For general information regarding scholarships, please see the Scholarships and Research Grants webpage.
These Summer Scholarships are available to suitable undergraduate students.
Aim: To give students research experience and acquaint them with the research activities of the Department and with potential graduate supervisors.
Emolument: $5,000 non-taxable.
Duration: Full-time employment for eight to twelve weeks (400 hours).
Updated Friday 4 September to hide projects Under Offer to Selected Student.
Updated Monday 7 September to include three new projects from Health and Medical Science.
This project is offered by the Clinical Trials Research Unit, School of Population Health, The University of Auckland.
A project underway as part of a collaboration between the Clinical Trials Research Unit (Auckland University) and the George institute (Sydney, Australia). The general aim of the research group is to study the burden of disease due to high heart rate in selected countries. The specific aim is to provide estimates of the burden of disease attributable to non-optimal heart rate by age and sex for adults aged ≥ 30 years in key world regions.
The methodology utilised for these analyses will follow that established for the World Health Organisation Global Burden of Disease study. This methodology was in part developed by members of the research team, who have been involved in several burden of disease analyses to date.
The analyses are important for gauging the potential for prevention and treatment strategies, and provide input into decisions around strategies to improve population health, and developing health policy and research priorities. Burden of disease analyses give a tangible idea of health risk by providing a measure of the public health impact of a risk factor such as heart rate. Furthermore, there is a demand for developing expertise in this area to assist health researchers with biostatistical techniques required for these types of analyses. The aim of this summer studentship is to assist with carrying out these analyses. Supporting development of these types of skills could be of great practical value in the area of public health research within New Zealand.
The successful student would take an active role in data analysis and interpretation of this study data and work alongside researchers at the CTRU. The key task would be to conduct analyses under supervision and interpret the analysis. At the end of the project we expect the student to present a report outlining their findings and assist with preparation of a manuscript for publication.
Through the project the student will
Contact: Stephen Vander Hoorn, s.vanderhoorn@stat.auckland.ac.nz
This project is offered by the Clinical Trials Research Unit, School of Population Health, The University of Auckland.
Reliable evaluation of new smoking cessation strategies and monitoring of existing programmes are important components of tobacco control. Self-reported abstinence should be verified to accurately assess effectiveness. However, current methods of verification (e.g. measuring carbon monoxide in expired breath or cotinine in saliva or urine) are not practicable or affordable for large cessation intervention trials or population surveys. Smokers have a lower fundamental frequency of voice compared to non-smokers, and there is some evidence that acoustic parameters change when smokers quit. However, there are no studies to date on how accurately this change can be measured.
A feasibility study has been carried out to address this research area and data has been collected from smokers and recent quitters to answer questions around this issue. The aim of this summer studentship is to assist with several new analyses of the RASP data to help address this issue.
The successful student would take an active role in data analysis and interpretation of the RASP data and work alongside researchers at the CTRU. The key task would be to conduct analyses under supervision and interpret the analysis. At the end of the project we expect the student to present a report outlining their findings and assist with preparation of a manuscript for publication.
Through the project the student will
Contact: Stephen Vander Hoorn, s.vanderhoorn@stat.auckland.ac.nz
This project is offered by the Clinical Trials Research Unit, School of Population Health, The University of Auckland.
A large national survey of children and young people’s physical activity and sedentary behaviours has just been completed. Physical Activity and sedentary behaviours were measured both using a computerised self-report recall questionnaire (Multimedia Activity Recall Questionnaire for Children and Adolescents) and accelerometers (movement devices). Each approach has its own costs and benefits; however comparison between the two methods is required to determine the preferred method. The aim of this study is to assist with several analyses of the Mission-On data to address this issue.
The successful student would take an active role in data analysis and interpretation of the Mission-On data and work alongside biostatisticians at the CTRU. The key task would be to review the available Mission-On data, conduct analyses under supervision and interpret the analysis. At the end of the project we expect the student to present a report outlining their findings and assist with preparation of a manuscript for publication.
