People often ask us to support them as they decide what data to collect on projects. It is essential to take time getting this right at the start of a project. Over the years, we’ve seen the problems and cost implications that occur when data requirements change during a project. So here are some tips on how to avoid this angst.
1) Focus on needs rather than wants
There are endless possibilities of data you could collect. It’s easy to get carried away with ‘nice-to-haves’ but it takes time, and money, to sift through all the data and pick out what you actually need. Instead, be really clear at the start of the project about what you are trying to achieve. Then link every piece of data you collect back to the project goals.
2) Think about who will use the data and how
Once you have an initial list of data, look at it again through the lens of ‘who’ will use the data and ‘how’ the data will be used. This will help you refine the list of data further to what is completely necessary.
3) Consider how your questions will generate accurate data
It’s amazing how easy it is to write a question and not provide an adequate answer option. You should consider the potential barriers to eliciting an accurate answer. For example:
o Is the question answerable? Make sure that every question can be answered, even if it is ‘not known’ or ‘respondent doesn’t want to answer’, otherwise your data may be distorted.
o Who are your interviewers and the interviewees? How could their demographics affect the data you receive (e.g. is the question culturally sensitive? Are language or literacy barriers at play?)
o How will your chosen format affect the data generated (e.g. if you are collecting data in remote areas, will a tablet or mobile phone battery last long enough? In some areas, perhaps a more informal paper field questionnaire would be more appropriate?)
o What kind of training do the interviewers need and how will you verify their work? We ensure that all data uploaded on our systems go through a data verification processes. For larger projects we sometimes also conduct a data audit to check that the data gathered is correct and assess whether interviewers require additional training or support.
4) Put yourself in the data analysis mode
The choice of what data to collect needs to be driven by the analysis required. It is often the case that people collect lots of data and then try to work out how to use it to show the impacts they want to measure. However, a more effective method is to start with the impacts you want to measure, then ask what you need to know in order to measure that impact and keep drilling down until you reach a piece of data that can be collected. You also need to consider what you want to measure throughout the duration of the project.
At GeoT we’ve helped organisations collect data on over 180,000 people. Feel free to contact us to discuss how we can help you with your data needs.