Statistical software

Deze pagina lezen in het Nederlands.

Author: dr. ir. N van Geloven
Authorship

This page was translated from the Dutch-language original using AI and has not been fully reviewed. Where the two differ, the Dutch version takes precedence.

Contents

• General statistical packages
  • SPSS
  • R
  • Stata
  • SAS
• Randomization software
• Sample size calculations and power analyses
  • nQuery Advisor
  • PASS
• Meta-analyses

General statistical packages

SPSS

SPSS (short for Statistical Package for the Social Sciences) is by far the most widely used package in medical research because of its user-friendliness (almost all analyses can be run via menus).

• Manual for basic analyses in SPSS written by Saskia le Cessie (LUMC) - This manual is limited to the statistical methods most commonly used in medical research, such as descriptive statistics, creating graphs, statistical methods for comparing two or more groups of observations, calculating correlations, and performing linear, logistic, and Cox proportional hazards regression.

R

R is a freely available statistical package that can be downloaded at http://www.r-project.org. Go to Download -> CRAN, select a 'mirror', and install the package. Because R has a so-called open source structure, new statistical developments are directly available via new downloadable packages. It is a very powerful and flexible package.

• Introductory manual: Simple statistical analyses in R.pdf. If you work through the first few pages of this manual, you will know the basics of the package.
• Short refcard Handy cheat sheet with commonly used functions in R.
• Handy YouTube instructional videos on importing data from SPSS into R: video 1, video 2, video 3.
• YouTube channel 'Intro to R'.

Stata

SAS

Randomization software

It is preferable for the randomization of an RCT to be computer-driven and/or performed by an independent person. Two computer-driven options for this are ALEA and Castor.

ALEA is an application designed by the Netherlands Cancer Institute. ALEA is validated for use in GCP trials. With ALEA, you can carry out the randomization procedure online. When a new patient is included, the website can be queried to determine which study arm that patient has been randomized to. Patients are randomized via the ALEA website within their own protected study environment. ALEA can support all types of randomized studies.

Castor is an online platform for building Electronic Case Report Forms (eCRFs). This platform can also perform a form of randomization: stratified block randomization with varying block sizes.

Sample size calculations and power analyses

nQuery Advisor

For calculating sample sizes, the program nQuery Advisor is available within Amsterdam UMC. nQuery is a validated and user-friendly program that offers the ability to calculate the desired group size and statistical power for various study designs and types of data.

PASS

PASS is an alternative program for calculating sample sizes. It supports more different designs than nQuery. A free trial version of the program can be downloaded from the PASS website.

Meta-analyses

Within the general statistical packages, there are several options for performing meta-analyses:

• There are various packages for R for performing meta-analyses. A good starting point is the CRAN task view on meta-analysis.

• STATA has had a meta-analysis command since version 16, and various packages for meta-analysis have also been written for it. You can read more about this on the STATA website.

• The most recent versions of SPSS also offer several pre-programmed routines. This SPSS documentation shows which ones.

In addition to the general statistical packages, Review Manager (RevMan) is a commonly used option. RevMan is designed for writing a Cochrane Review and, uniquely among those listed here, facilitates both data extraction and various types of meta-analyses. The (offline) desktop version (RevMan 5), however, is no longer supported by Cochrane. It has been replaced by RevMan Web, for which a Cochrane account is required.

Note: For diagnostic meta-analysis, the meta-analysis routines in RevMan and SPSS version 28 are not sufficient. In that case, two outcome measures are relevant (for example, proportion of true positives and proportion of false positives); these are dependent on each other and need to be analyzed simultaneously/within a single model. Such more advanced analyses are offered elsewhere, including in SAS, R, and STATA and in R-based (web) applications.

---

Finished reading? You can go to the Overview of all topics, or to Choosing a test for help picking a suitable test or analysis. Want to learn more about biostatistics? Then follow the AMC e-learning Practical Biostatistics. Did you find something on these pages that isn't correct? Does a link not work? Or would you like to contribute to the wiki? Then get in touch with us.

This biostatistics wiki is an initiative of the former statistics helpdesk of Amsterdam UMC, location AMC. Amsterdam UMC staff can request statistical support via intranet.