Workflow & Information Management

As mentioned earlier, when we introduced the concept of Open Science, your group is expected to follow proper file and data management practices. Having your files and data organized appropriately will save you time. The best rule of thumb for lab experiments is to NEVER assume that you will remember exactly what you did days, weeks or months prior… you won't, and your science (as well as your lab grade) will suffer. So being organized and meticulous is absolutely vital to the success of your project.

Considering how you are going to keep track of everything is one of the first decisions that you need to make as a group. If you don't write things down right from the moment you start developing your experimental plan, how will you be able to run the experiment the exact same way twice? For example,

  • Are you going to use paper or electronic files?
  • How will you share methods and data with each other?
  • How are you going to keep track of the changes you make over time, in case you change your plan and need to back up to an earlier version of your experimental plan?
  • Once you know the data you want to collect, how are you going to record it?
  • How should that file be organized to make it as easy as possible to accurately record the data your experiment generates?
  • How will you make sure you know which test subject/ trial each data point belongs to at all times?
  • Once your data are recorded, how will you explore them to check for mistakes, such as typos? And how exactly will you deal with such mistakes to make the data "clean"?
  • With the "clean" data in hand, were you able to implement the statistical analysis as originally planned? Or did you need to modify your analysis in any way?
  • In Lab 5 (online, asynchronous lab) you will be exploring best practices for file management and naming conventions. For now, here is an example of naming a data file for an experiment investigating how mealworm movement is affected by the presence of light.
    • 20200626_MealwormProject_Light-movement-data.csv
  • Here is an example of naming a written proposal for the same experiment:
    • 20200724_Mealworm-project_Proposal_V01.docx

As you design your experiment, and develop (and troubleshoot!) the method you expect to use, it can be very helpful to draw out your proposed methodology into a flow chart. This can help you better visualize what you want to do, and may help you realize where you need to think a little more critically about your proposed plan and how it will work.

visual representations of your methods work really well on posters! So, if you start working on it right from the start, it should be awesome by the time you're ready to build your poster!

As stated earlier, all your decisions about experimental design, analyses, and data management and cleaning should be made and documented (see below) before you collect any data, otherwise you may subconsciously (or consciously) let the data influence your actions, which will bias your conclusions.

Following these steps should ensure the experiment is repeatable by anyone that wishes to do so (including you!). Others will not get exactly the same results (because of sampling error), but they should have no trouble replicating the experiment - your documentation should be clear and thorough.