- EpiData-list - University of Manitoba Mailing Lists

Select top 3 among 8
by epidata-list＠lists.umanitoba.ca 05 Aug '13

05 Aug '13

Hello all. I have a small constraint in my chk file and I wonder how I could program it. I have 8 options (8 variables V1 - V8). Interviewers are asked to chose their top 3 options. So that the legal values are 1, 2, 3 and there must not be the same value twice. I dont know how to write short instructions to check that. Thanks for your advices. Serge. -- *M. Aymar Serge SOMDA, MSc. **Biostatisticien* * Responsable du Centre de Calcul Centre MURAZ Bobo Dioulasso, Burkina Faso http://www.centre-muraz.bf/ Bureau des Essais cliniques Institut Claudius Regaud Toulouse, France http://www.claudiusregaud.fr/ * * *

1 3

odd error the 6th time
by epidata-list＠lists.umanitoba.ca 01 Aug '13

01 Aug '13

I use EpiData3.1 because I have used to the qes/chk/rec model. Now I encountered a strange problem as follow:My problem: In my project there are a series of y/n variables (0: no 1: yes 9:unkown) with default value 0. When I enter default 0 for about 40 times the EpiData goes funny.all qes,chk,rec files are in the attached. The error message after entering 40 of 0 for y/n variable:Access violation at address 00002020. Read of address 00002020. The qes and chk files are compressed into file named ontar.rar which is attached .Many thanks and bestwishes!hshxue@aliyun.com22:21 2013/8/1

1 0

Size limit in the new epidata version
by epidata-list＠lists.umanitoba.ca 25 Jul '13

25 Jul '13

Hi, I've tried to locate the answer to my question on your web pages, but since I couldn't find it, I hope you don't mind answering a quick question: Does the new epidata manager have a limit of 999 entry lines in the same way as the old qes-file? Or is there some way of extendig the number of entry lines in a qes-file? Thank you!

1 1

How to select text/string variables
by epidata-list＠lists.umanitoba.ca 23 Jul '13

23 Jul '13

I have a database which I imported from excel into EpiData stat, the unique IDs are string e.g (cdh01/09, cdh02/09......cdg01/09,cdg/02/09....). How can I select a data set with only "cdh" for further analysis? Walter

1 1

I am desperated
by epidata-list＠lists.umanitoba.ca 23 Jul '13

23 Jul '13

I use EpiData3.01 because I have used to the qes/chk/rec model. Now I encountered a strange problem as follow:My problem: In my project there are a series of y/n variables (0: no 1: yes 9:unkown) with default value 0. When I enter default 0 for about 40 times the EpiData goes funny.Qes and chk files are attached(I deleted rec file because the message with rec file was rejected yesterday). The error message after entering 40 of 0 for y/n variable:Access violation at address 00000000. Read of address 00000000.Many thanks and best wishes!hshxue@aliyun.com22:28 2013/7/23 ------------------------------------------------------------------发件人：epidata-list-request(a)lists.umanitoba.xn--ca:2013723-qs5qj53hohuja674bkku440d 1:00收件人：epidata-list@lists.umanitoba.ca;主　题：EpiData-list Digest, Vol 117, Issue 10 Send EpiData-list mailing list submissions to epidata-list(a)lists.umanitoba.caTo subscribe or unsubscribe via the World Wide Web, visit http://lists.umanitoba.ca/mailman/listinfo/epidata-listor, via email, send a message with subject or body 'help' to epidata-list-request(a)lists.umanitoba.caYou can reach the person managing the list at epidata-list-owner(a)lists.umanitoba.caWhen replying, please edit your Subject line so it is more specificthan "Re: Contents of EpiData-list digest..."EpiData-list mailing list___________________________________Today's Topics: 1. odd error (epidata-list@lists.umanitoba.ca)----------------------------------------------------------------------Message: 1Date: Mon, 22 Jul 2013 11:31:16 +0800From: epidata-list(a)lists.umanitoba.caSubject: [EpiData-list] odd errorTo: <epidata-list(a)lists.umanitoba.ca>Message-ID: <c8395064-d34a-4ebd-933e-b4a23e9dd584(a)aliyun.com>Content-Type: text/plain; charset="utf-8"I use EpiData3.01 because I have used to the qes/chk/rec model. Now I encountered a strange problemas follow:My problem: In my project there are a series of y/n variables (0: no 1: yes 9:unkown) with defaultvalue 0. When I enter default 0 for about 40 times the EpiData goes funny.all qes,chk,rec files are in the attached. The error message after entering 40 of 0 for y/nvariable:Access violation at address 00000000. Read of address 00000000.Many thanks and best wishes!hshxue@aliyun.com11:29 2013/7/22聽-------------- next part --------------A non-text attachment was scrubbed...Name: =?UTF-8?B?b250YXIucmFy?=Type: application/octet-streamSize: 3876 bytesDesc: not availableUrl : http://lists.umanitoba.ca/pipermail/epidata-list/attachments/20130721/c7e0b… of EpiData-list Digest, Vol 117, Issue 10*********************************************