Through the project the student will
Contact: Stephen Vander Hoorn, s.vanderhoorn@stat.auckland.ac.nz
This project is offered by the Woolf Fisher Research Centre, The University of Auckland.
A National project was set up to conduct the first nation-wide evaluation of Schooling Improvement Initiatives. One key goal was to examine the improvements in achievement across these initiatives. The first part of the summer scholarship involves a literature review on value-added measures appropriate for measuring educational improvements (e.g., cost-benefit analysis of educational programmes/studies, treatment of missing data and its influence on value-added studies, educational studies that have applied segmentation methods or other multivariate analysis methods for survey data in relation to student achievement). Based on the literature review, in the second part of the scholarship, the student will undertake a series of exploratory studies on the data (e.g., applying suitable multivariate analysis methods to the data in relation to the student achievement, studying the student-level absence/transience patterns and then relating the patterns to student achievement using a value-added specification from the predictors collected in the study).
This would suit students who have familiarity with SAS and R and have good marks in STATS 302 and/or STATS 330.
Contact: Mei Kuin Lai, mei.lai@auckland.ac.nz
The Wellcome Trust Case Control Consortium genotyped cohorts of 2000 individuals with each of 7 common complex diseases (such as diabetes and hypertension) and 3000 control individuals on 500,000 genetic markers. They published analysis of these data using standard methods.
In this project, the student will take one of the disease cohorts and the controls and analyze the data with new methodologies (haplotypic analysis and/or identity by descent detection) with the goal of finding new associations between genetic variation and the disease, in order to improve understanding of genetic factors contributing to susceptibility to the disease. The student will perform a literature search in order to find published associations from other data sets on the same disease, and will compare these results from other studies with the results of his/her analysis on the Wellcome Trust data set.
Contact: Sharon Browning, browning@stat.auckland.ac.nz
This project is offered by Plant and Food Research.
You get your hands on real large scale data sets. We need your help to analyse and make sense of the information contained in the data.
This project involves full-time statistical analysis of previously collected data. It primarily involves the development of statistical protocols that can be repeated with similar types of data at a later stage. The developed routines must be easy to run in standard software, preferably SAS.
The successful candidate must be able to handle large scale data sets and be proficient in tasks such as descriptive statistics, and univariate analyses. Some experience in multivariate analyses, notably PCA/ Factor analyses, MANOVA, discriminant analysis and cluster analysis is required. There will also be issues to consider relating to variable selection and weighting of cases and variables. The successful applicant will also be expected to show initiative and suggest additional analyses that may uncover systematic patterns in the data.
The successful candidate will work in close collaboration with the scientists that collect the focal data, as well as with experienced biometricians.
Contact: Nihal de Silva, http://careers.plantandfood.com/jobseeker/, Job reference: 1186
This project is offered by the School of Biological Sciences, The University of Auckland.
Metabolomics is the study of the chemical fingerprints left behind by specific cellular processes. Gas chromatography coupled with mass spectrometry (GC-MS) provides a powerful analytical technique for the identification and quantification of the metabolites, or compounds, within a cell, tissue and biofluid of an organism. During GC-MS, information is collected on each compound’s GC retention time and its specific fragmentation pattern, or mass spectrum. Around only 30% of compounds are identified by cross-referencing this information against a mass spectral library, severely limiting our ability to understand the biological functions of cellular processes. The aim of this project is, therefore, to explore methods which will enable us to separate ‘real’ mass spectral peaks from noise. These peaks will then be analysed using univariate and/or multivariate statistical techniques to explore how they change under different experimental conditions.
Key techniques/skills that the student will learn: Understanding of how GC-MS metabolomics data is generated; Application of statistics to GC-MS metabolomics data; Statistical analysis and programming using the R software.
Prerequisites: Minimum Stage 2 statistics (although Stage 3 statistics is preferred). Experience with the R software and some knowledge of biochemistry would be an advantage.
