Dear all
Since release of last trial version of the new EpiData Manager in
January around 1000 copies have been downloaded from the testing.php
website at epidata.dk.
Since then we (Torsten Christiansen and I) have worked on ways of
implementing the more subtle parts of the old rec file system in the new
format, which is described at
http://www.epidata.org/dokuwiki/doku.php/documentation:datafileformat:xml_v0
The description is not completely up to date, but will be updated soon.
The challenge was to also include relate, encryption and principles for
translation of data forms in a structured way. This has now been
completed and I expect a new release for testing within some weeks.
In the final file format all contents will be contained in one physical
file formulated in XML and in compliance with the w3c.org validator for
an xml file. The basic blocks implemented will be:
- user access control block
- a list of control structures, e.g. specification of relationship btw.
files
- a list of tables contained (each like an old rec file)
- in each table a number of sections
- in each section a number of fields.
- each field having a number of attributes, such as type (integer,
string etc), controls etc.
The file will contain the combination of all current rec and chk files
for a given project. User control will be at the section level.
Many users will now think "Oh-no" they changed the simplicity principle
of REC-CHK-QES and created a monster system with many problems. This is
however NOT the case. For a simple project the file content structure
is of no interest to the user and will appear simple. But by containing
all in one physcal file it should become much easier to keep track of
the data for a given project. Basicly with the new structure files can
also be edited in a standard editor.
The complete rewriting is expected to be completed in about one year
from now, and funding is still welcomed for the process.
regards Jens Lauritsen
EpiData Association
Hi!
When trying to create a Codebook I get an access violation error. I have tried to carefully revise the check file with no result. The error seems to be related to only fields named with the pattern "xxxcom" which all are string variables.
-If I choose all fields before "xxxcom" the Codebook for those variables will be created without error.
-If I only check "xxxcom" --> access violation
-If I check variables before and after "xxxcom" --> access violation
-If I check only variables after "xxxcom" -->no error
Error message:
"Access violation at address 00403347 in module 'EpiData.exe' Read of address 393938FB"
Epidata 3.1 (270108)
Windows XP Pro
Thanks for help
Jonas Malmstedt, MD
Department of Vascular Surgery, N1:06
Karolinska University Hospital
SE-171 76 Stockholm
PS I have also noticed another problem: smtimes EpiData Will not read *.not and *.rec files. I get an error telling me that the file name are not valid or the file is used. When I try to open the same file not NotePad - no problem! DS
When you prepare the rec file for double entry better to leave the box
"Match by record" unchecked. However if the double entry appears to slow or
cause some frustration, perhaps it might be better to use copy structure and
enter the data for comparison.
Tieble
On 29 April 2010 18:06, <epidata-list-request(a)lists.umanitoba.ca> wrote:
> Send EpiData-list mailing list submissions to
> epidata-list(a)lists.umanitoba.ca
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.umanitoba.ca/mailman/listinfo/epidata-list
> or, via email, send a message with subject or body 'help' to
> epidata-list-request(a)lists.umanitoba.ca
>
> You can reach the person managing the list at
> epidata-list-owner(a)lists.umanitoba.ca
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of EpiData-list digest..."
>
>
> EpiData-list mailing list
> ___________________________________
>
> Today's Topics:
>
> 1. RE: Double Entry for Related Files
> (epidata-list(a)lists.umanitoba.ca)
> 2. Access violation error with Codebook command in EpiData
> (second try) (epidata-list(a)lists.umanitoba.ca)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 28 Apr 2010 13:48:41 -0500
> From: epidata-list(a)lists.umanitoba.ca
> Subject: [EpiData-list] RE: Double Entry for Related Files
> To: <epidata-list(a)lists.umanitoba.ca>
> Message-ID:
> <679902537E215D43805DEA9BE833090E042A5B80(a)NORCEX1.norc.org>
> Content-Type: text/plain; charset="us-ascii"
>
> I understand the two different methods of double-entering data, and I do
> indeed want to "prepare for double-entry". The problem is that when I
> go to prepare the questionnaire for double entry, I can only choose to
> prepare either the parent OR the child file. If I choose the parent
> file, then a double-entry file is only created for the parent file -
> when I open it to double enter the data, it relates to the original
> child file, so it opens the parent double entry file along with the
> original child file. Since only one of the two files is a double entry
> file, the program thinks I am entering the data for the first time, and
> does not compare the alues entered to the values from the original file.
