A guide to performing Statistical Analysis with Jupyter Notebooks
Statistical Analysis with Jupyter Notebooks
This version of SEEK implements the usage of Jupyter notebooks in order to analyze spreadsheet data.
Step 1: Selecting cell ranges
Cell ranges can be selected via marking the cells with the mouse normally, or through input text selections.
A text selection box is available in the Spreadsheet interface, making it easier to select multiple non-adjacent groups of cells, by using Excel-like syntax and separating groups with a comma.
Examples:
- selections within the active sheet: A1:D20, A30:D40
- selections in different sheets: sheet2!A1:C80, sheet3:sheet4!F1:J80 (this is currently not useful, but later on plotting across sheet will be implemented).
Note: backward selections such as A80:C1 would work too.
Step 2: Choose your analysis from the menu
Once some data is selected, click the “Analyze” button (see below). Clicking it will open a menu of the available statistical tests and analyses which are currently available in this SEEK instance.
Step 3: Tweak specialized analysis parameters in the pop-up menu
Additional analyses
It’s fairly simple to add new analyses to “Analyze” menu by implementing the necessary (python) code and taking the necessary steps to add it into SEEK. We will provide information about that per request.
However, if you need a new analysis, please contact your SEEK administrator or project developer. For security reasons, only they can add newly implemented analyses to SEEK, after personally reviewing that these do not contain malicious code.