Contact: Dr Kathy Ruggiero, k.ruggiero@auckland.ac.nz
University of Auckland researchers have been monitoring subtidal reef communities and sedimentation levels on shallow reefs in the Hauraki Gulf for the Auckland Regional Council (ARC) since 1999. This provides a valuable long-term baseline to assess the effects of continued urbanization of coastal catchments on the adjacent reefs. Increased sedimentation from development is considered to be one of the major threats on coastal ecosystems. Sediment traps are used to estimate levels of sedimentation on the reef and relate to variation in the reef communities among sites over time. However, it is poorly understood how well sediment trap data actually relates to sedimentation on the reef versus other environmental drivers (e.g., rainfall, wind and waves). Gaining a better understanding of how sediment trap data relates to sedimentation and other environmental factors is necessary for management to evaluate the monitoring program and interpret changes in reef communities. This project will involve a spatial and temporal analysis of how sediment trap data relates to a suite of environmental factors including data on water quality (e.g., chlorophyll, turbidity, nutrients), rain fall and wave exposure. Analyses will also be carried out to investigate how these variables relate to variation in reef communities among sites in space and time. There may also be opportunities to work in the laboratory and in the field (potentially involving scuba diving) with other researchers. Contact: Nick Shears, nickshears@xtra.co.nz
Project will involve a lot of data extraction and processing as well as statistical analysis. This project would suit someone who has done one of STATS 220, STATS 779, STATS 380 or STATS 782 and wants to learn something about SQL, R, Perl, cluster analysis and forensic genetics.
Contact: James Curran, curran@stat.auckland.ac.nz
In 1995/1996 I wrote a programme which demonstrated graphically various population genetic phenomena such as inbreeding, drift, migration and mutation. The programme, called Popgen, ran under Windows 3.11 and Windows 95, and was written in C++.
I would like to revive Popgen and make it publically available for download.
In this project, the student would re-write and update Popgen in a modern (portable) language such as Java or C#. The skills needed are moderate to strong programming with at least the rudiments of computer graphics (i.e. you understand the concept of drawing on the screen, changing colours etc). From this project you will learn some population genetics, random number generation, and some ideas about writing efficient fast programs.
Contact: James Curran, curran@stat.auckland.ac.nz
This project is to search for and document expected information matrices in the statistical literature, and then implement these within my VGAM package for R. For each of these, the derivation of initial values and random variate functions (dpqr) where possible are needed. The consequence of this project is that many multivariate distributions will be implemented within the useful framework provided by the supervisor. We will also implement mixture models such as two negative binomial distributions. The project would be suitable for a student with a good grasp of statistical theory (including logistic regression) and R programming.
Contact: Thomas Yee, t.yee@auckland.ac.nz
Ordination is a multivariate technique for modelling multispecies-environmental data simultaneously. Currently there are no methods to handle ordinal species data, which is a very common form.
This project is to implement a new method for ordinal ordination recently developed by the supervisor. It will firstly involve converting general Fortran/Ratfor code to C. Applications to several animal and vegetation data sets will be made. Time permitting, the switch from LINPACK to LAPACK will be investigated and implemented. The project would be suitable for a student with very good programming skills (R and C, and ideally Fortran too but not necessary) and experience with logistic regression.
Contact: Thomas Yee, t.yee@auckland.ac.nz
The rates of evolution, i.e., the pace at which molecular sequences accumulate substitutions, vary extensively along genes and chromosomes. For the sake of simplicity, modern phylogenetic methods assume that rates can vary freely along genomes (the substitution rate of a given gene is considered as an independently and identically distributed random variable). However, experimental evidence suggests that genes that are involved in similar functions tend to cluster together on the genome. Such clustering is expected to impact on the autocorrelation of substitution rates along genomes, because genes involved in a common function need to evolve in a concerted fashion. The goal of this project is to investigate the patterns of autocorrelation of substitution rates along genomes. The statistical tools that will be developed during the first part of the project will then be applied to the analysis of a large yeast data set for which information on gene and chromosome positions available.
