explanation of research-driven and participant-driven issues. The second aim was to include Italian women in health behaviour research and to contrast the Italian sample with the Anglo-Australian sample to determine if differences exist in relation to their beliefs. In Australia many studies in health behaviour research do not include women whose first language is not English. The third aim was to evaluate the Anti-Cancer Council of Victoria’s (ACCV) Community Language Program (CLP) by: (a) identifying the strengths and weaknesses of the program as seen by the participants; and (b) assessing the impact of the program on women’s knowledge and beliefs about breast cancer, early detection of breast cancer, self-reported and intended breast screening behaviours. The CLP is an information service that uses women’s first language to convey information to women whose first language is not English. The CLP was designed to increase knowledge about breast and cervical cancer. The research used a pre-test-intervention-post-test design with 174 Italian-born and 138 Anglo-Australian women aged 40 years and over. Interviews for the Italian sample were conducted in Italian. The intervention was an information session that related to breast health and screening mammography. Demographic variables were collected in the Pre-Test only. Qualitative open-ended questions that related specifically to the information session were collected in the Post-Test phase of the study. Direct logistic regression was used with the participants’ beliefs and behaviours to identify the relevant variables for language (Italian speaking and English-speaking), attendance to an information session, mammography screening and breast self-examination (BSE) behaviour. Pre- and Post-Test comparisons were conducted using chi-square tests for the non-parametric data and paired sample t-tests for the parametric data. Differences were found between the Italian and Anglo-Australian women in relation to their beliefs about breast cancer screening. The Italian women were: (1) more likely to state that medical experts understood the causes of breast cancer; (2) more likely to feel that they had less control over their personal risk of getting breast cancer; (3) more likely to be upset and frightened by thinking about breast cancer; (4) less likely to perceive breast cancer as serious; (4) more likely to only do what their doctor told them to do; and (5) less likely to agree that there were times when a person has cancer and they don’t know it. A pattern emerged for the Italian and Anglo-Australian women from the logistic regression analyses. The Italian women were much more likely to comply with medical authority and advice. The Anglo-Australian women were more likely to feel that they had some control over their health. Specifically, the risk variable ‘dread’ was more applicable to the Italian women’s behaviour and internal locus of control variable was more relevant to the Anglo-Australian women. The qualitative responses also differed for the two samples. The Italian women’s comments were more general, less specific, and more limited than that of the Anglo-Australian women. The Italian women talked about learning how to do BSE whereas the Anglo-Australian women said that attending the session had reminded them to do BSE more regularly. The key findings and contributions of the present research were numerous. The focus on one cultural group ensured comprehensive analyses, as did the inclusion of an adequate sample size to enable the use of multivariate statistics. Separating the Italian and Anglo-Australian samples in the analyses provided theoretical implications that would have been overlooked if the two groups were combined. The use of both qualitative and quantitative data capitalised on the strengths of both techniques. The inclusion of an Anglo-Australian group highlighted key theoretical findings, differences between the two groups and unique contributions made by both samples during the collection of the qualitative data. The use of a pre-test-intervention-post-test design emphasised the reticence of the Italian sample to participate and talk about breast cancer and confirmed and validated the consistency of the responses across the two interviews for both samples. The inclusion of non-cued responses allowed the researcher to identify the key salient issues relevant to the two groups. The limitations of the present research were the lack of many women who were not screening and reliance on self-report responses, although few differences were observed between the Pre- and Post-Test comparisons. The theoretical contribution of the HBM and the TRA variables was minimal in relation to screening mammography or attendance at the CLP. The applicability of these health behaviour theories may be less relevant for women today as they clearly knew the benefits of and the seriousness of breast cancer screening. The present research identified the applicability of the risk variables to the Italian women and the relevance of the locus of control variables to the Anglo-Australian women. Thus, clear cultural differences occurred between the two groups. The inclusion of the illness representations was advantageous as the responses highlighted ideas and personal theories salient to the women not identified by the HBM. The use of the illness representations and the qualitative responses further confirmed the relevance of the risk variables to the Italian women and the locus of control variables to the Anglo-Australian women. Attendance at the CLP did not influence the women to attend for mammography screening. Behavioural changes did not occur between the Pre- and Post-Test interviews. Small incremental changes as defined by the TTM and the stages of change may have occurred. Key practical implications for the CLP were identified. Improving the recruitment methods to gain a higher proportion of women who do not screen is imperative for the CLP promoters. The majority of the Italian and Anglo-Australian women who attended the information sessions were women who screen. The fact that Italian women do not like talking or thinking about cancer presents a challenge to promoters of the CLP. The key theoretical finding that Italian women dread breast cancer but comply with their doctor provides clear strategies to improve attendance at mammography screening. In addition, the inclusion of lay health advisors may be one way of increasing attendance to the CLP by including Italian women already attending screening and likely to have attended a CLP session. The present research identified the key finding that improving Anglo-Australian attendance at an information session is related to debunking the myth surrounding familial risk of breast cancer and encouraging the Anglo-Australian women to take more control of their health. Improving attendance for Italian women is related to reducing the fear and dread of breast cancer and building on the compliance pattern with medical authority. Therefore, providing an information session in the target language is insufficient to attract non-screeners to the session and then to screen for breast cancer. Suggestions for future research in relation to screening mammography were to include variables from more than one theory or model, namely the risk, locus of control and illness representations. The inclusion of non-cued responses to identify salient beliefs is advantageous. In addition, it is imperative to describe the profile of the cultural sample in detail, include detailed descriptions of the translation process and be aware of the tendency of Italian women to acquiesce with medical authority.

