- EpiData-list - University of Manitoba Mailing Lists

How to Delete Records Based on a Variable
by epidata-list＠lists.umanitoba.ca 22 Jul '11

22 Jul '11

Hello, I am trying to delete a subset of records in my dataset based on the value of one of the variables (hhid) in either EpiData Entry or Analysis. I am guessing it is easy, but I cannot figure it out. Can someone please advise? Thanks, Sam

1 1

permission
by epidata-list＠lists.umanitoba.ca 21 Jul '11

21 Jul '11

Dear the Epidata community Please allow me to post on your list Regards, -- Lê Phương Loan Project Assistant - ACT 2 0167 28 55 567

1 1

New test version for EpiData Manager, EntryClient and EpidataCMD
by epidata-list＠lists.umanitoba.ca 15 Jul '11

15 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. The development during most recent weeks has focused on solving all reported bugs and fixing issues related to keyboard shortcuts across operating systems. Keyboard short cuts will be operating system typical (or should be) from now on. Bugs reported recently are also fixed. The list of major changes in the versions release today: =============================================================== Changes to Manager: - bug fix: * creating sections on MAC was nearly impossible, and difficult at times on Windows. * Assigning jumps to a UPPERCASE string field would cause the program to crash. * Typing an invalid date/time in range, would not always give a warning. On MAC this would lead to a large black screen, show for a few seconds. Changes to EntryClient: * Fixed bug where closing at times could save a "New Record" without data. * Major changes in short-cuts on MAC. * Performing "New Record" on a "new record" will clear content (but ask user first). =============================================================== Kind regards, Torsten Bonde Christiansen. Jens Lauritsen EpiData Association.

1 0

New test version for EpiData Manager, EntryClient and EpidataCMD
by epidata-list＠lists.umanitoba.ca 15 Jul '11

15 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. The development during the last week have mainly been focus on solving some large outstanding bugs in the programs. This mean that today release does not have any new feature, but will hopefully add to the feeling that we rapidly approaching the first stable release (Version 1) of EpiData Manager and EpiData EntryClient. With todays release we have also tried to address the problem of how different Operating Systems use different sets of keyboard short-cuts. Our initial goal was to have the programs use the same set of short-cuts across all OS's, but we have found this to be virtually impossible. Instead we have now focused on have a small set of common short-cuts, but try to use the "normal" short-cut for regular functions (such as search, copy & paste, etc.) Hopefully this will in the future lead to a point where we integrate a list of command the programs are capable of and let users customize the associated shortcuts as they wish. The list of major changes in the versions release today: =============================================================== Changes to Manager: * Fixed the longest outstanding bug: Creating sections on MAC was nearly impossible, and difficult at times on Windows. * Assigning jumps to a UPPERCASE string field would cause the program to crash. * Typing an invalid date/time in range, would not always give a warning. On MAC this would lead to a large black screen, show for a few seconds. Changes to EntryClient: * Fixed bug where closing EntryClient at times could same the "New Record". * Major changes in short-cuts on MAC. * Performing "New Record" on a "new record" will clear content (but ask user first). =============================================================== Kind regards, Torsten Bonde Christiansen. EpiData Association.

1 0

Re: Managing two related unique IDs in data entry
by epidata-list＠lists.umanitoba.ca 12 Jul '11

12 Jul '11

> > Jamie, Unfortunately this solution is not working here. Did you tests it? Did it work with you? The problem is that like this EpiData only accepts sample2 in blank, ie, missing data for this field. Also, when the warning comes up, the (@sample1, @sample2) does not show any data. My guess is that EpiData is not creating the checksamp variable correctly or is being able to type sample1 comment correctly into sample2. Another good guess, might be that the error is above the chair in front of the keyboard. Thanks anyway for the ideia. I will try to make this warning work later and let you know. > , > > before file > define checksamp ### > hide checksamp > end > > sample1 > comment legal use recfile.rec > type comment checksamp > end > > sample2 > after entry > if sample2 <> checksamp then > help "This pair of sample numbers (@sample1, @sample2) do not match > those assigned. \nPlease check and re-enter both." type=error > goto sample1 > endif > end > end > > Jamie > > 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

Troubleshooting skip pattern crash
by epidata-list＠lists.umanitoba.ca 12 Jul '11

12 Jul '11

Hello, I recently posted in regards to an error/crash message I was receiving with entering skip patterns. These errors were in the previous Manager version and I have since upgraded to the most recent upgrade. Unfortunately the trouble has persisted. Is anyone else experiencing this issue? Suggestions for getting around it? The epx file is a hypothesis-generating questionnaire, therefore it is relatively large. In addition, the skip patterns are being generated before value labels have been given...is it possible that one of these two factors is contributing to the error? Any help would be appreciated. Regards, Jonathan Edwin -------------------------------------------------------------------------------------------------------- Outbreak Management Division Public Health Agency of Canada Unit 120, 255 Woodlawn Road West Guelph, Ontario N1H 8J1 jonathan.edwin(a)phac-aspc.gc.ca

1 0

Need help: Check codes truncated
by epidata-list＠lists.umanitoba.ca 11 Jul '11

11 Jul '11

Dear colleagues, I would apprciate your help in solving check codes truncation proble. I am currently creating one database using epidata. I have one of the check codes of about 600 charecters long(example below), however after applying chk codes, it is automatically truncated to 207 charecters when I open to see next time. Any clues or ways to code alternatively? *check code:* IF (Sex=2) AND (Msex=0) OR (Msex=9) OR (Msex=99) AND (Fsex=0) OR(Fsex=9) (Fsex=99) AND (IDU=0) OR (IDU=9) OR (IDU=99) AND (CFac=0) OR (CFac=9) OR (CFac=99) AND(Trans=0) OR (Trans=9) OR (Trans=99) OR (coag=0) OR (coag=9) OR (coag=99) OR(HetSexEND=0) OR (HetSexEND=9) OR (HetSexEND=99) OR (ConSusHIV=0) OR (ConSusHIV=9) OR(ConSusHIV=99) AND (OccuExp=0) OR (OccuExp=9) OR (OccuExp=99) AND (OthMedExp=1) OR(OthMedExp=0) OR (OthMedExp=9) OR (OthMedExp=99) AND (NMNOcc=1) OR (NMNOcc=0) OR(NMNOcc=9) OR (NMNOcc=99) *THEN* * **Risk="OtherMedExp-10"* Thank you. Bijay Adhikari, PhD CD Epiodemiologist Saskatchewean Ministry of Health

1 2

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