Design of studies for quantitative research. Causal inference and models, culturally appropriate design, western and tikanga research ethics, complex surveys, experiments, and observational studies addressing causation and prediction.
Understanding the design of a research study – the data generation process – is critical for formulating appropriate statistical analyses and interpreting the results. Optimising research design to ensure reliable answers research questions is an important area of statistics, both for methodological research and in the application of statistical methods. The aim of this paper is to provide students with the knowledge and skills to translate a research aim into specific study objectives, construct a culturally appropriate study design to address the objective(s), and determine a suitable statistical analysis. Topics include causal inference, research ethics (western and tikanga), complex surveys, experiments, and observational studies addressing causation and prediction.
| Paper title | Research Design and Methods |
|---|---|
| Paper code | STAT424 |
| Subject | Statistics |
| EFTS | 0.1667 |
| Points | 20 points |
| Teaching period | Semester 1 (On campus) |
| Domestic Tuition Fees (NZD) | $1,206.91 |
| International Tuition Fees | Tuition Fees for international students are elsewhere on this website. |
- Prerequisite
- STAT 401 or (STAT 270 and STAT 310), or equivalent (contact department for further information)
- Restriction
- STAT 311
- Contact
Professor Katrina Sharples katrina.sharples@otago.ac.nz
- Teaching staff
Professor Katrina Sharples and Associate Professor Phillip Wilcox
- Paper Structure
The main topics are:
- Introduction to the course, Vision Mātauranga and Māori research.
- Overview of case studies.
- Research questions, aims and objectives; types of design.
- Causal inference and statistical models.
- Measurement.
- Surveys.
- Experiments.
- Observational studies addressing causation and prediction.
- Māori research and ethical frameworks, and culturally informed co-design; research ethics.
- Evaluating and improving the design (sources of variation, controlling random error, minimizing bias).
- Teaching Arrangements
In common with STAT311: 3 lectures per week and weekly tutorials where students work in groups with a mix of backgrounds to reflect research design in practice.
STAT424 material: fortnightly seminars will be held relating to advanced technical aspects of design.
- Textbooks
Readings will be provided.
- Graduate Attributes Emphasised
Global perspective, lifelong learning, scholarship, communication, critical thinking, cultural understanding, ethics, environmental literacy, information literacy, research, self-motivation, team-work.
View more information about Otago's graduate attributes.- Learning Outcomes
On completion of this course students will be able to:
- Distinguish different types of research questions and identify appropriate research objectives and designs for a given research question.
- Describe the key aspects of causal inference as it relates to research design, including causal effects, roles of confounders, mediators and effect modifiers, and determine an appropriate estimand and statistical model to address a research objective.
- Describe the ethical considerations (western and tikanga), the role of culturally informed co-design in the design, conduct, analysis and reporting of a research study.
- Identify the relevant sources of variation, correlation, the units of replication and potential sources of bias in a proposed complex study design, and use this knowledge to independently optimise the planned design, conduct and analysis.
- Carry out an advanced statistical evaluation of a study design, including power and sample size determination.
- Design and plan the analysis for a complex survey, an experiment, and an observational study.
- Work effectively in a multidisciplinary group to develop a study design.
- Explain design choices clearly to a general scientist audience in both oral and written communication.
