Similar to the principle related to assumptions is the one related to data treatment. It is really uncommon to employ a dataset exactly as it comes out of the box. You always have to perform data cleaning and validation.

 Quite common results of this activity are exclusions of records or substitutions of missing data. Both of them are examples of data treatment to be communicated within your report. Why?

First of all, because this will increase the level of reproducibility of your work; that is, it will make it clear how to reperform the analysis you conducted. This will provide assurance on the quality of your analysis and make your stakeholders more confident at drawing conclusions from it.

Moreover, documenting these treatments and their rationales will be of great help for you as well in the eventuality of a future need to reperform the same analysis with updated data.