Data Hangover Part II: Going Retro

In the previous post, I covered some basic steps for preventing Data Hangover, however all of my suggestions pertained to actions you can take prior to data collection. What if the project is already well underway?  How does one conduct retroactive data management? This is a much more complicated and tricky question, and the answer will vary with things like

  • type(s) of data and their format(s)
  • number of collaborators/contributors
  • current organization and management system
  • ideal organization and management system
  • resources (personnel and financial)

Although listed last, Resources are going to be the most accurate predictor of how much retroactive data management you can accomplish.  Personnel resources can better be interpreted as time.  Data management in general is quite time consuming, and it is certainly difficult to correct mistakes after a project is well underway.  Time is therefore likely to be the biggest hurdle to retroactive data management.

So where should you start? My best suggestion is to pretend that you can travel back in time and create a data management plan.  Carefully consider things like: What is the ideal organization scheme for file structure? Where should you archive the data? What metadata standards should you use? Build your data dictionary and describe what codes, units, ranges etc. apply to the data.

deLorean from BTTF

Grab Michael J. Fox and get those datasets cleaned up! Photo from Flickr by F1RSTBORN

Use this data management plan to get a little bit closer to your ideal data situation.  What steps can you take today? Tomorrow? Come up with a timeline and plan for working on your data management.  If there are multiple people involved in the project, assign specific tasks related to re-organization and standardization.

If your personnel and/or time are limited, there are a few options:

  1. Consider asking for funds to facilitate data management. This is especially important if you are attempting to make sense of a large project spanning multiple years with an abundance of collaborators and data types.  There are sometimes calls for proposals specifically for these types of “small-scale” data projects.
  2. Hire someone. This person might be from a local information school (although why limit it to local?). They may also be a graduate student in need of funding or a technician you can hire for a couple of months.
  3. Break down the gigantic task into small, manageable chunks.  Often, data management is tedious and repetitive.  I loved having projects like this on my desk while I was a graduate student- they were great for breaking up the more mentally challenging tasks, and provided a mental break while allowing me to still be productive.  Consider replacing your YouTube time with data management time.

Although I don’t envy you the task of retroactive data management, you can be sure that the sense of satisfaction you will receive from well-managed data will make all of your efforts worthwhile.

Tagged , ,

One thought on “Data Hangover Part II: Going Retro

  1. […] I also knew I needed some organizational tips of my own going into our initial meeting. Although I was taught about planning for data management from the beginning of an experiment, we never really covered managing the vast amounts of data that have previously been collected. Carly Strasser offers some key tips in Part 2 of the Data Hangover Series (“Going Retro”). […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: