Dear Epi 218 Students,
One of you recently asked how to parse REDCap field values into multiple new fields. This would be like splitting a date field into separate fields for day, month, and year; taking the dash out of phone number; or separating a single name field, "Lastname, Firstname", into two fields, "Lastname" and "FirstName".
The answer is that you can't do this in REDCap.
The same student has two REDCap projects for the same study. One project has a single record per study subject and one has multiple records per study subject. (It is often necessary or at least convenient to create two separate REDCap projects like this.) The student asked how to join the data between these two projects for purposes of querying, creating derived variables, monitoring for completeness, etc.
The answer is that you can't do this in REDCap.
A different student asked how to create a report that counts study subjects by sex, race, and ethnicity. Periodic reports like this are required by the NIH, so we call this an NIH Report. The student wanted to be able to create the report in REDCap so it could be run the way we ran REDCap reports in Lab 3.
The answer is that you can't do this in REDCap.
REDCap is great for quickly building web-based data collection instruments, but it is not a good system for data storage and management. You should collect your data in REDCap and transfer it to your production data repository, which should allow parsing of field values, derivation of new fields, creation of one-to-many relationships, enforcement of referential integrity, querying, and reporting. We recommend Microsoft Access. For your final project’s data management plan, we will (reluctantly) accept a set of Stata “.dta” tables or data tables for another statistical package. For the reasons discussed on the first day of class, one or more Excel files is NOT an acceptable data repository.
Best,
Michael