Effects of invertebrate predators and a pesticide on temporary pond microcosms used for aquatic toxicity testing

The effects of increased trophic complexity, through the addition of predatory notonectids (Anisops deanei), on temporary pond microcosms used for aquatic toxicity testing were studied. Replicate microcosms were established using sediment from a dried temporary pond, and treated with one of four concentrations of the organochlorine pesticide endosulfan (0, 1, 10 or 50 μg/L), in the presence or absence of six A. deanei. The tanks were sampled regularly for nine weeks following the addition of the predators and the entire contents of each tank counted after 12 weeks. Analysis using non-metric multidimensional scaling (MDS) and non-parametric MANOVA showed that both Anisops and endosulfan at concentrations >10 μg/L significantly altered community structure. However, an interaction between the effects of Anisops and the effects of endosulfan was not detected. The addition of Anisops did not increase the variability of response and thus did not reduce the sensitivity of the test method.

Pixel-based correspondence for facial images

A pixel-based correspondence method is presented for representation of facial images. The proposed method consists of two modules: face-image matching and face-image morphing. In the face-image matching module, the correspondence between two images are calculated for all pixel locations. A novel area-based matching method is proposed that makes use of the concept of the fractal dimension, and develops a non-parametric local transform as a basis for establishing correspondence between two face images. In the face-image morphing module, a mapping is performed for deformation of the source face image onto the target face image. This is done to map the pixels in the source face image to the location of their corresponding pixels in the target image.

Australian teaching and learning centres through the eyes of their directors : characteristics, capacities and constraints

This paper is based on research to identify common factors that contribute to the effective strategic leadership of teaching and learning centres. The second of three phases of data collection involved a survey of Directors of Australian teaching and learning centres. The data collected were quantitatively analysed using a range of descriptive, parametric and non-parametric techniques. Based on a response rate of 81.6 percent, we present a contemporary, comprehensive and representative quantitative snapshot of Australian teaching and learning centres, as seen through the eyes of their Directors. The time since last restructure, incumbency of the current Director and total Directorship experience of the current Centre Director all have mean values of ‘sometime in the previous one to three years’. Most Centres would consider their work in the areas of ‘recognition and reward’ and ‘professional development of staff’ as high impact functions, and they would be pleased with their efforts in the former area, and wish to perform better on the latter. The principal constraint identified by Centres was ‘lack of staff time’, both in the Faculties and in the Centre, to engage in teaching and learning improvement activities. Overall, Centres feel well included in relevant university committees and other activities.

