Menu ENN Search
Language: English Français

ENA software

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

Diallo zakaria

etudiant en securité alimentaire et nutrition

Normal user

5 Jun 2017, 12:09

Hello everyone
After completing the anthropomotric data entry in ENA, I verified the plausibility report and it shows 26% and what is problematic. Are the results of this survey reliable? thank you

Damien Pereyra

Independant Nutrition Evaluation Expert

Normal user

5 Jun 2017, 17:04

Dear Zakaria,

Is this "anthropomotric data entry in ENA" from a SMART survey? It is even from a nutritional survey? ENA for SMART propose you an intersting tool to analyse a dataset which is the plausibility check. If the final score is 26%, you should first of all, deeply analyses this repport to understand where the problem are/is? It could be in the sampling process, sample size not reached, poor antropometric measurment, etc.
You should share this repport and give us some more detail about where and how the data have been collected...
Looking forward for some more details
Best
Damien

Diallo zakaria

etudiant en securité alimentaire et nutrition

Normal user

5 Jun 2017, 17:57

Thank you Mr Damien
In fact, it was in the framework of a study on the determinants of malnutrition in children from 6 to 59 months in a commune at the same time we wanted to evaluate the nutritional status of these children. The sample is of the reasoned type, that is to say the villages are chosen on the basis of a certain number of criteria to know the prevalence of GAM and SAM following the active screening by MUAC in this municipality. So the sampling was not based on ENA at the outset. thank you

Abbas Kedir Jelde

Information Management Specialist

Normal user

6 Jun 2017, 11:11

Dear Colleague
If you have conducted standard nutrition survey using cluster sampling you are expected to check every item that contributes to having the 26% report.  ENA for SMART makes the following points:
• Flagged data (% of out of range subjects) should not exceed 7.5% . Otherwise estimation of Age or measurement of height or weight may be wrong. Check the values from your plausibility output.
• P-value for Overall Sex ratio should not be less than 0.001 (Significant chi square). Otherwise one sex is dominant in your data. Under normal conditions both sexes have 50% chance to be included in the survey. But a slight difference may not incur extreme penalties.
• P-value for Age ratio (6-29 vs 30-59) should not be less than 0.001 (Significant chi square). If p-value exceeds those, either children aged less than 30 months are dominant or above 30 months is the most frequent observed figure. This may result if they did not estimate age of child accurately using local and event calendar. Unless differences in age of children are supported by evidence, family planning and inter-generation gap
• Digital preference score penalties for weight, height and MUAC should not exceed 20. Otherwise your enumerators favour only some digit like 0.5, 0.0 etc. Check this for consistency
• Standard Deviation for WHZ score should less than 1.20 if data is normally distributed. This mainly based on the quality of the data
• Skewness (biased to left or right from normality)and Kurtosis (flatness or steep shoot) of the data for WHZ need to be less than or equal to <±0.6 for better quality of survey
• P-value of Poisson distribution of WHZ-2 (GAM rate)need to be >0.001

Sameh Al-Awlaqi

Public Health and Nutrition Consultant

Normal user

6 Jun 2017, 12:07

Hi Abbas,

If you used the standard planning page on ENA software, the two-stages sampling and conducted a good standardization test for your team with , you shouldn't have problematic results. In case you have done that, and you are still collecting data on a daily basis, go on and insert the rest of data ( complete set) then check the plausibility.

In case you have a completed set of data, revise the data set you have before copying it into ENA, after copying into ENA, identify the flagged data in purple colour. Make sure to tick smart flags on the 4th tab on ENA( exclusion of z-scores with) before generating the plausibility and the word report. Make sure to use July 9th 2015 ENA version( the latest).

Other factors that may cause problematic findings mentioned on other colleagues comments above.

Hope that helps,

Regards,

Sameh

Sameh Al-Awlaqi

Public Health and Nutrition Consultant

Normal user

6 Jun 2017, 12:12

Apologies my response above is addressed to Zakaria who originally posted the question.

Back to top

» Post a reply