July 2011 - EpiData-list - University of Manitoba Mailing Lists

Managing two related unique IDs in data entry
by epidata-list＠lists.umanitoba.ca 11 Jul '11

11 Jul '11

Hello EpiData friends, Im involved in a reliability study. I planned to collect two blood samples at same time to measure the very same biomarker from the same patient. I would like to measure the variability of this biomarker when the lab responsible does not know what sample comes from what patient. Thus I have a (patient) ID and two samples ids. My control/recruiting qes would look like patientid Unique identifier ##### sample1 ### sample2 ### My results qes would look like sampleid Sample identifier ### sampres Sample result ####### Later, I intend to connect these two files. I generated random numbers that would be stick as blood sample stamps from 1 to 500. Thus, when a patient collects a pair of blood sample these numbered stamps are stick on the container, before they get to the lab. My problem is that, for the patient to be blind to the lab professionals, each patient should have two different sample number. Im concerned that the professional that enroll the patient might give the samples a correct pair of stamps to the blood samples but mistype this pair in the enrollment control data. FOr example: random numbers 1 106 303 2 236 168 3 302 228 4 107 197 I made a rec file with these pairs of stamps numbers like this sample1 #### sample2 #### If the enrollment professional recruits a patient with medical chart 19992 and gives him/her the third pair of stamps, this means that he should type 302 in the sample1 variable. I would like EpiData to fill or to check if the sample2 for this patients is 228. If not a warning should come to check these numbers. I did try using the chk command as sample1 comment legal use recfile.rec type comment sample2 end sample2 legal 303 168 228 197 ... end end It looks it works nicely. But I reckon if there is any more elegant way to do it. Is there? Kind regards . Abraço forte e que a força esteja com você, Dr. Pedro Emmanuel A. A. do Brasil Instituto de Pesquisa Clínica Evandro Chagas Fundação Oswaldo Cruz Rio de Janeiro - Brasil Av. Brasil 4365 Tel 55 21 3865-9648 email: pedro.brasil(a)ipec.fiocruz.br email: emmanuel.brasil(a)gmail.com ---Apoio aos softwares livres www.zotero.org - gerenciamento de referências bibliográficas. www.broffice.org ou www.libreoffice.org - textos, planilhas ou apresentações. www.epidata.dk - entrada de dados. www.r-project.org - análise de dados. www.ubuntu.com - sistema operacional

1 1

want to delete all records in epi data
by epidata-list＠lists.umanitoba.ca 11 Jul '11

11 Jul '11

Dear colleagues, I would really appreciate if you could help me out in deleting all records from rec file in epidata. I used the pack file funcationin in *the Tools * menu.but still i couldnt delete all records. see the attached error. thanks and best regards muhammad

1 2

new test version for Manager and EntryClient
by epidata-list＠lists.umanitoba.ca 06 Jul '11

06 Jul '11

Dear All Today a new set of test versions have been placed on the Http://www.epidata.dk/testing.php for general download. =============================================================== Changes to Manager: * HTML Report window stays open. * Added additional reports. * Fixed bug on MAC where valuelabels could not be deleted at all in the ValueLabel Editor. * Fixed bug on Windows where valuelabels, comparison and "write valuelabel to" definitions could be added randomly to a field. Changes to EntryClient: * Added a WorkTool bar to main form. * Fixed bug where skip records (F5/F8) would cause program crash if dataset was empty. * Fixed bug where jumping to a field would not work under certain circumstances. * Fixed long standing bug where pasting text into uppercase field was not automatically made uppercase. =========================== The bug fix since last test versions are essential to Windows users, which in the previous version could mess up use and definition of Value Labels and field definitions. Another issue was only relevant for Mac users. Please notice that files produced with the previous version might contain errors. This underlines the fact that supporting multiple operating systems effectively creates a lot more work and complications, since the same problems - now fixed - was not seen in a Linux computer. It also underlines the need for a batch testing procedure, which we are developing now, such that functionality simulating user behaviour will be implemented sometime soon. We plan to use the scripting engine to automate tests as shown below. More on this later. The simulations will be done on all supported operating systems. Once we have a few scripts developed they will be posted on the web such that users can enhance and extend the tests. regards Jens Lauritsen EpiData Association List of tasks for simulator: 1. Create user simulation and gui automation tests. 1. Create a macro/script template for benchmarking tests. 1. create datafile structure (all field types first, then a new section with all field types) 2. define value labels 3. extend definition of datafile with more "checks" from field property page 1 4. extend definition of datafile with more "checks" from field property page 2 (extended) 5. extend definition of datafile with more "checks" from field property page 3 (jumps) 6. extend definition of datafile with more "checks" from field property page 4 (calculations) 2. Import/export data between the versions of epidata (in different formats) and check import/export success. - specifically import and export of bromar, lifetable and tiramisu datasets - import as reg file or stata file (where labels are attached directly). 1. Create a scrip to automate this 1. data integrity checks: e.g. go to record no xxx and compare the actual data with the imported value. Check data type (int,float,string,Date-time) 2. Check inconsistencies between xml/rec formats (if any) and see how gui interprets the xml data

