- EpiData-list - University of Manitoba Mailing Lists

Formatting and exporting data
by epidata-list＠lists.umanitoba.ca 05 Dec '05

05 Dec '05

Thanks Jens for your prompt and detailed response to my question about exporting numeric fields for use in Stata. While the fix you suggest in Stata (for var * : capture format X %9.0g) is simple for experienced Stata users, it's a bit painful as a general approach for new users. I'm definitely in favour of exporting according to type and sticking to %9.0g formatting, if this is possible. Cheers, Suzanna Suzanna Vidmar Research Fellow Clinical Epidemiology and Biostatistics Unit Royal Children's Hospital Flemington Rd Parkville Victoria 3052 telephone: +61 3 9345 6372 facsimile: +61 3 9345 6000

1 0

Re: Appending based on "dbf" files
by epidata-list＠lists.umanitoba.ca 03 Dec '05

03 Dec '05

A user reported problems with appending files which were exported from Excel as Dbase (dbf) I suggest you export as csv instead since some of the DBF files Excel creates are NOT of the correct format. An alternative could be to try different DBF formats (dbaseII DbaseIII DbaseIV) Some mails to the list regarding this problem were sent as HTML. The list will from today convert such mails to text format. But it is preferable that mails to the list are sent as plain text, since these mails are shorter - a nice gesture to persons with slow modem lines. In some mail programs, e.g. Thunderbird http://www.mozilla.com/thunderbird/ which is a freeware programme one can define which type of mails a certain receiver of mails can read. Jens Lauritsen EpiData Association

1 0

Re: Display format and Stata export
by epidata-list＠lists.umanitoba.ca 03 Dec '05

03 Dec '05

For NON-Stata users this mail is not relevant. Regarding the display format I forgot to show a solution: 1. export data to Stata format Either use EpiData and choose the version of Stata you prefer or use EpiC 2. Read the data into Stata 3. Set the display format as you desire: e.g. for a file myfile.rec to export to Stata and set numerical format %9.0g for all variables: epic export stata test\testdata\bromar.rec * replace Start Stata and issue the commands: use bromar.dta for var * : capture format X %9.0g ci ....... The reason for capture is that if you have a string variable there would be an error. Instead of the "for var *" you can use for var v1 v3 v13 etc and name the variables explicitly. In newer versions of Stata an alternative is foreach Kind regards Jens Lauritsen EpiData Association

1 0

Formatting and exporting data
by epidata-list＠lists.umanitoba.ca 03 Dec '05

03 Dec '05

The question of how to format data by Susanne Widmar is a good one. Allow me to give a lengthy explanation since this is an issue in which we should agree on a uniform principle as well for import, export and use of data in EpiData Analysis. This relates to two aspects: 1. Accuracy of data - data types (binary, integer, real, double...) - we wish to replicate the precision of the collected data 2. How many decimals are shown on output from commands - we wish to see as much as needed for the interpretation - often less than aspect 1. In Stata the format is for some commands controlling the number of decimals, such as the CI command, whereas for other commands the display format is part of the command specification. The same "confusion" is now in EpiData Analysis where there is only partial consistency of showing and controlling formatting. In my view output formatting of the output should be decided by the command not by the data. In v1.1 build 54+ of EpiData Analysis (now available for testing) I have implemented for frequencies new options: freq v1 /D0 // would give percentages with no decimals freq v1 /D1 // would give percentages with one decimal For cross tables there are the settings for tables: TABLE PERCENT FORMAT COL P1() TABLE PERCENT FORMAT ROW P1{} TABLE PERCENT FORMAT TOTAL P1[] in which the user can define the number of decimals for the percentages For other commands (describe, means..) the display format of commands is either fixed or depending on the size of the estimates. User specification of output format will be implemented later. For import and export to Stata we could rewrite the procedures to work on datatype rather than format as it is now, if this is possible. We could then set all numerical variables to format %9.0g Another strategy could be to allow EpiData Analysis to save data in Stata format (which Bill Gould CEO of StataCorp has given permission to do). I suggest that users comment on the uniform strategy: a. All calculation is based on maximum accuracy with the data at hand b. Output formation should be part of each command. c. Data storage (type of variable) should be as strict as possible. Accomplishment of this is more important than speed of import and export. Regards Jens Lauritsen EpiData Association

