Thing 9: Statistical Analysis Software
This week’s post will help you to learn about a selection of different statistical packages, how to access them (on campus and outside the university) and where to get help.
Statistical analysis is an integral part of evaluating research data and presenting results in any research field. There is a broad range of software packages available that facilitate this. Some are freely available from the internet, but most of the advanced and more comprehensive statistical software are available only as commercial packages.
The more popular statistical packages among the university community are listed in the table below.
Choosing which package to use:
Not all the software packages do every type of analysis. Some packages are very powerful and flexible, hence can perform a wide variety of statistical analysis. In contrast, others have specialised functions like creating graphs, sample size calculations, genetic analyses etc. Therefore, your choice of a statistical package will depend on the nature of your research and the type of analysis you will be doing. Factors to be considered when choosing the right statistical tool for your particular research:
- The number, size and complexity of data sets you have to be analyzed, and the capacity of the package.
- The functionality and the output you require.
- Packages used by others in your department and/or others doing similar research.
- What is currently available at the University of Melbourne, or what can be purchased through your department.
|Main Features||Considerations||Availability at University of Melbourne|
|NVivo||Designed for qualitative research with rich text-based and/or multimedia information. Intended to assist researchers to organize and analyze non-numerical unstructured data. Used mainly in social sciences and marketing||
|SPSS||Primarily for statistical analysis. Especially useful for analyzing large-scale survey data.||
Minitab is a command- and menu-driven software package for statistical analysis. Analysis can be performed using drop-down menus or syntax.
Support for design and analysis of experiments (including factorial, response surface, mixture, and Taguchi designs).
SAS (Statistical Analysis System).
In addition to statistical analysis, it also allows programmers to perform report writing, graphics, business planning, forecasting, quality improvement, project management and more.
MATLAB is intended primarily for numerical computing.
MATLAB allows matrix manipulations, plotting of functions and data, implementation of algorithms, the creation of user interfaces, and interfacing with programs written in other languages, including C, C++, C#, Java, Fortran and Python.
It is primarily used in the fields of engineering, science, and mathematical economics.
Genstat is used in a number of research areas, primarily in agriculture, biology, genetics, ecology, environment (forestry and soil), food science and medical and pharmaceutical. But it is also used in finance, industry, engineering, statistics, and mathematics.
For those with more technical knowledge, it offers more power and flexibility through a command language interface.
|STATA|| STATA is a general-purpose statistical software package used primarily in the fields of economics, sociology, political science, biomedicine, and epidemiology.
Its capabilities include data management, statistical analysis, graphics, simulations, regression, and custom programming.STATA uses a point-and-click interface as well as command syntax.
R provides access to a variety of statistical and graphical techniques including linear and non-linear modelling, classical statistical tests, time-series analysis, classification, and clustering.
Functionally, the program is considered comparable to SAS.
However, R has the advantage of being open-source and thus freely available.
Have a look at the MIT Library Guide for statistical packages. Compare the strengths and weaknesses of the statistical packages they list. Does that help you in choosing one that will best fit your research?
Also look at UCLA’s Institute for Digital Research and Education (IDRE) website. Under the ‘software‘ tab you can read about the features of some statistical packages. Look especially at the information under ‘Statistical Analysis‘ that goes into detail about the statistical features of each software.
University of Melbourne Library Guides
- Statistical and Mathematical Software – includes links to support materials and further reading.
- Statistics and Data –
Need help with a stats issue or want some training?
Training and support options for University of Melbourne students and staff
- The Statistical Consulting Centre provides a range of services to the University of Melbourne academic community.
- The Melbourne Statistical Consulting Platform (MSCP) provides advice on statistical methods to graduate researchers and staff from any discipline across the university.
- Consultants can provide one-on-one advice to assist with any stage of a quantitative research project from planning and data collection to analysis and reporting.
- The MSCP deliver short courses in statistical methods in collaboration with the Statistical Consulting Centre.
- The Graduate Students Association
- offers a number short courses including for NVivo throughout the year.
- Facebook group for R users
- The Research Platform Services team has a new help desk service for researchers, by researchers.
- The Help Desk is located on level 3, Eastern Resource Centre Foyer – adjacent to the ERC Library. Regular hours yet to be advised, however, it is open on Monday 18 September from 10.00am – 3.00pm and on Thursday 21 September from 11.00am – 3.00pm.
- Researchers are welcome to drop in at any time to get help and support, including free training and advice on tools such as R and R Studio, Matlab, Python, Omeka as well as Data storage, HPC and compute.
- Contact Research Platform Services for more information.
- The Social and Cultural Informatics Platform (SCIP) supports the investigation, development, and use of advanced research tools to analyse or visualise data and information in different forms. SCIP offers training and support consultations to all University researchers and collaborators. More information or contact SCIP.
Other free statistical software packages and resources
- Free Statistics – free open-source statistical software and comparison
- StatsSci.org – a portal for statistical science, with special reference to Australia
- StatPages.net – free software packages you can install on your computer for stand-alone (offline, non-Internet) computing. Useful categories, e.g. biostatistics and epidemiology; surveys, testing, and measurement.
- Epidemiologic software and calculators – compiled by staff at McGill University.
- Choosing the correct statistical test in SAS, STATA, SPSS and R – compiled by UCLA’s Institute for Digital Research and Education.
This post was written by Dr. Harshanie Habarakadage (Library Cadet, Brownless Biomedical Library) and Guido Tresoldi (Liaison Librarian – Science and Engineering)