Statistical Resources from the Sydney Informatics Hub
Training Resources for Statistical Consulting Service, Sydney Informatics Hub, The University of Sydney.
- Chris Howden, Team Lead
- Jim Matthews, Senior Statistical Consultant
- Kathrin Schemann, Statistical Consultant
- Alex Shaw, Statistical Consultant
Read more information about our team. We work within the Sydney Informatics Hub Core Research Facility team to support researchers and research students at the University of Sydney.
We invite those within the university and beyond to browse this list of resources, which include the slides from our workshops and links to our favourite external resources.
Our Workshops and Workflows
Statistical Consulting within the Sydney Informatics Hub offers a modular training programme made up of 1.5 hour workshops, each focusing on a single statistical method. Statistical Workflows giving practical step-by-step instructions applicable in any software are used and include experimental design, exploratory analysis, modelling, assumption testing, model interpretation and presentation of results. They are integrated into Training Pathways to give a holistic understanding of data analysis from a statistical perspective. Researchers are also encouraged to design a custom programme tailored to their research needs.
We continue to develop and improve our workshop offering using feedback and the needs we identify in our statistical consulting service. If you have some feedback about our slides, you can email the author. The workshops will be periodically updated. The version number is shown next to the slides for download.
If you would like a synopsis of each workshop, or if you would like to know the next date that our workshops will be delivered to Sydney University staff and students, click here.
Download workshop slides:
- Research Essentials: Analysing your data v4.1 Attachment: Statistical analysis roadmap for SPSS
- Experimental Design v1.23
- Power and Sample Size Calculation by Jim Matthews 8/22
- Linear Models 1 v1.22
- Linear Models 2 v2.8
- Linear Models 3 v1.41
- Surveys 1 v2.5
- Surveys 2 v1.2
- Meta-Analysis by Jim Matthews 6/22 Example Data file Example R script
- Survival Analysis by Jim Matthews 6/22
- Statistical Model Building v1.5
- Multivariate Statistical Analysis 1: Dimension Reduction v1.5
Acknowledgement of our workshops
The continued acknowledgment of the use of Sydney Informatics Hub facilities including statistical consulting ensures the sustainability of our services. Suggested wording for the use of our workshops and/or workflows:
“The authors acknowledge the Statistical workshops and workflows provided by the Sydney Informatics Hub, a Core Research Facility of the University of Sydney.”
We provide tailored statistical advice to University of Sydney researchers, research students, and affiliates. How we can help
Our team can provide:
- advice on experimental design
- support with data analysis and interpretation
- support with reviewing statistics in your project or grant application
- support with calculation of sample size and experimental power that will be needed in your grant application
If you are an eligible Sydney University student or staff member, you can book a consult by requesting project help
Acknowledgement of statistical consulting service
The continued acknowledgment of the use of Sydney Informatics Hub facilities including statistical consulting ensures the sustainability of our services. Suggested wording for acknowledging help provided during statistical consulting:
“The authors acknowledge the Statistical Consulting service provided by the Sydney Informatics Hub, a Core Research Facility of the University of Sydney.”
Acknowledging specific staff:
“The authors acknowledge the Statistical Consulting service provided by [name of staff] from the Sydney Informatics Hub, a Core Research Facility of the University of Sydney.”
For textbook references, those with Sydney University Library access can use the @Sydney Uni library links
Basic statistical theory
Alex says: A visual, storytelling and R coding-driven approach to introducing basic statistical concepts.
Alex says: Beautiful interactive animations that illustrate important concepts in probability and statistical theory.
Kathrin says: This resource from the University of Melbourne nicely illustrates the calculation and interpretation for measures of associations in observational studies using animation. A good understanding of measures such as odds ratio (OR) and risk ratio (RR) is required for many different analyses, including logistic regression and meta-analysis.
Alex says: Interactive visualisations that illustrate some really important and often misunderstood concepts to do with hypothesis testing.
Research Essentials: Analysing your data
Broman, K.W. and Woo, K.H. “Data Organization in Spreadsheets”
Kathrin says: Many researchers collect data in a spreadsheet like Excel. Unfortunately, there are many pitfalls for analysis when importing data from Excel into statistical software. This paper outlines these pitfalls and gives good advice on data organisation to set you up well for analysis.
Using Statistical Software: R
Kathrin says: This LinkedIn Learning course is a great introductory resource to learn R software coding for statistical analysis. The “tidyverse” is a collection of some of the most versatile R packages for cleaning, processing, modelling, and visualising your data. The course is free with a University of Sydney login.
Using Statistical Software: SPSS
Jim says: This LinkedIn Learning course is an ideal way to introduce yourself to the SPSS environment and learn everything from the basics of data manipulation and visualisation right through to linear and logistic regression. Having access to exercise files, speaker transcripts and captions, Q&A and your own notebook make it really accessible and user friendly. You’ll be browsing the other content on LinkedIn Learning after you have finished this one. The course is free with a University of Sydney login.
McCormick, K. et al. “SPSS Statistics for Data Analysis and Visualization” @Sydney Uni library
Jim says: There are many textbooks on SPSS that cover the basics in detail (my favourite is the one by Andy Field), but not many that cover the more advanced features of SPSS. So once you have learnt the basics, this is a great reference to learn more about advanced statistics, data visualisation, predictive analytics and advanced programming techniques.
SPSS Contextual Help
Jim says: When you open a dialog box for an analysis and you are faced with a variety of options that seem to be written in a foreign language, click on the help button to see the details. It is actually quite helpful. If that is not enough detail, go to the main menu “Help>Documentation in pdf format” and download the manuals.
Power Analysis (and Sample Size Calculation)
G*Power software Available for free download at the G*Power website
Jim says: There are lots of software options for sample size calculation depending on your needs. G*Power does the job for a wide range of scenarios with a minimum of effort.
Julious, S. A. Sample Size for Clinical Trials @Sydney Uni library
Jim says: Clinical trials often involve testing variations on the classic hypothesis of a difference between groups. The research question in a clinical trial might require a test of “equivalence”, “superiority” or “non-inferiority” for example. This book by Julious sets out how to handle each option, including useful worked examples.
Linear Models (ANOVA, Simple linear regression, multiple regression, etc)
Chris says “A series of short usually 5-15min videos which simply explain a lot of the basics”
Chris says “A series of short 5-20min videos ranging from a simply introduction to more complex details”
Borenstein, M. “Introduction to meta-analysis” @Sydney Uni library
Jim says: Borenstein is a world renowned expert on meta-analysis and is responsible for developing the Comprehensive Meta-Analysis (CMA) software tool. His book is well written and is always my first choice when looking for answers to meta-analysis questions.
Higgins and Green (Editors) “Cochrane Handbook for Systematic Reviews of Interventions” @Sydney Uni library
Jim says: Cochrane is an international network founded in 1993 that is without peer in promoting good scientific practice for systematic reviews of medical research. If you follow the Cochrane handbook you can be assured your methods are rigorous, defensible and following best practice.
Harrer, M “Doing Meta-Analysis in R” website of book by the same name
Jim says: If you are planning to use the R language with a package such as meta or metafor then you will find this website really useful. It covers some topics that are not included in basic textbooks such as Network meta-analysis and Bayesian meta-analysis.
Collett, D. “Modelling survival data in medical research” @Sydney Uni library
Jim says: This is a good all-round reference for survival analysis. It is written at an intermediate level (of statistical knowledge) but in an easy to read style. A highly recommended starting point.