Data Analysis / Statistics
- We translate research ideas into formal scientific questions and investigate them with the appropriate study design and statistical methods.
- We command a wide range of statistical, machine learning, and data science methods.
- We supplement our statistical results with meaningful interpretations.
- We visualize data and summarize results in clear figures and clean tables.
As you begin planning your project, we are happy to discuss your ideas with you. We can help you pinpoint your research questions and select the appropriate endpoints and study design. In addition, we can help perform a feasibility analysis to identify potential pitfalls in the study and provide a rough estimation of the study’s required size.
Once the research question and study design have been determined, we can calculate the sample size required to answer your research question. Together with you, we develop a rough analysis plan for assessing the primary endpoint. When we perform a sample size calculation for you, we do not simply present you with a number, but also provide a detailed description of how we derived it. Finally, we draft a rough analysis plan together with you for all other analyses to be performed, including any interim analyses, and write the corresponding sections of the study protocol.
When the research project has been approved by the ethics committee (or at a later point in time before database closure), we write a detailed statistical analysis plan for all primary and secondary analyses, including descriptions of how the results will be presented. Pre-planning analyses is crucial to avoid bias.
Our expertise covers a wide range of statistical methods and machine learning tools:
After database closure, we usually perform a number of consistency checks on the data and perform any steps required to prepare the data for the intended analyses. All analyses are conducted as described in the detailed statistical analysis plan. When we analyse the data for your study, we not only provide you with results tables, but also give guidance on interpreting the results and write descriptions of the applied methods. Moreover, as co-authors of the resulting manuscript, we take responsibility for writing the statistical paragraphs of your manuscript, providing publication-ready tables and figures, and handling all reviewer comments and requests regarding statistics.
Statistical Significance is dead – let’s embrace uncertainty
A subject close to our hearts ...