[EpiData-list] Epidata Manager / Calculate

epidata-list at lists.umanitoba.ca epidata-list at lists.umanitoba.ca
Sat Jan 26 16:05:23 CST 2013


These are good questions from Bernard, as always. I have in the past put a lot of calculations into .chk files, especially before Analysis was released. Now I try to keep calculations (in classic Epidata) to a minimum and only program calculations that are useful to guide or immediately validate the entry of data. With full-time professional data entry personnel, normally  you do not want to interrupt the flow of entry with any checks other than some skips. With researcher-entered data, checks and skips are common and often useful.

Calculation of time (in hours or minutes) would not often be of use at the time of data entry; it is very easy to do at the time of data cleaning or analysis, either with Epidata's Analysis or any other software.

Bernard's question about age calculation led me to create a small test file that compared calculated age with entered age. This is easy, but we are missing on the Extended tab a comparison that the value of the current field must be equal to another field. This would be useful and exceedingly simple to implement. I would drop the word "than:" from the properties box to make the english reading of the comparison consistent.

Classic Epidata is still a good choice for many applications - especially if one has made a relational database or when Epidata is being used as a patient tracking and management tool. It is certainly stable enough and robust enough for this. I have been an advocate of using Manager/Entry Client to manage data for small studies where the main objective is to get data into Analysis or another software program for analysis because the process is so much simpler to teach. I love .chk programming, but very few of the people I teach in public health have any patience for programming per se.

Jamie

On 2013-01-26, Bernard wrote:

> - In the "Edit/Calculate", I cannot find time difference, but fractions of a day. Is this expected ? 



More information about the EpiData-list mailing list