>
>
>
>
> I cannot seem to find a way to prepare both the parent and the child
> file for double entry together. If I prepare them for double entry
> separately, I run into similar problems.
>
>
>
> Sam
>
>
>
>
>
> ----------------------------------------------------------------------
>
>
>
> Message: 1
>
> Date: Tue, 27 Apr 2010 19:13:58 +0200
>
> From: epidata-list(a)lists.umanitoba.ca
>
> Subject: Re: [EpiData-list] Double Entry for Related Files
>
> To: epidata-list(a)lists.umanitoba.ca
>
> Message-ID: <4BD71B56.7020800(a)tbrieder.org>
>
> Content-Type: text/plain; charset=UTF-8; format=flowed
>
>
>
> Hi Sam:
>
>
>
> There are two approaches to double-enter: 1) independently, 2) the
> second file depending on the first. You seemingly chose the latter
> ("prepare for double-entry"). If you choose that option, you must enter
> the records in the same sequence (I believe) and the value of each field
> in the second record will be compared to the corresponding one in the
> first and a discordance reported, so you can verify. Thus, no need to
> enter twice.
>
>
>
> Alternatively, you enter your related file pairs independently between
> first and second entry. You must have a unique identifier in both the
> parent and the child file. The sequence then doesn't matter of course.
>
>
> After you have entered parent1-child1 and parent2-child1, you validate
>
> parent1-vs-parent2 on the unique identifier, then child1-vs-child2 on
> that identifier.
>
>
>
> Hans
>
>
>
> On 20:59, epidata-list(a)lists.umanitoba.ca wrote:
>
> > Hi,
>
> >
>
> > I have a questionnaire that uses two related .rec files for data
> entry.
>
> > I want to double enter the data by checking for discrepancies as the
>
> > data is entered the second time. Is it possible to do this for
>
> > questionnaires that span two .rec files? When I try to prepare for
>
> > double entry, the program creates a "file1_dbl.rec" file, but relates
>
> > it to the "file2.rec" file rather than creating a "file2_dbl.rec" file
>
>
> > to compare against.
>
> >
>
> > I am thinking that I will just have to un-relate the file, and enter
>
> > and double-enter the survey in two parts, but any help would be
> appreciated.
>
> >
>
> > Thanks,
>
> >
>
> > Sam
>
> >
>
>
>
> --
>
> Hans L Rieder, MD, MPH
>
> Jetzikofenstr 12
>
> 3038 Kirchlindach
>
> Switzerland
>
>
>
> Tel: +41 31 829 4577
>
> Mob: +41 79 321 9122
>
> Web: http://www.tbrieder.org
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> --
>
> Samuel Haddaway
>
> Research Assistant, International Projects
>
> National Opinion Research Center
>
> haddaway-sam(a)norc.org
>
> Phone: 301.634.9423
>
>
>
>
>
> ------------------------------
>
> Message: 2
> Date: Wed, 28 Apr 2010 21:50:13 +0200
> From: epidata-list(a)lists.umanitoba.ca
> Subject: [EpiData-list] Access violation error with Codebook command
> in EpiData (second try)
> To: epidata-list(a)lists.umanitoba.ca
> Message-ID:
> <OF120C2D35.551C97A0-ONC1257713.006CF7D1-C1257713.006CF7D4(a)sll.se>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Hi!
> When trying to create a Codebook I get an access violation error. I have
> tried to carefully revise the check file with no result. The error seems to
> be related to only fields named with the pattern "xxxcom" which all are
> string variables.
> -If I choose all fields before "xxxcom" the Codebook for those variables
> will be created without error.
