Use of multiple regression methods in health sciences research. Development of linear, logistic, Poisson and Cox regression models for estimation and prediction including covariate adjustment, dummy variables, transformations and interactions.
This distance paper will introduce students to the main regression methods in health
sciences research and is highly recommended for all students that want and/or need
to analyse quantitative data. The paper builds on knowledge and skills learned in
725 and also has a strong applied component. From a public health point of view,
students will learn how to generate and interpret statistical models to adjust for
confounders as well as identifying the variables that have a statistical effect on
the outcome of interest. The regression topics covered include: analysis of variance,
correlation, reliability studies, multiple linear regression and logistic regression.
For this paper, students must have a computer with an Internet connection and be computer
Students will learn R, an open-source, free statistical software under the terms of the GNU General Public License.
|Paper title||Applied Biostatistics 2 - Regression methods|
|Teaching period||1st Non standard period (8 July 2019 - 23 August 2019)|
|Domestic Tuition Fees (NZD)||$1,400.75|
|International Tuition Fees (NZD)||$4,934.75|
- HASC 413 or PUBH 725
- HASC 415, STAT 241, STAT 341
- Limited to
- MA, MAppSc, MClinPharm, MHealSc, MPH, MPharm, MPHC, MSc, DPH, PGDipAppSc, PGDipArts, PGDipHealSc, PGDipMLSc, PGDipPharm, PGDipSci, PGCertPH
- (i) The prerequisite may be waived for students with an equivalent level of knowledge. (ii) MPHC students require approval from the Board of Studies in Primary Health Care to enrol for this paper. (iii) This paper runs for the second half of first semester. (iv) Please note that from 2019, this paper will be offered in the first half of the second semester.
- Students who have completed an undergraduate degree in any discipline or recognised equivalent.
- Department of Preventive and Social Medicine, Dunedin campus: email@example.com
- More information link
- View more information on postgraduate studies in Public Health
- Teaching staff
- Paper Convenor: Dr Josie Athens
- Paper Structure
- Correlation and Reliability
- Simple Linear Regression
- Multiple Linear Regression
- Statistical Modelling
- Logistic Regression
- Participation and contribution: 10% of the marks for this paper will derive from your contribution to Zoom sessions and discussion forums. The marks will not be awarded for the correctness of your contributions, but for making an effort to engage with the question at hand and to use the reading and other learning that you have done to progress the discussion.
- Assignment 1: This assignment, worth 40% of the mark for the paper, assesses your ability to analyse and report statistical models where the outcome is continuous. Emphasis is made on checking assumptions and model simplification. The assignment will also assess your ability to analyse relatively simple reliability studies.
- Assignment 2: This assignment, worth 50% of the mark for the paper, assesses your understanding to analyse and report statistical models where the outcome is influenced by two or more predictors. Emphasis is made on model simplification and understanding interactions.
- Teaching Arrangements
- Compulsory webinar sessions: Tuesday afternoons 4pm-6pm.
- Block day: To be determined (in week 4 or 5 of the Term).
1. Kirkwood, Betty R., and Jonathan AC Sterne. 2003. Essential Medical Statistics. Second Edition. Blackwell.
2. Dalgaard, Peter. 2008. Introductory Statistics with R. Second Edition. Springer.
- Graduate Attributes Emphasised
- Interdisciplinary perspective, Lifelong learning, Scholarship, Critical thinking,
Information literacy, Research, Self-motivation.
View more information about Otago's graduate attributes.
- Learning Outcomes
- Students who successfully complete the paper will be able to
- Demonstrate an understanding of the application of regression methods to help answer scientific questions and the assumptions inherent in the models
- Demonstrate skills in model selection, model fitting and model interpretation for estimation and prediction
- Apply skills to develop a regression model and perform an analysis to help answer a scientific question
- Describe and test the underlying assumptions of the model and carry out simple sensitivity analyses