November 5, 2018 – For participants in the Golden Retriever Lifetime Study, it’s both a sense of accomplishment and relief when they complete their lengthy annual questionnaire, comprising more than 200 items covering everything from diet to household exposures to activity, and lots more.
Once the questionnaire is complete, data is housed and backed up with an off-site partner. Scientists at the Foundation receive data on a preset schedule and process it into a variety of forms.
Although some of the data is useable exactly as it arrives, many data points need to be processed before they can be analyzed. Two ways the study research team processes data are normalization and cleaning.
Normalization is the process of standardizing data. For example, the questionnaire asks people to measure how much food they give their dog each day. Some participants answer this question in cups and others in ounces. This data needs to be converted to a standard unit of measurement so that a researcher can use it.
Examples of data cleaning are correcting misspellings or unexpected answers, such as a respondent inadvertently indicating that a male dog had a litter of puppies! These inconsistencies need to be resolved before the data is ready for analysis.
In addition to processing, the data needs to be packaged. Some data might be put into spreadsheets for monthly reports sent to the study’s oversight committee. Some data is put into statistical programs for further analysis and publication.
What is the future of this data? Ultimately, it will be available as a dynamic database for internal and external researchers to use as a tool that will help them better understand factors influencing the health of our dogs. This means getting results out faster to veterinarians and pet owners, fulfilling the mission of the study and the Foundation – bridging science and resources to advance the health of animals.