Course for undergraduates (including BSc Public Health):
COURSE DESCRIPTION: This course will provide an overview of the use of basic statistical concepts and methods in the public health field. Emphasis will be placed on conceptual understanding of statistical approaches to medical data rather than on theory and equations. Students will learn the importance of the correct use of statistical techniques in addressing questions of public health importance such as (1) How can we tell if global warming really happening what is causing it? (2) How does diet affect individual risk of chronic diseases? (3) Is gambling in casinos always a losing proposition? (4) Does exposure to other people’s cigarette smoke really cause health problem? The course will enable students to develop the statistical literacy necessary to interpret the vast amounts of information they must process for personal decision making and less vulnerable to manipulation. Students will also be introduced to a user-friendly software for performing basic statistical analyses.
LEARNING OUTCOMES:
By the end of this course, student should be able to: have an understanding of the fundamental concepts of probability and basic statistical analysis. They will be able to apply this knowledge towards improved understanding of reports on research findings and better personal decision making.
Courses for all CUHK Medical Faculty Research Postgraduate students:
COURSE DESCRIPTION: This course introduces basic statistical concepts and methods. The emphasis of the course is on practical applications: choosing the correct method for particular datasets and correct interpretation of the analysis results. Examples from different disciplines of public health including chronic and infectious disease epidemiology, environmental health, and health policy will be used to illustrate the use of biostatistical methods in answering important public health questions.
LEARNING OUTCOMES:
Student should be able to:
1. understand the importance of biostatistics in public health and medical research.
2. develop a conceptual understanding of basic biostatistics,
3. critically read and understand the statistical methodology and results sections of medical and public health research papers.
4. be capable of carrying out basic statistical analyses using SPSS statistical software.
COURSE DESCRIPTION: This course introduces basic statistical concepts and methods. The emphasis of the course is on practical applications: choosing the correct method for particular datasets and correct interpretation of the analysis results. Examples from different disciplines of public health including chronic and infectious disease epidemiology, environmental health, and health policy will be used to illustrate the use of biostatistical methods in answering important public health questions.
LEARNING OUTCOMES:
Student should be able to:
1. understand the importance of biostatistics in public health and medical research.
2. develop a conceptual understanding of basic biostatistics,
3. critically read and understand the statistical methodology and results sections of medical and public health research papers.
4. be capable of carrying out basic statistical analyses using SPSS statistical software.
Common course for SPH taught postgraduate programmes
BIOS5001: Introduction to Biostatistics (3 units)
COURSE DESCRIPTION: This course introduces basic statistical concepts and methods. The emphasis of the course is on practical applications: choosing the correct method for particular datasets and correct interpretation of the analysis results. Examples from different disciplines of public health including chronic and infectious disease epidemiology, environmental health, and health policy will be used to illustrate the use of biostatistical methods in answering important public health questions.
LEARNING OUTCOMES:
Student should be able to:
1. understand the importance of biostatistics in public health and medical research.
2. develop a conceptual understanding of basic biostatistics,
3. critically read and understand the statistical methodology and results sections of medical and public health research papers.
4. be capable of carrying out basic statistical analyses using SPSS statistical software.
M.Sc. & PgD programmes, MPH concentration in Epidemiology and Biostatistics
COURSE DESCRIPTION: This course will provide a foundation for the practical analysis of data for which the primary outcome is a continuous variable. The course will begin with an introduction to ‘real-world’ data analysis with a motivating example looking at predictors of infant birthweight in Hong Kong. Methods for multivariate analysis of predictors of continuous outcomes including one-way and two-way ANOVA and multiple linear regression will then be discussed in detail with an emphasis on correct use of these methods in practice.
LEARNING OUTCOMES:
Student should be able to:
1. understand and evaluate the use of linear models in the medical literature in an intelligent manner.
2. develop skills in analyzing epidemiological data with continuous outcomes using linear models and to understand the basic principles that underlie research designs and statistical inference.
3. perform fundamental statistical procedures for research projects involving continuous outcomes and interpret results.
BIOS5003: Categorical & Survival Data Analysis
COURSE DESCRIPTION: This course will provide a foundation for the practical analyses of categorical and time to event (survival) data. The course will cover the use of logistic regression models for use with binary outcomes and Cox proportional hazards regression models for time to event outcomes. Practical application of these models will be emphasized and model building and the checking of model assumptions will be covered in detail.
LEARNING OUTCOMES:
Student should be able to:
1. understand the concepts, assumptions and logic involved in statistical methods commonly used in medical research including categorical data analysis and time-to-event data analysis.
2. develop appropriate statistical models for the data and correctly interpret the results.
COURSE DESCRIPTION: The objective of this course is to provide students with a theoretical and practical knowledge of the issues involved in the design, conduct, analysis and interpretation of randomized clinical trials. We will discuss the basic principle of randomization and its importance, proper randomization and blinding procedures, choice of control arm, the importance of clear definition of endpoints, methods to calculate sample size, other statistical considerations and ethical issues in clinical trials. Attention will be given to the problems of conducting clinical trials in both single center and multi-center, and covers trials initiated by industry as well as trials in academic setting. Students will be trained to develop skills to properly design clinical trial, critically analyze and carry out research and to communicate effectively.
LEARNING OUTCOMES:
Student should be able to:
1. understand the advantages and disadvantages of various designs in clinical research.
2. understand the concepts of randomization in controlled clinical trials.
3. develop a protocol for a clinical trial to address the research questions.
4. have a general knowledge of the statistical issues commonly encountered in clinical trials.
5. be aware of the ethical issues in clinical trials.
6. have an appreciation of the Good Clinical Practice (GCP) requirements in the operation of clinical trials.
7. learn some basic elements of data management and quality assurance in multi-center clinical trial set up.
M.Sc. in Epidemiology and Biostatistics
BIOS6001: Topics in Linear Models (2 units)
COURSE DESCRIPTION: This course will cover advanced statistical modeling techniques for use with complex datasets. Topics will include Poisson and Negative Binomial regression for count outcomes, repeated measures ANOVA, GEE models and multilevel models for longitudinal data, multilevel models for clustered data.
LEARNING OUTCOMES:
Upon completion of this course students will understand the reasons that more complex statistical models need to be used for datasets for which the assumptions of linear or logistic regression are not valid, such as datasets with ordinal or count outcomes, longitudinal or clustered data, and data with non-linear associations between variables. They will understand which models should be used for each of these situations, how to fit and interpret these models, and how to check the assumptions of these models.
BIOS6002: Topics in Multivariate Analysis (2 units)
COURSE DESCRIPTION: This course will cover methods importance in the analysis of data collected from questionnaires. Both exploratory and confirmatory factor analysis (under the framework of Structural Equation Models) will be discussed.
LEARNING OUTCOMES:
Upon completion of this course students will understand the reasons that more complex statistical models need to be used for datasets for which the assumptions of linear or logistic regression are not valid, such as datasets with ordinal or count outcomes, longitudinal or clustered data, and data with non-linear associations between variables. They will understand which models should be used for each of these situations, how to fit and interpret these models, and how to check the assumptions of these models.