September 2008 - EpiData-list - University of Manitoba Mailing Lists

Mailing List
by epidata-list＠lists.umanitoba.ca 04 Sep '08

04 Sep '08

Sir, Please include my email in the mailing list Olaniyan, M.A madeolaniyan(a)gmail.com

1 0

"run error" in analysis test build 145/46 fixed
by epidata-list＠lists.umanitoba.ca 03 Sep '08

03 Sep '08

The error mentioned on the list of "graph dialog and "run" has been fixed. For this reason a new build 147 is now on the test site. Usually we will not put out this many builds, but since the dialogs are a crucial part of the testing I decided to do so. Regarding the other comment : "2. Correlation coefficient “r” information needs some face lift. Can the coefficient of determination, r2 that is the measure of meaningfulness of “r “ is provided?" I can only agree. The whole linear regression and correlation part will be rewritten, but I do not expect this to happen before early 2009. Some issues in relation to translation are being worked upon. V2.1 will not be released in general before these are settled. Kind regards Jens Lauritsen

1 0

copying data btw. file levels in related entry (type comment)
by epidata-list＠lists.umanitoba.ca 03 Sep '08

03 Sep '08

Pedro wrote: "The first problem here is that the field name can not automaticaly be inserted in the projectresult.rec, such as with TYPE COMMENT. " There is a way to transfer data which could then be added as shown in the samples with EpiData Entry, - house-person-visit, where name and status of each person is transferred and shown on the screen next to the entry field. In the visit.chk file: (notice global) BEFORE FILE DEFINE varNAME _________________________________ GLOBAL DEFINE varAGE ### GLOBAL DEFINE varSEX _ GLOBAL DEFINE varILL _ GLOBAL HELP "Please enter names etc. for all persons living in \n \n @varADDRESS \n @varCITY" END in the visit.chk: BEFORE FILE define info _________________________________________________ END Before record * these come from the person file: let info = varNAME + " Age: " + varAGE + " Sex: " + varSEX + " Ill: " + varILL end VISIT BEFORE ENTRY type "@info" END END The same technique could take your name entered at an upper level and saved also in the next level. (instead of the "type" here) regards Jens Lauritsen EpiData Association

1 0

I think a found a bug
by epidata-list＠lists.umanitoba.ca 03 Sep '08

03 Sep '08

Jens, I think a found a bug. I was having a difficult time to look at a simple dataset at Stata to control patients inclusion. After several try outs and different sorts, I found that some dates were changed. That is... paciente that recruited in june had dates in july, some that were at the 25aug2008 were shown as 1aug2008. At first, I tought that was typing error. Than I checked one by one and most of it were correct at the epidata dataset but wrong in the stata dataset. Thereofre I think there is something wrong in the export tool to stata format. Later I will try other ways, and softwares of export data to see if the same dates keep coming wrong. Did anyone have any experience like this? Also I found yesterday that EpiData does not import from Stata 10, only Stata9/8. Any update soon? -- Abraço forte e que a força esteja com você, Pedro Emmanuel

1 0

Epi Analysis test release 146 (2nd Sept) - Few suggestions
by epidata-list＠lists.umanitoba.ca 03 Sep '08

03 Sep '08

Dear Jens, It is heartening to see that the development of Epidata Analysis is continuing at rapid pace. Few suggestions; 1. The graphs have been disabled. None of graph dialog box is opening. 2. Correlation coefficient r information needs some face lift. Can the coefficient of determination, r2 that is the measure of meaningfulness of r is provided? It is only a square of r, one might say, but multiple r2 calculations from one database can be cumbersome. EPI6 was providing both r and r2 with 95% confidence intervals also. 3. Epi Analysis development is hard work but you are fast approaching the release of a wonderful product that will be far better then EPI6 while preserving its full functionality. With regards, Shavinder Singh

1 0

Re: EpiData-list Digest, Vol 58, Issue 10
by epidata-list＠lists.umanitoba.ca 02 Sep '08

02 Sep '08

Dear EpiData friends, I figure out a workaround for my problem at last, the only problem now is that my partner has a preference in typing the name of the patient two or three time then to work with the solution. To remind: Im typing data from patients in which each colects 3 blood samples so my qes looks something like: identifier #### name<A > samplenumber # result # Assuming that my patients will be inserted in the data file three times each, I wanted to avoid differences in the typing of the names, such as missing letter or middle names comes in one record but not in the other. so the solution was: projectID.qes identifier #### name<A > projectID.chk identifier key unique end name key relate identifier projectresult.rec end projectresult.qes identifier #### samplenumber # result # projectresult.chk identifier key 1 Autosearch list identifier END The first problem here is that the field name can not automaticaly be inserted in the projectresult.rec, such as with TYPE COMMENT. By consequence, later I will have to merge the two files side by side to have the name of the patient or other id fileds in the results file (not very difficult but not automatic). The work around to see the name while typing is to tile the related windows horizontaly, then one can see the name and the number of the patient at the top of the screen and the data at the botton at the same time. The second problem is that autosearch does not work in the child file. It is necessary to navigate (F7 or F8) by the (few) records of each patient to reach a new empty record for that patient, and at last, for some unknown reason my partner, who is typing the data, likes better to work only with one screen. The idea of pressing F8 to find a new record and F10 each end of child record to return to the parent record did not seem to be atractive. Maybe Jens could consider a "autocomplete" check command for string fileds for a future update of EpiData entry. Best regards for all, Pedro Brasil 2008/8/30 <epidata-list-request(a)lists.umanitoba.ca> > 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: Repeat values in other records > (epidata-list(a)lists.umanitoba.ca) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sat, 30 Aug 2008 01:18:53 +0200 > From: epidata-list(a)lists.umanitoba.ca > Subject: [EpiData-list] Re: Repeat values in other records > To: epidata-list(a)lists.umanitoba.ca > Message-ID: <48B883DD.7010603(a)epidata.dk> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > "I looked for in several chk commands but didnt figure out a solution > yet. Isit possible, when the cursor is at the name field and a type > "Mary", EpiData shows all the "Mary" already in the rec file, so I could > pick one of them ....." > > Look in helpfile for EpiData Entry under "Autosearch" > AUTOSEARCH > Autosearch is a look-up function that checks the remaining data file for > records with certain data. Look-up can be defined to be on one field or > a combination of several fields. > > Let us know on the list your solution - and experiences > > Regards > Jens Lauritsen > EpiData Association > > > ------------------------------ > > ________________________________________ > EpiData-list(a)lists.umanitoba.ca > http://lists.umanitoba.ca/mailman/listinfo/epidata-list > > > End of EpiData-list Digest, Vol 58, Issue 10 > ******************************************** >