1 0

Display format
by epidata-list＠lists.umanitoba.ca 01 Dec '05

01 Dec '05

This is a question/comment for the folk at EpiData. Much of the data here is entered in EpiData and then analysed in Stata. I've noticed that in Stata the display format for a field is the same as the data entry format in EpiData. For example if the field "smoke" (0=no, 1=yes) has the field type set to # in EpiData, in Stata the display format is %1.0f. This can cause confusion with Stata commands that base the display format of their output on the display format of the field, e.g. the Stata command "ci". If I type the following into Stata: ci smoke, binomial The output I get looks like this: Variable | Obs Mean Std. Err. [95% Conf. Interval] -------------+--------------------------------------------------------------- sex | 44 0 0 0 0 because the mean, SE, and CI have been rounded to the nearest integer. Is it necessary for the output/exported format of fields to be the same as the data entry format? Can numeric fields be exported with a format of at least 9.0g? Cheers, Suzanna Suzanna Vidmar Research Fellow Clinical Epidemiology and Biostatistics Unit Royal Children's Hospital Flemington Rd Parkville Victoria 3052 telephone: +61 3 9345 6372 facsimile: +61 3 9345 6000

1 0

re: appending files properly
by epidata-list＠lists.umanitoba.ca 30 Nov '05

30 Nov '05

1 0

Re: How to append datasets properly
by epidata-list＠lists.umanitoba.ca 29 Nov '05

29 Nov '05

To assist Kate with the appending problem we must know more: a. Are you using EpiData Entry or EpiData Analysis b. Which version and build See about box or in analysis: Version c. Show a listing of the variable types of the four variables in each of the two files. To do this: In EpiData Entry: Document - file structure In Analysis: read the file and then issue the command "var" Cut and paste into for the relevant four variables into a new mail to the epidata-list Regards Jens Lauritsen EpiData Association

1 0

Appending
by epidata-list＠lists.umanitoba.ca 29 Nov '05

29 Nov '05

Kate: Could it be that one of the variables at some places has an illegal value? I observed it with dates: while EpiData will accept a year "203" for instance, it seems that if you append in EpiData Analysis, you could get this type of error message. You could go to the variable(s) in question in EpiData Analysis, sort on the variable, browse and look at the beginning an the end whether there is anything that could be fishy. This of course after you used the command "variables" in both datasets to ensure that the structure is indeed identical. Hans I have a question regarding appending files properly. I'm attemping to append 2 datasets, and the program states that 4 variables among the 2 datasets are incompatible. The variables appear to be consistent across both datasets, so I'm at a bit of a loss. Any help would be appreciated. Thanks, Kate -- Hans L Rieder, MD, MPH Jetzikofenstr. 12 3038 Kirchlindach Switzerland Tel: +41 31 829 4577 Mob: +41 79 429 9945 Web: http://www.tbrieder.org

1 0

How to append datasets properly
by epidata-list＠lists.umanitoba.ca 28 Nov '05

28 Nov '05

1 0

How to: Create SPC graphs with same x period axis value for two consecutive periods
by epidata-list＠lists.umanitoba.ca 23 Nov '05

23 Nov '05

Question to info at Epidata: I have two successive periods: 1 and 2 I wish a graph where the first 6 observations in each period are included and on the graph the x axis should say 1 2 3 4 5 6 in both of them Using the Ichart.rec example included in installation of EpiData Analysis: file: ichart.rec which has 27 observations variables: lege : counts for a given doctor (lege is doctor in Norwegian) dag : consecutive number of observations (values are 1 - 27) Assume that period 1 in the study was observation 1-14 and observation 15-27 was period 2. Read the data en calculate as indicated after the "//" is comments ................................................................................................ read ichart define d ## d = _n // _n is the same as recnumber the observation number of the data if d > 15 then d = d -15 // now d is the correct number withing each period select if d < 7 // only include first 6 observations from each period ichart dag lege // this would show x axis with values of variable dag ichart dag lege /break=7 /xlabel=d // this would make a break for the two periods and use d as label ................................................................................................ Similarly for date variables: ................................................................................................ ichart tid lege /break="01/01/2003" /xlabel=d // if the value of "tid" was "01/01/2003" for observation 7 ................................................................................................ The format of the break value should for dates be the same as the x axis variable. The graph above will show a gap btw. the two periods since the x-axis variable has values from 1 to 27 If one has consecutive numbers e.g. 1-12 and you wish to have larger separation of the two periods: ................................................................................................ define d1 ## d1 = _n if d1 > 6 then d1 = d1+2 // experiment a bit if you wish 1, 2 or more added ichart d1 lege /break=7 /xlabel=d // this would move the two parts of the graph apart ................................................................................................ Regards Jens Lauritsen EpiData Association ps. Unfortunately the /xlabel option is not in the current commands and functions reference. Will be added to next release.

1 0