Menu ENN Search
Language: English Français

Epi Info command syntax error

This question was posted the Assessment forum area and has 5 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Suzanne Fuhrman

Nutrition in Emergencies, WFP

Normal user

13 May 2014, 08:45

I am analysing SMART data using EpiInfo 7. We used cluster sampling, so during the analysis I am using complex sample statistics. When I try to run a complex sample frequency I get 'command syntax error unexpected token: 5.' This happens when I try to run a frequency on any variable. The cluster is put as the PSU. What am I doing wrong here?

When I open different datasets and do the same steps for analysis I am not having a problem. Is it something with the format of the data?

Thank you.

Anonymous 730

Nutrition and Food Security Officer

Normal user

13 May 2014, 23:21

To be able to understand your question and provide appropriate assistance,you need to give a bit more detail.If you could mention what your PSU is and possibly paste the command here and the message as it appeared in EPI INFO that would be good.
Another thing to note is that someone advised me not to use EPI INFO 7 but rather EPI INFO 3.5.4 etc.
I'm not sure if it is related to such issues as yours.

Mark Myatt

Consultant Epideomiologist

Frequent user

14 May 2014, 08:08

I am not a Windows user and have avoided using EpiInfo after ongoing problems and data-loss a few years ago. I can, therefore, only suggest a few things things to try based on what I have picked up from colleagues:

(1) Check your syntax (i.e. what you typed). The message implies that the fifth object is badly specified. Check for odd characters if file and variable names. Check spaces and other separators. Check for missing or extra quotes and brackets.

(2) Try sorting the dataset on PSU before asking for a complex analysis.

(3) Take a close look at the data. Do you have missing values in your PSU variable? Do you have a cluster with only missing values in your analysis variable? Do you have a cluster with just a single record? Are there empty records in the dataset (including a blank record at the end)? Most procedures expect identifying data (e.g. PSU, weight, stratum IDs) to have no missing values. Another issue may be that you have odd data-types (e.g. a letter in a number field). Do a line listing / browse of your dataset to check for these issues.

(4) Revert to an earlier version of EpiInfo. I have heard of problems with version 7. BTW ... I have just been using the old DOS version of CSAMPLE for a legacy application and this runs well in DOSBOX on Windows and on the the Macintosh and on other UNIX machines.

You should report this problem to the EpiInfo helpdesk. They have a telephone helpline (USA 404-498-6190, 0830-1630 EST). They might be able to solve your issue in minutes.

Please let us know how you get on with this so we can document the problems and solution to help others i

Tamsin Walters

en-net moderator

Forum moderator

14 May 2014, 13:02

From Nina Dodd (FAOSO):

Probably the first row of data has no data values(cluster number) for the cluster variable. Ensure the first row has a data value.

Nina

Suzanne Fuhrman

Nutrition in Emergencies, WFP

Normal user

14 May 2014, 21:24

Hi, I ended up just downloading Epi Info 3.5.4 and there was no problem running the command there. I gave up trying to figure out how to fix it in Epi Info 7!

Anonymous 730

Nutrition and Food Security Officer

Normal user

15 May 2014, 01:45

Good!

Back to top

» Post a reply