Should process or tools define research data collection?

Data collection is a crucial part of clinical research. Clinician scientists carefully design chart review forms (CRFs) to capture various types of information. There are multiple tools available for data entry including MS Excel, MS Access, OpenOffice, FileMaker, OpenClinica and RedCap.

The question I would like to address in this post is 'Do you define the data collection process and then hunt for a tool or obtain a tool with capabilities and develop processes around it?' 

Merriam-webster defines process as a series of actions that produce something or that lead to a particular result and tools as something that is used to do a job or activity. Processes and tools go hand in hand, so which comes first? Let us consider both cases.

• Tools First: Researchers tweak their process based on the tool on hand. For example, a research fellow used MS Excel to capture birth weight data. Naturally, the he/she created a number column.
• Process First: Researchers establish their process and configured the tool to suit the process. Using the same example as above. The research fellow created a process for capturing birth weight data, which includes rounding to 2-decimal places and entering birth weight in pounds (lbs).

As data collection progress, the research fellow discovers that birth weight in patient medical charts is recorded in various units (pounds, grams, kilograms).

• In the first scenario, the research fellow would tweak his/her process based on MS Excel capacity to capture this type of data and decided to convert the column to text. The research fellow successfully collected and entered the data but found that the birth weight data in its current text format is not usable in the analysis phase.
• In the second scenario, the research fellow would configure MS Excel to capture the birth weight data as indicated in the established process. He/she created two columns - one is in number format with 2-decimal places and the other is in text with drop-down list (lbs, grams, kg). The research fellow successfully collected and entered the data. More importantly, he/she was able to easily convert all birth weight data into lbs and use it for analysis.

Taking the time to establish the process will save you valuable time. Tools are very important but it does not define the data types you need to answer your research question.

"Perfection has to do with the end product but excellence has to do with the process" - Jerry Moran.