This document is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.
In the last chapter we looked at HTML, a computer technology for producing reports, which is the very last step in a data project. In this chapter, we go back to the beginning of a data project, to the collection of the data set, and to the very important stage of getting the data set from its original analogue state into an electronic format.
Many large data sets, such as bank transactions, computer network activity, and satellite weather recordings are collected directly by machines. In those cases, the data recording is about as good as it gets. The data are immediately electronic and the data that are recorded are a faithful representation of what the recording mechanism “sees”.
However, a huge amount of data is still collected by humans via interview or written surveys and this sort of data collection requires a data entry step in order to get the information into an electronic format.
There are good reasons for using humans to collect data. For example, humans are better than computers at interviewing other human subjects, at explaining procedures and answering arbitrary questions. Humans are also better at tasks like observing animal behaviour in the field, at judging criteria such as “level of aggressiveness”, and assessing variables that are hard to define or capture precisely in a mechanical manner. But there are serious disadvantages to having humans collect data.
When it comes to recording or copying information, humans are slow and incaccurate, so there are efficiencies to be gained from ensuring that the data are recorded only once and that the data are checked as they are entered.
In this chapter, we will look at electronic forms, a computer technology for assisting data entry.
Paul Murrell
This document is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.