Google Refine is a great free data cleaning tool that you can use to prepare your collected data for analysis with R and R Commander. Anyone who has worked with survey data input by human beings (so anyone who has worked with survey data) has encountered the fact that not everyone has the same understanding of how data should be input and that everyone occasionally suffers from fat finger syndrome (FFS). Annual incomes may be entered as $30,000, $30k, or 30000, and computer thinks that these are different values. Monterey could end up being Moneterey, monterey, monterey ca, Monterrey, or any number of other versions.
Rather than combing through hundreds or thousands of responses to ensure that each is identical, Google Refine groups responses that it identifies as being similar and allows you to merge them into a single, consistent response. Rather than potentially producing 5-10 distinct responses, you can be certain that every “Monterey” is counted as “Monterey.” This workshop will teach you the basics of uploading uncleaned data into Google Refine, using “facets” to group and combine responses, applying filters to more quickly find certain kinds of data, and exporting data that is ready for analysis in R.
The worksop will be held on Thursday November 8th from 12:30 to 1:30 at the Digital Learning Commons (420 Calle Principal). Please come prepared to learn, with Google Chrome (go.miis.edu/chrome) and Google Refine (go.miis.edu/refine) installed on your computer and your data readily available to work with. See you there!