This project will be supervised by Stéphane Guindon (Department of Statistics). The student will be hosted by the Computational Evolution group and will therefore interact with graduate students in statistics, mathematics and computer sciences. Also, the student is expected to attend the seminars organised by the Bioinformatics Institute (School of Biological Sciences). A background in statistics, biology or bioinformatics is required. Programming skills are also expected.
Contact: Stéphane Guindon, guindon@stat.auckland.ac.nz
PhyML is a widely used software program that estimates phylogenetic trees from alignments of biological sequences. It is written in C ANSI and relies on a standard unix-type command line user interface. The goal of this project is to provide a graphical user interface to PhyML that could be used on most operating systems.
This project therefore requires basic programming skills but the student will also be given the opportunity to learn about maximum likelihood estimation of phylogenetic trees and molecular evolution.
Contact: Stéphane Guindon, guindon@stat.auckland.ac.nz
I have some existing code, written in R, for generating random environments and also random walks in those random environments. I would like people to be able to do the same kinds of simulations interactively via my webpage. For an example for a different problem see http://www.stat.auckland.ac.nz/~mholmes/javastuff/urn_model/urn_model.html
This project therefore involves writing simulation code in an appropriate language (e.g. Java) as well as creating an appropriate web interface from which to run the simulations.
Contact: Mark Holmes, mholmes@stat.auckland.ac.nz
This project is offered by The Department of Statistics and Centre of Methods and Policy Application in the Social Sciences (COMPASS), Faculty of Arts, The University of Auckland.
The New Zealand Social Science Data Service (NZSSDS) provides access to a number of survey data sets through the use of Nesstar software. One-way frequency tables and descriptive statistics are freely viewable while cross tabulations, basic regression analyses and data downloads are available through registration and the meeting of other criteria, terms & conditions.
The use of NZSSDS in teaching has been piloted, using a subset of data from the New Zealand Quality of Healthcare Survey. Students signed our standard ‘user undertaking’ agreement forms and accessed the data in a lab environment. However, ideal would be to completely avoid any issues of security and have data sets that were freely available for students to download and work with on their own – this would also support the use and learning of standard statistical packages that is going on anyway, at least in the statistics department.
To this end, data sets would likely need to be subsetted, and then ‘perturbed’ in other ways so as to make them ‘safe’ for release while still giving believable results.
In the first instance we would be targetting four surveys that have been discussed extensively in STATS 740 over the years:
The first three of these have already been archived in some form on NZSSDS, and the last would be expected to be there before the commencement of a studentship. This will provide the basis for teaching data sets to be archived – the metadata will be marked up and data viewable within the Nesstar Publisher software, in the use of which skills will need to be developed. It would also be most useful for this to be combined with writeup of documentation for its use for marking up and archiving data.
Contact: Peter Davis, pb.davis@auckland.ac.nz
This project is offered by The Department of Statistics and Centre of Methods and Policy Application in the Social Sciences (COMPASS), Faculty of Arts, The University of Auckland.
A new collaborative approach in research is to make traditional academic papers obsolete by creating research objects that contain all the material needed to understand a piece of research, including underlying data, metadata and research outputs. In practice, this combines a data repository with a research method making published research more transparent, reusable and reproducible.
The NZ Social Science Data Service (NZSSDS) is an ongoing initiative being undertaken by COMPASS. The vision is that data sets, all well documented with metadata for users, will be made publicly available for examination, basic analysis online and authorised download (via Nesstar software) to the wider research community. Key published papers will be documented along with computer code for data manipulation and analysis. There are already over 20 data sets available on the NZSSDS website. The scholarship will contribute to this system, so that the service can easily be used by people coming to the website.
The student will have good computing and statistical skills.
Contact: Peter Davis, pb.davis@auckland.ac.nz
These Summer Scholarships are available to suitable undergraduate students.
Aim: To give students research experience and acquaint them with the research activities of the Department and with potential graduate supervisors.
Emolument: $5,000 non-taxable.
Duration: Full-time employment for eight to twelve weeks (400 hours).
For general information regarding scholarships, please see the Scholarships and Research Grants webpage.