Key steps for choosing a proper study design are understanding the scientific aims of the study and what is required to achieve them.
In this seminar there is a brief overview of some aspects of selected study designs. It is important to understand what sources of variation exist in your data and how your study design can address it. Threats to internal validity are measurement error, selection bias, imprecise estimation, confounding, missingness, and inappropriate analyses. Some design mistakes can be irremediable.
Presented by Dr. Marianne Huebner, CSTAT and Department of Statistics and Probability
On a related note, here is a 5 minute video of the presentation “Only fools rush in! – Initial data analysis is required for developing and validating prediction models”. It does give an outline of an initial data analysis plan (needed in addition to a statistical analysis plan for the study) and potential consequences. Watch Here.