1 0

Re: EpiData-list Digest, Vol 117, Issue 9
by epidata-list＠lists.umanitoba.ca 23 Jul '13

23 Jul '13

Thank you, Katie, for the suggestion! I think missing data was the issue with the Stata format. We found a way to successfully import data into SPSS. Export the data file from EpiData Manager as a .csv file (I used the Linux setting). Using Excel, I opened a blank spreadsheet and imported the .csv file via the "add from text" option under "data". I specified that it is delimited by commas, and then changed each column from "general" format to "text" format. This was tedious but necessary because Excel automatically deletes any leading 0s upon import. If you are using these notes as instructions, DO NOT CLICK IN ANY CELL. It will revert to general format, and delete leading 0s. I saved the file as a .txt file (using MS-DOS format). Then I opened in Microsoft Word. I deleted the variable names. I then used find and replace to find "^t" and replace with "^p". SPSS doesn't like tabs, so I needed to replace all tabs with paragraphs. You can use the syntax EpiData Manager produces when exporting to SPSS, I think. Or write your own. I wrote my own, because, as noted earlier in this thread, I had variables embedded within the lines. Worked like a charm! -Natalie

1 0

odd error
by epidata-list＠lists.umanitoba.ca 22 Jul '13

22 Jul '13

I use EpiData3.01 because I have used to the qes/chk/rec model. Now I encountered a strange problem as follow:My problem: In my project there are a series of y/n variables (0: no 1: yes 9:unkown) with default value 0. When I enter default 0 for about 40 times the EpiData goes funny.all qes,chk,rec files are in the attached. The error message after entering 40 of 0 for y/n variable:Access violation at address 00000000. Read of address 00000000. Many thanks and best wishes!hshxue@aliyun.com11:29 2013/7/22

1 1

Field Length Issue
by epidata-list＠lists.umanitoba.ca 18 Jul '13

18 Jul '13

Hello, I am currently working to try to import data into SPSS. I created the .dat file with StatTransfer from the Stata-formatted file. Looking at the .sps file, it appears that one of my variables which should be 70 characters (and is specified as such in the EpiData Manager) contains only 1 character. I noticed this issue while rewriting the syntax to pull the data into SPSS. Our variables are embedded within the string variables (e.g., EPIVAR = 56 -> VAR1 = 5 VAR2 = 6). We used spaces as missing data. Could it be an issue with missing data? It shouldn't be, if the field is specified as 70 characters! How can this be? It is making me very anxious.

1 2

Scripting in Manager
by epidata-list＠lists.umanitoba.ca 16 Jul '13

16 Jul '13

Using: Test version EpiData Manager: v1.5.0.10 3 NOENTER fields: id1, length 5, integer id2, length 5, integer id3, length 5, string 1 MUSTENTER field serno, length 4, integer In the field serno I type in SCRIPT, AFTER ENTRY: id1 := serno; DEFINE id2Temp: integer; id2Temp := serno+1000; id2 := id2Temp; id3 := string(serno); Result in EpiData Entry Client: id1: correctly done id2: correctly done id3: nothing done (except a "." written into the field) I cannot get strings working, neither from string to string nor (as above) trying to transform an integer to a string. I tried different ways, nothing worked. Consulting the Wiki was not revelatory. Dates constructed from date components (with 1 or more components set to missing, and then making approximations) also work impeccably. Floats not yet tested. -- 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 0

Can I "tell" RELATE how many children records to create for a specific household
by epidata-list＠lists.umanitoba.ca 15 Jul '13

15 Jul '13

I have a household survey that asks the head of the household several general questions and then also interviews all women of reproductive age within that household. The number of eligible women can thus range from 0 to several. Is there a way in Epi Data to essentially tell the program that there will be 2 women's records for household A and 3 womens records in household B, and so on. I remember in Epi Info (dos) we could specify the number of "children" records in the KEY UNIQUE field (if I put a "2" on that line then after 2 "childrens' records were created then Epi would bounce back to the main rec file. Thanks for any suggestions. Conrad

1 1