> -If I only check "xxxcom" --> access violation
> -If I check variables before and after "xxxcom" --> access violation
> -If I check only variables after "xxxcom" -->no error
>
> Error message:
> "Access violation at address 00403347 in module 'EpiData.exe' Read of
> address 393938FB"
>
> Epidata 3.1 (270108)
> Windows XP Pro
>
> Thanks for help
>
> Jonas Malmstedt, MD
> Department of Vascular Surgery, N1:06
> Karolinska University Hospital
> SE-171 76 Stockholm
>
> PS I have also noticed another problem: smtimes EpiData Will not read *.not
> and *.rec files. I get an error telling me that the file name are not valid
> or the file is used. When I try to open the same file not NotePad - no
> problem! DS
>
> ------------------------------
>
> ________________________________________
> EpiData-list(a)lists.umanitoba.ca
> http://lists.umanitoba.ca/mailman/listinfo/epidata-list
>
>
> End of EpiData-list Digest, Vol 78, Issue 17
> ********************************************
>
Sam,
I thought about you related file problem and perhaps there is a workaround.
I thougth for example of edaverse event notification of medication use. Each
patient is using one or more medications and while using these may have two
or more events simultaneously such as rash nausea and vomiting. Each event
will be a record in a related file. But the unique identifier is the patient
record and may cause trouble in double entry.
Perhaps if each related record may be a combination, you me be able to run
double entry independently.
Example:
Main qes file ---
ID #####
name ___
sex___
medication ___
related file ----
ID #####
event __
eventID ________
description _____
Related chk file
eventID
before entry
let eventID = ID + event
end
end
If patient 123 has three events
then eventID would be 123a 123b and 123c and may uniquely identify the
events and double entry may be possible by this eventID.
I did not test it although.
But this means you would have to type all again if you already typed it all.
Or perhaps exporting editing and importing may make this easier.
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
I understand the two different methods of double-entering data, and I do
indeed want to "prepare for double-entry". The problem is that when I
go to prepare the questionnaire for double entry, I can only choose to
prepare either the parent OR the child file. If I choose the parent
file, then a double-entry file is only created for the parent file -
when I open it to double enter the data, it relates to the original
child file, so it opens the parent double entry file along with the
original child file. Since only one of the two files is a double entry
file, the program thinks I am entering the data for the first time, and
does not compare the alues entered to the values from the original file.
I cannot seem to find a way to prepare both the parent and the child
file for double entry together. If I prepare them for double entry
separately, I run into similar problems.
Sam
----------------------------------------------------------------------
Message: 1
Date: Tue, 27 Apr 2010 19:13:58 +0200
From: epidata-list(a)lists.umanitoba.ca
Subject: Re: [EpiData-list] Double Entry for Related Files
To: epidata-list(a)lists.umanitoba.ca
Message-ID: <4BD71B56.7020800(a)tbrieder.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Hi Sam:
There are two approaches to double-enter: 1) independently, 2) the
second file depending on the first. You seemingly chose the latter
("prepare for double-entry"). If you choose that option, you must enter
the records in the same sequence (I believe) and the value of each field
in the second record will be compared to the corresponding one in the
first and a discordance reported, so you can verify. Thus, no need to
enter twice.
Alternatively, you enter your related file pairs independently between
first and second entry. You must have a unique identifier in both the
parent and the child file. The sequence then doesn't matter of course.
After you have entered parent1-child1 and parent2-child1, you validate
parent1-vs-parent2 on the unique identifier, then child1-vs-child2 on
that identifier.
Hans
On 20:59, epidata-list(a)lists.umanitoba.ca wrote:
> Hi,
>
> I have a questionnaire that uses two related .rec files for data
entry.
> I want to double enter the data by checking for discrepancies as the
> data is entered the second time. Is it possible to do this for
> questionnaires that span two .rec files? When I try to prepare for
> double entry, the program creates a "file1_dbl.rec" file, but relates
> it to the "file2.rec" file rather than creating a "file2_dbl.rec" file
> to compare against.
>
> I am thinking that I will just have to un-relate the file, and enter
> and double-enter the survey in two parts, but any help would be
appreciated.
