[SMARTKeymaster]

The papers below may be of assistance for your proposed research endeavour.

Hong Yang & Mercedes de Onis (2008). Algorithms for converting estimates of child malnutrition based on the NCHS reference into estimates based on the WHO Child Growth Standards. BMC Pediatrics, 8:19.
http://www.biomedcentral.com/1471-2431/8/19

Andrew Seal & Marko Kerac (2007). Operational implication of using 2006 World Health Organization growth standards in nutrition programmes: secondary data analysis. BMJ 334:733.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1847893/

Please find below a link to download the CDC statistical calculators (along with instructions) that are used to help interpret prevalence results from nutritional surveys.

http://smartmethodology.org/survey-planning-tools/smart-methodology/interpreting-smart-results-cdc-calculator/

[SMARTKeymaster]

The updated ENA for SMART software is available in English, French, and Spanish. To change languages select ‘Extras’ followed by ‘Change of Language’ in the top left corner of the ENA for SMART software.

[SMARTKeymaster]

This will not have any effect on the Crude Death Rate or Under 5 Death Rate. Some of the demographic information will be affected, but quite minimally. If there are a significant number of HHs with over 20 individuals in your survey then this can be mentioned in the report.

• This reply was modified 10 years, 8 months ago by  SMART.
• This reply was modified 10 years, 8 months ago by  SMART.

[SMARTKeymaster]

There is not a defined acceptable range of design effect for W/H. In a non-cluster simple/systematic random sample the design effect is always 1. In a cluster sample the design effect of W/H is typically between 1 and 2. When planning a survey if there is no information available for design effect the rule of thumb is to use 1.5. Kindly also refer to the following presentation: 3.SM.04 Sample Size Calculation (slides 17-23) in the Survey Planning Tools/ SMART Capacity Building Toolbox/ Survey Manager Section

[SMARTKeymaster]

As you have correctly stated, in the top left section of the ‘Death Rates-Data Entry Individual Level’ tab, there are 20 rows (1 row per individual) that represent 1 household. On the bottom half of the screen is the ‘Data of all households’ (1 household per row) section. As you fill in data in the top left section, the information is automatically recorded below.

If there are more than 20 people in a household (based on your survey definition), for example, let’s say Cluster 1, Team 2, HH 1 has 25 individuals you can enter the data for the first 20 individuals in the top left section of the ‘Death Rates-Data Entry Individual Level’ tab. Next, you can manually select the next row on the ‘Data for all households’ section on the bottom half of the screen and the individual HH section in the top left will reset. In this section you can enter the remaining 5 individuals except you would write it as HH2 (Cluster 1, Team 2, HH 2) and make a note of this on your questionnaire that HH1 and HH2 are the same HH.

[SMARTKeymaster]

The level of precision used for your survey depend on the estimated Crude Death Rate (number of deaths/10,000/day). If previous high quality mortality survey results are available it is best to use those results as a starting point and then estimate if the CDR has increased, decreased, or stayed the same.

Refer to 8.SM.02 ACF-CA Calculating Sample Size for Mortality for regional CDR estimates along with the following recommendations:

For most settings with death rates less than 1.0 deaths/10,000/day, a precision of ±0.3 is appropriate.
LOW: Where death rates are low (0.3 or less) greater precision maybe necessary: ± 0.2
HIGH: Where death rates are very high (> 1.0), a precision of ± 0.4 or 0.5 may be sufficient.
Do not use precision less than ± 0.2 unless there is strong justification.

[SMARTKeymaster]

The ENA for SMART software is generally intended for children under 5 years but can be used to enter data for any age (for example MUAC). Regarding z-scores, the software uses the WHO Child Growth Standards as the reference population and can calculate z-scores for W/H under 5 years, W/A under 10 years and H/A under 19 years.

[SMARTKeymaster]

This scenario does not occur that often but it IS possible to measure a child opposite to what they are supposed to be.

1. In the ENA for SMART ‘OPTIONS’ tab select ‘SHOWING COLUMNS FOR MEASURE, CLOTHES, AND WEIGHING VARIABLES’ followed by ‘SAVE’.

2. In the ‘DATA ENTRY ANTHROPOMETRY’ tab enter “l” for length and “h” for height in this column if a child was not measured according to standard protocol. For example, if a child was measured lying down when they should have been measured standing up, or vice versa, you will indicate this in this column. Only enter H or L if the child was measured opposite to protocol (indicating how the child was actually measured). For example, if a child is 94 cm but he was measured lying down (malaria, sick, etc), an “l” should be entered in the corresponding box of the Measure column. Leave this column blank if child was measured correctly. ENA for SMART will automatically apply a correction factor of 0.7cm.

3. Make a note of this on the cluster control form and the questionnaire.

[SMARTKeymaster]

Kindly refer to the Sampling Methods and Sample Size for the SMART Methodology June 2012 (pg.31,32) which refers to using the Modified EPI method only as a last resort as well as some of the limitations. It can found at the following link:

https://smartmethodology.org/survey-planning-tools/smart-capacity-building-toolbox/

Manager-Complementary Tools and Resources-Handouts-A_Manuals, Case Study-A2 Sampling for SMART.

If other users know of any other documents/papers discouraging the use of the Modified EPI Method kindly post the link on this thread.

[SMARTKeymaster]

Dear Narendra,
There were a couple of minor glitches in the mortality section that were corrected in the June 28th Version on ENA.
Thanks

[Author]

Posts