Sweet talking to ethnic customers : a field experiment on the links between entrepreneurial business planning, cultural sensitivity and sales growth

The purpose of this study was to test whether calculated inclusion of cultural sensitivity in a selected entrepreneurial business planning (EBP) process could increase sales growth in a test market and to explore the implications of a positive answer for the theory and practice of entrepreneurial business planning. Execution of a pretest-posttest control group experimental design measured and compared the implemented effectiveness of a planned entrepreneurial initiative based on cultural sensitivity. Though small in scale and limited in focus, the initiative qualified as an example of entrepreneurial business planning (EBP) and could be used to apply, test and extend aspects of the developing theory in this field of entrepreneurship research. Since the initiative was planned to overcome a culturally-defined impediment to business growth, it also offered opportunity to explore the specific importance of cultural variables in the context of EBP.

A planned sales-promotion was offered to a control group (receiving information in English) and a treatment group (who received the information in the language of ethnic origin). The sixty subjects had been chosen at random from a population of route-trade retailers of defined ethnic origins (Greek, Lebanese and Chinese) and randomly assigned to control and treatment groups. Monthly sales averages of the promoted product were measured before and after treatment. A Chi Square test was used to evaluate the relative proportion of the control and treatment groups who accepted the promotional offer. A two sample t-test procedure and complementary non-parametric Mann-Whitney test were performed to compare the mean sales-performance change of the two groups. Analysis showed that there was a significant increase in mean sales when the planned entrepreneurial initiative was communicated in the relevant language of origin.

The experimental results have specific practical relevance to revitalising the deteriorating route-trade segment of the Australian confectionery market through increasing the sales growth of wholesalers who are prepared to act entrepreneurially and include cultural sensitivity as an element in planning and implementation. By introducing cultural sensitivity as a necessary extension of a plan’s communications role, the results also have general theoretical implications for the developing paradigm of entrepreneurial business planning.

Application of sequential nonparametric confidence bands in finance

In a nonparametric setting, the functional form of the relationship between the response variable and the associated predictor variables is assumed to be unknown when data is fitted to the model. Non-parametric regression models can be used for the same types of applications such as estimation, prediction, calibration, and optimization that traditional regression models are used for. The main aim of nonparametric regression is to highlight an important structure in the data without any assumptions about the shape of an underlying regression function. Hence the nonparametric approach allows the data to speak for itself. Applications of sequential procedures to a nonparametric regression model at a given point are considered.

The primary goal of sequential analysis is to achieve a given accuracy by using the smallest possible sample sizes. These sequential procedures allow an experimenter to make decisions based on the smallest number of observations without compromising accuracy. In the nonparametric regression model with a random design based on independent and identically distributed pairs of observations (X ,Y ), where the regression function m(x) is given bym(x) = E(Y X = x), estimation of the Nadaraya-Watson kernel estimator (m (x)) NW and local linear kernel estimator (m (x)) LL for the curve m(x) is considered. In order to obtain asymptotic confidence intervals form(x), two stage sequential procedure is used under which some asymptotic properties of Nadaraya-Watson and local linear estimators have been obtained.

The proposed methodology is first tested with the help of simulated data from linear and nonlinear functions. Encouraged by the preliminary findings from simulation results, the proposed method is applied to estimate the nonparametric regression curve of CAPM.