1 0

EpiData data entry competitors. Where are they?
by epidata-list＠lists.umanitoba.ca 06 Jul '11

06 Jul '11

Hello EpiData friends, Last week I did teach the 6th edition of "introduction of softwares for research porpuses" here at Rio de Janeiro. Usually we present to the students three softwares: EpiData, zotero, SPSS (although Im have been trying to replace SPSS for R-project for a while). Because of this course, yesterday, the principal of the institution I work at came to me and asked if it is worth to get engaged at this initiative: http://www.project-redcap.org/ Actually, it seems to be pretty similar to epidata projects in content, except that it is web based and it seems that it is necessary to become a partner to use it. Since then I have looked around and trade a couple of emails with some friends and some other stuff came up... http://www.rch.org.au/dmo/services.cfm?doc_id=13999 This is one is pretty good, but very expensive, and we had a tough time trying to manage the operator limitations (need to be highly skilled to operate this software) and as more projects are running more licenses are required to get the job done in time and more money is necessary. http://www.sphinxdevelopment.co.uk/about.html this one has not many information at the webpage, but a friend of mine told me that they have a software similar to teleform, which scans the form and fill the the dataset from the scanned image. Once in a while a hear someone is using pendragon: http://pendragonsoftware.com/ But this one is for portable devices such as palmOS iPad Android and stuff. Im writing all this stuff to ask you guys if you have any bad/good experience with other dataentry softwares? This time particularly with redcap. Every time someone asks me about other dataentry software, I take some time trying it and I always come back to EpiData. I believe that there isnt really a good competitor to EpiData so far... free, small, portable, easy to use, and does the job. Also, I imagine that looking at what other data entry softwares do, may give some insight to improve EpiData functionality. This web idea may be something to think about. Kind regards, Abraço forte e que a força esteja com você, Dr. Pedro Emmanuel A. A. do Brasil Instituto de Pesquisa Clínica Evandro Chagas Fundação Oswaldo Cruz Rio de Janeiro - Brasil Av. Brasil 4365 Tel 55 21 3865-9648 email: pedro.brasil(a)ipec.fiocruz.br email: emmanuel.brasil(a)gmail.com ---Apoio aos softwares livres www.zotero.org - gerenciamento de referências bibliográficas. www.broffice.org ou www.openoffice.org - textos, planilhas ou apresentações. www.epidata.dk - entrada de dados. www.r-project.org - análise de dados. www.ubuntu.com - sistema operacional

1 6

Corruption error
by epidata-list＠lists.umanitoba.ca 04 Jul '11

04 Jul '11

Hello. I am working with the new EpiData Manager, and every now and then when I click within a field I receive the following corruption error: "epidatamanager.0.8.2.0.i386.win32 Press OK to ignore and risk data corruption. Press Cancel to kill the program" I am forced to kill the program, as clicking OK blocks me from making edits because the same message pops up. Does anyone have any suggestions or fixes to handle this issue? Many thanks. ------------------------------------------------------------------------------------------------------- Jonathan Edwin Public Health Agency of Canada

1 0

a Short status on development (Manager and EntryClient) - what to expect the coming month
by epidata-list＠lists.umanitoba.ca 01 Jul '11

01 Jul '11

Dear All Just now development has focus on removing bugs and clarifying issues, which prevent release as first public v1.0 In recent days a number of important issues have solved, such as: a. Pasting into upper case fields (fixed) b. Strange behaviour sometimes with jumps (fixed) (resulted in jumping to a wrong field) The plan is to release the next test versions towards the end of next week, just before we close for vacation until August 1st. regards Jens Lauritsen EpiData Association

1 0