>
> Thanks,
>
> Sam
>
--
Hans L Rieder, MD, MPH
Jetzikofenstr 12
3038 Kirchlindach
Switzerland
Tel: +41 31 829 4577
Mob: +41 79 321 9122
Web: http://www.tbrieder.org
--
Samuel Haddaway
Research Assistant, International Projects
National Opinion Research Center
haddaway-sam(a)norc.org
Phone: 301.634.9423
Hello epidata friends,
Im occasionally reading emails in the list about limitations in the related
rec files. The last I remember was about double entry, which I did not read
a proposition of solution. Here goes another problems that I could not solve
yet.
Im working with a dataset entry with two purposes: 1) to estimate a
trajectory of a particular liver enzyme over time while patients is under
therapy and 2) estimate the risk of liver toxicity while in therapy. The tow
objectives are time dependent.
In the first objective the possible result is a graph with a line conecting
dots which represents the enzyme values in each time point.
The second is a cox model or a exponential survival model which the outcome
is liver toxicity due to therapy. In this one, the medications may change
over time.
Therefore I was imagining creating a main recfile with clinical
characteristics of each patient that are permanent, that is, do not change
over time such as HIV serology status, and create tow related recfiles with
1) data about each liver panel with correspondent date and 2) the medication
in use with initial date and end date with the end status (cure, default,
death etc); both may be fed with several data from different dates for each
patient.
Now ... one common problem with dates is that very often typing mistakes
make dates incompatible such as:
1) the second liver panel was collected BEFORE the first.
2) The end of medication was registered BEFORE the beginning or the second
medication started BEFORE the end of the first.
To control this sort of mistake is easy with conditional checking when the
dates are in the same recfile. But if I make related files the conditional
checking with dates should be done in the same field (or different fields)
in different records. That is ... I should ask epidata to go in other record
with the same key (ID) and check if the collection date is before or after
the date of this open record.
The question is ... is this possible? Does anyone have a sugestion to
workaround (possibly having all the dates in the main recfile and related
file)?
An example:
******** Main qesfile
ID ####
name _____
hiv #
start <dd/mm/yyyy>
end <dd/mm/yyyy>
toxicity #
******* First related file - liver panel
ID ####
coldate <dd/mm/yyyy>
panelnum #
alt ####
ast ####
****** Secodn related file - medication
ID ####
TBdrugs _______
HIVdrugs _______
combstart <dd/mm/yyyy>
combend <dd/mm/yyyy>
outcome #
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
Hi,
I have a questionnaire that uses two related .rec files for data entry.
I want to double enter the data by checking for discrepancies as the
data is entered the second time. Is it possible to do this for
questionnaires that span two .rec files? When I try to prepare for
double entry, the program creates a "file1_dbl.rec" file, but relates it
to the "file2.rec" file rather than creating a "file2_dbl.rec" file to
compare against.
I am thinking that I will just have to un-relate the file, and enter and
double-enter the survey in two parts, but any help would be appreciated.
Thanks,
Sam
--
Samuel Haddaway
Research Assistant, International Projects
National Opinion Research Center
haddaway-sam(a)norc.org
Phone: 301.634.9423
Hello Epidata Friends,
I know this will sound repetitive and passed over but Im not finding a
workaround. I have a very ancient rec file created by EPI6, this dataset was
abandoned in 1999. My intention was to update the rec and chk file for new
data to come in and update the up to 2009. I was trying to export using
epidata entry tool to Stata and do several editing within Stata and then
import it again from Stata to epidata to keep on feeding the dataset.
My problem is that the dates of birth and follow-up start are configured as
dd/mm/yy , that is Year with tow digits. When I open the data from within
Stata, weird dates come. Just the years look weird such as 01apr2020. All
the years are 2020 or 2019 and I know the yars may range from 1986 to 1999.
It seems that is not only a problem to add the 19.. before the dates year so
far.
So I guess, since Epidata does not accept dates with tow year digits, it is
exporting with error. Therefore Im accepting any suggestions to workaround.
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