[EpiData-list] Importing data to Epidata (v.

EpiData development and support epidata-list at lists.umanitoba.ca
Tue Jun 23 03:41:31 CDT 2015


I am a newcomer to Epidata, so I apologize if my questions are not
sophisticated. I am exploring the latest version and it is fantastic with
many ingenious features, great work and great programming philosophy, thank

As the version is new, I understand that the documentation is not yet fully
developed, and hence if someone can help me with these questions, I think
they can be of use to many others as well.

What I want to do is to import a spreadsheet with field labels and data to
Epidata. The spreasheet contains field names on the top row and records on
the following rows. Once I have imported the initial data, I would like to
ammend the records with additional fields, i.e. have the initial
spreadsheet data as the seed data in records. I think this is a very normal
practice to do, but I have run into several difficulties:

1. I have tried to import the spreadsheet with "File/Import file" but I
don't seem to get it working. I have turned the spreasheet from Excel to
CSV, but the dialogue box that opens doesn't recognize data. I made a
workaround by copying a part of the spreadsheet and used "File/Import from
clipboard" and that did work. The downside is that you have to do it part
by part, and the fundamental problem is why it doesn't work in the first
place. Is Epidata Manager very picky as to what field separators are used
in CSV? I have tried changing field separators but with no luck. Clipboard
seems to be the only workaround, but even still I have to do it in parts.
(I am using Ubuntu and the latest Linux 64-bit version)

2. With the copy-paste method I noticed that Epidata was smartly able to
interpret some fields with the right type (e.g. date, but not all
date-fields, but I believe this is the issue of my spreadsheet), but not
all. I have not discovered a way to change the field type afterwards, is it
at all possible? The only hack I figured out was to manually change the
field type in XML code directly, but I would be very hesitant to do it. If
I touch XML code directly, will that mess up everything, or is it the
simple fix to the problem?

3. Value-labels: My spreadsheet contains no value labels, obviously. It has
for example field "Country of origin" and the value is string "Nepal". Is
there a way for me to easily convert strings of text into value labels as a
part of the import process? String values are consistent (the spreadsheet
is an output file from a database), so I should imagine it would be
possible do so.

4. Provided I get questions 1,2 and 3 adequately resolved, I should be fine
with the initial data import. But what do I do if I want to later import
additional records from the same spreadsheet structure to my Epidata
project? Is this something that Append is supposed to do? There I have the
same issue with the spreadsheet compatibility: The CSV file does not open
in a recognisable way, hence I haven't been able to play around and test it

Help is much appreciated, thank you!


More information about the EpiData-list mailing list