February 2007 - EpiData-list - University of Manitoba Mailing Lists

EpiData-List: List index out of bounds
by epidata-list＠lists.umanitoba.ca 15 Feb '07

15 Feb '07

When I try to open updated REC file, I get the message "List index out of out of bounds (420)" How can I solve this problem? Thank you.

1 0

BUG REPORT : minus sign for MISSINGVALUE with more
by epidata-list＠lists.umanitoba.ca 15 Feb '07

15 Feb '07

It is not a bug, DATE is a reserved name, you cannot use it as a field name. Furthermore, there is nothing like "## ## ###" for a field definition, the definition for dates is <dd/mm/yyyy> (if you choose European date format) Regards, Hans Hello, I noticed a strang bug in Epidata Entry when I use the '-' sign to input=20 a MISSINGVALUE in a field. I created three fields on one line to enter a date with incomplete value = : Date ## ## #### Age ## Sexe # In the check file I used the following DATE MISSINGVALUE 99 END DATE1 MISSINGVALUE 99 END DATE2 MISSINGVALUE 9999 END -- 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

1 1

Multiple response questions
by epidata-list＠lists.umanitoba.ca 14 Feb '07

14 Feb '07

Dear Epi-data users: What is the best way to deal with multiple response questions? Thanks, Kristine Hopkins Population Research Center, University of Texas at Austin

1 0

BUG REPORT : minus sign for MISSINGVALUE with more than one field on the same line
by epidata-list＠lists.umanitoba.ca 14 Feb '07

14 Feb '07

Hello, I noticed a strang bug in Epidata Entry when I use the '-' sign to input a MISSINGVALUE in a field. I created three fields on one line to enter a date with incomplete value : Date ## ## #### Age ## Sexe # In the check file I used the following DATE MISSINGVALUE 99 END DATE1 MISSINGVALUE 99 END DATE2 MISSINGVALUE 9999 END When I type a '-' followed by enter or tab in the field - DATE, the focus goes to DATE2. - DATE1, the focus goes to the field AGE - DATE2, the focus goes to the field SEXE This only happens when I use the minus sign and when there are several fields on the same line. -- Jean-Marie Chrétien CHU d'Angers - Centre de Recherche Clinique 4 rue Larrey 49933 ANGERS Cedex 9 Tel : 02.41.35.59.76 Fax : 02.41.35.59.68

1 0

��dryangyun��
by epidata-list＠lists.umanitoba.ca 14 Feb '07

14 Feb '07

where is your master websites,thank you

1 0

sorting rec file
by epidata-list＠lists.umanitoba.ca 13 Feb '07

13 Feb '07

"Is there any possibility to sort the file without exporting the data?" Within EpiData Entry: Define the ID field as a key ID key unique end then after entering all data, export data (#6 in toolbar) to EpiData file and mark "use index as sort order" In EpiData Analysis read myfile sort id savedata newfile copyfile myfile.chk newfile.chk Regards Jens Lauritsen EpiData Association

1 0

Sorting of rec File ?
by epidata-list＠lists.umanitoba.ca 13 Feb '07

13 Feb '07

Dear all, Because of our study design and the randomization procedure in our trial the ID can not be entered in the correct order. Later on it would be more convenient to enter data in the order of the ID. Is there any possibility to sort the file without exporting the data? Thanks in advance, Markus

1 0

QES with two colmns?
by epidata-list＠lists.umanitoba.ca 12 Feb '07

12 Feb '07

Hello all, As times goes by, Im getting the idea that data entry is an art more then a science. Many ptojects have to keep a lot of fields in the paper form and at the end the paper is not big enough to have them all. The consequence is that the QES file should be as similar as possible with paper forms. These last couple weeks Im tryin to make a QES file with tow columns but I cant get to straight up the second column in the REC file. When I make the QES I line up the second by hand, since menu EDIT\line field comand seems to work fine only with one column. But whe the REC file generated the variables, labels and filds of the second column are not align as the QES file (sometimes the labels are uncomplete). Exemple: QES p71 RECUPRACAO # p72 SEQUELA # p73 OBITO# p74 CAUSA MORTIS <A > p75 NECESSITOU INTERNACAO # p76 PROLONGOU A INTERNACAO # p77 MEDICAMENTO SUSPENSO # p78 HOUVE MELHORA # p79 POSOLOGIA ALTERADA # p710 NECESSITOU TRATAMENTO ADICIONAL # p711 HOUVE REEXPOSICAO # p712 REACAO REAPARECEU # p713 PRIMEIRA NOTIFICACAO # p714 NOTIFICADO PARA INDUSTRIA # -- Exemple of resulting REC. p71 RECUPRACAO # p72 SEQUELA # p73 OBITO# p74 AUS<A > p75 NECESSITOU INTERNACAO # p76 PROLON# p77 MEDICAMENTO SUSPENSO # p78 HOUVE MELHORA # p79 POSOLOGIA ALTERADA # p710 NECESSITOU TRATAMENTO ADICIONAL # p711 HOUVE REEXPOSICAO # p712 REACAO REAPARECEU # p713 PRIMEIRA NOTIFICACAO # p714 NOTIFICADO PAR # The most amazing to me is that I´ve worked on some QES files with tow columns before and sometimes it works fine and sometimes it just does not work. At this file for example, a few segments work fine and some dont. Any tip to work around this matter? Thank you, Pedro Emmanuel Brasil Abraço forte e que a força esteja com você, Pedro Emmanuel

1 2

Web Site problem corrected. - new graph command ready v2 b106 analysis
by epidata-list＠lists.umanitoba.ca 08 Feb '07

08 Feb '07

There has been a problem with the website download principle on www.epidata.dk during the last 24 hours. The problem has been resolved, but affected downloads as well as the "Mantis" documentation system. In the latest test build for analysis (b106) a new graph type was included, called "dplot" This will plot a cumulative frequency curve for a given variable. Example of usage: * Show a cumulated frequency percent plot for a variable (here height): dplot height dplot height /yline=50 - the values contained are the same as those showed at : freq height /CP * With option /P: Show a probit plot. * If the data are distributed as Gaussian (Normal) then a probit plot will show data on a straight line. X axis: Midpoints of all data. (e.g. if x = 1 3 ...., then first observation is at x=(1+(3-1)/2)=2 Y axis: The cumulated percentage is converted to probability and graphed as the inverse normal probability dplot height /P There is an unresolved issue with small numbers (incorrect showing) Regards Jens Lauritsen

1 0

How to edit a graph and set further options
by epidata-list＠lists.umanitoba.ca 07 Feb '07

07 Feb '07

Gilles responded correctly on the question of further defining graph options: Try the command: SET GRAPH EDIT = ON which will work in v1 of analysis In the test version v2.0 the "set graph edit" is not allowed. Instead you have to add /edit to any graph command. e.g. scatter v1 v2 /edit But many users will (I assume) use the dialogs from the workprocess toolbar. And in these the /edit is by default added to all graph commands. Regards Jens Lauritsen EpiData Association

1 2