1 0

New EpiData Analysis ready for test - again
by epidata-list＠lists.umanitoba.ca 02 Sep '08

02 Sep '08

Some problems with menu dialogs showed up in b145 which made me release b146 today of EpiData Analysis. The main difference from yesterday (b145) is changes to CIPLOT dialog. But unfortunately today we have discovered a number of problems with the dialogs as well in b146: "RUN" button does not work - use "execute" instead. means dialog takes wrong variables to the command line For some graphs the x and y texts are missing. The main problem this shows is that in addition to the batch testing which we do on estimates etc. some principle of testing user interface should be added. Another issue reported is very small font size in printing. If someone have experienced this let us know - in particular if there is a solution regards Jens Lauritsen EpiData Association

1 0

input data in selected fields
by epidata-list＠lists.umanitoba.ca 02 Sep '08

02 Sep '08

i would suggest the following procedure: 1. create a data entry file with all fields and add your checks for 81-85 2. HIDE all fields except 81-85 3. JUMP to 81 after opening a record 4. AUTOSAVE after 85 hope this helps, Markus epidata-list-request(a)lists.umanitoba.ca schrieb am 25.08.2008 19:08: > 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. input data in selected fields (epidata-list(a)lists.umanitoba.ca) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 25 Aug 2008 10:47:48 -0500 > From: epidata-list(a)lists.umanitoba.ca > Subject: [EpiData-list] input data in selected fields > To: epidata-list(a)lists.umanitoba.ca > Message-ID: <200808251556.m7PFu8PX002301(a)tunku.uady.mx> > Content-Type: text/plain; charset="iso-8859-1"; format=flowed > > Dear members in the list > > I have a database with about 2000 records, each > one with 90 fields. I want to input information > that is missing in fields 81 to 85. I would like > to use an entry screen that shows me only field > 81 to 90. This is because I have a chk file that > has command that produce information in fields 86 > to 90, based on the information I input in fields 81 to 85. > Is it possible to do that to make a simplified > entry screen to use only a fraction of the fields in the data base? > I will appreciate your help > > Jose > Mexico > > > Dr. Jose Arturo Farfan Ale > Centro de Investigaciones Regionales > "Dr. Hideyo Noguchi" > Universidad Autónoma de Yucatán > Av. Itzáes por 59, num. 490 > Mérida, Yucatán, México 97000 > Tel: (999) 923-22-74 > (999) 924-57-55 > Fax: (999) 923-61-20 > > > > ------------------------------ > > ________________________________________ > EpiData-list(a)lists.umanitoba.ca > http://lists.umanitoba.ca/mailman/listinfo/epidata-list > > > End of EpiData-list Digest, Vol 58, Issue 7 > ******************************************* >

1 0

New EpiData Analysis ready for test
by epidata-list＠lists.umanitoba.ca 01 Sep '08

01 Sep '08

Just uploaded: Version 2.0.5 build 145 Release candidate for v2.1 * Extended or adapted function o lifetable :New command for life tables and Kaplan Meier plots o New syntax for compact tables: tab outcome variables /ct [/OA /O /RR] o Uniform syntax for lifetable, epicurve, compact tables, ciplot cmd outcome ..... The outcome variable is the first one o Uniform options for No Confidence Interval: /NOCI (lifetables, ciplot) o SPC charts: to exclude information on central values: /NCVI o More table headers in control of user, see: set table ct ..., set lifetable header ... * Bugs fixed o /c=xxx can now be used for value labels with string variables o percent signs cannot be used in lines with a comment * o Sorting tables now correct in dialogs and /fv tables o Table dialog for Pareto corrected to take string variables o Extra lines on graph PNG files removed (bottom and right side) o other bugs fixed See full information on the testing and analysis version files referenced on Http://www.epidata.dk/testing.php A decision has been made to include lifetables and Kaplan Meier plots in next release (v2.1). Please test this in the just uploaded build. The command accepts two different syntax forms: lifetable outcome time lifetable outcome startdate enddate We are still clarifying some aspects of translation, which necessitates postponing v2.1 release. Another issue being investigated just now is the possibility to include new SPC graph types. The main objective for v2.1 release is that after release a major rebuild will take place of structures not updated since the very first analysis parts and also allow for user defined add-on modules. regards Jens Lauritsen EpiData Association

1 0