Sequential fixed-width confidence bands for kernel regression estimation

We consider a random design model based on independent and identically distributed (iid) pairs of observations (Xi, Yi), where the regression function m(x) is given by m(x) = E(Yi|Xi = x) with one independent variable. In a nonparametric setting the aim is to produce a reasonable approximation to the unknown function m(x) when we have no precise information about the form of the true density, f(x) of X. We describe an estimation procedure of non-parametric regression model at a given point by some appropriately constructed fixed-width (2d) confidence interval with the confidence coefficient of at least 1−. Here, d(> 0) and 2 (0, 1) are two preassigned values. Fixed-width confidence intervals are developed using both Nadaraya-Watson and local linear kernel estimators of nonparametric regression with data-driven bandwidths.

The sample size was optimized using the purely and two-stage sequential procedure together with asymptotic properties of the Nadaraya-Watson and local linear estimators. A large scale simulation study was performed to compare their coverage accuracy. The numerical results indicate that the confidence bands based on the local linear estimator have the best performance than those constructed by using Nadaraya-Watson estimator. However both estimators are shown to have asymptotically correct coverage properties.

Sample size determination for kernel regression estimation using sequential fixed-width confidence bands

We consider a random design model based on independent and identically distributed pairs of observations (Xi, Yi), where the regression function m(x) is given by m(x) = E(Yi|Xi = x) with one independent variable. In a nonparametric setting the aim is to produce a reasonable approximation to the unknown function m(x) when we have no precise information about the form of the true density, f(x) of X. We describe an estimation procedure of non-parametric regression model at a given point by some appropriately constructed fixed-width (2d) confidence interval with the confidence coefficient of at least 1−. Here, d(> 0) and 2 (0, 1) are two preassigned values. Fixed-width confidence intervals are developed using both Nadaraya-Watson and local linear kernel estimators of nonparametric regression with data-driven bandwidths. The sample size was optimized using the purely and two-stage sequential procedures together with asymptotic properties of the Nadaraya-Watson and local linear estimators. A large scale simulation study was performed to compare their coverage accuracy. The numerical results indicate that the confi dence bands based on the local linear estimator have the better performance than those constructed by using Nadaraya-Watson estimator. However both estimators are shown to have asymptotically correct coverage properties.

Pilot Study of clinician attitudes to insulin pump therapy: international differences and the need for a greater understanding of the patient perspective

OBJECTIVES: To identify and survey health care professionals (HCPs) attitudes to insulin pump therapy (CSII).

METHODS: Eight specialists were interviewed to explore the attitudes and beliefs about CSII. Responses were analysed thematically and used to inform the design of a new 22-item questionnaire: the Attitudes to Pump Therapy (APT) Survey. The APT was pilottested among 95 HCPs (54% male; 75.5% diabetologists/DSNs, 13.8% general practitioners) at the International Diabetes Federation (IDF) conference, 2006. Results were analysed using non-parametric statistics with bonferroni correction.

RESULTS: Analyses of interview data identified 9 themes: biomedical, perceived control of care/diabetes, technology, quality of life, financial resources, training, education & support, suitability, and evidence-base. Items were designed to reflect these themes with responses scored on a 5-point Likert scale (strongly agree—strongly disagree). No statistically significant differences
were found by gender, HCP speciality, country (and continent) of origin or proportion of patients using CSII. Most notable differences were found in relation to gross domestic product (GDP) and the potential for pump therapy to achieve tight blood glucose control (lower GDP = more agreement: p = 0.001), and result in diabetic ketoacidosis (DKA) (lower GDP = less agreement: p < 0.005). Ranked mean scores showed a split between biomedical/clinical items (N = 11) and items concerned with patient experience (N = 11). Attitudes about biomedical/clinical issues were generally clear (i.e. for 7/11 items, the mean score was “agree”) but less decisive about patient experience (i.e. for 8/11 items, the mean score was “neither agree nor disagree”).

CONCLUSION: Few subgroup differences existed, but those that did may be explained by lack of access to treatment (directly corresponding to GDP). Clinicians’ were generally clear in their attitudes regarding biomedical aspects but less so regarding patient experience. Research focusing on patient-reported outcomes is likely to offer clinicians a greater understanding of the patients’ perspective of insulin pump therapy.