Discussions Discussions


RIMS Survey

Display Date: 6/5/12


Rural Microfinance and Livestock Support Program (RMLSP) of Afghanistan supporting tow components Microfinance and livestock in 8 provinces of the country. We are about to do RIMs survey for both components. I have gone through RIMS survey guideline and found it very useful. The confusion I have is about sample selection. This is because veterinary services sub component under Livestock component have been covered all communities of 58 districts in 5 provinces, on the other hand we should do RIMs survey for the Livestock all sub components simultaneously. 



That's great but what exactly is your question. If you can elaborate your question further, there will be someone who will help out.


Posted on 6/5/12 12:18 PM.


Posted on 6/10/12 8:05 AM.

Sampling is widely used for gathering information about a population so it is concerned with the selection of a subset of individuals from a population to estimate characteristics of the whole population. I can suggest some sampling methodologies that may helps in RIMS Survey
1.Systematic sampling relies on arranging the target population according to some ordering scheme and then selecting elements at regular intervals through that ordered list
2.Cluster Sampling, Sometimes it is more cost-effective to select respondents in groups ('clusters'). Sampling is often clustered by geography, or by time periods.
3.Panel sampling is the method of first selecting a group of participants through a random sampling method and then asking that group for the same information again several times over a period of time.


Mahmood Stanikzai

Posted on 6/10/12 10:25 AM.

RIMS guidelines say that you should sample all project households as a single groups - i.e. take a random sample of 900 project households selected from all project participants regardless of which components of the project covered them. However this approach does not aim to generate information on the impact of individual components - which could be really useful to know.

Is this survey to be carried out at the start-up, mid-term or completion stage of the project? This is important as surveys at mid-term and completion aim to compare data from a sample households that have benefited from the project with data from a sample of households at the start of the project who have not yet benefited from the project. The different between indicator measurements will give an idea about changes that the project may have bought about and hence the impact of the project.

Posted on 7/16/12 11:48 AM.

Thank you very much for your instruction. I did what was in RIMS guideline and that match your idea as well. we interviewed 900 project households. Now we are in data entry process if you remember we were told in Thia M&E and KM training that IFAD has particular database for RIMS data. I have tried to find it in portal but I could not. Could you pleas help me to access mentioned database and RIMS survey report if you have.


Posted on 7/25/12 5:53 AM in reply to Edward Mallorie.

The RIMS Survey Software can be downloaded by going to the IFAD website (http://www.ifad.org/operations/rims/).

The above link has all the resources required for the RIMS Survey including the Software which should be downloaded from the link (RIMSSetup.msi) which is on the right side of the webpage under "Hot links"and installed.


Posted on 7/25/12 6:26 AM in reply to Asadullah Battar.

The RIMS Survey Software can be downloaded by going to the IFAD website (http://www.ifad.org/operations/rims/).

The above link has all the resources required for the RIMS Survey including the Software which should be downloaded from the link (RIMSSetup.msi) which is on the right side of the webpage under "Hot links"and installed.


Posted on 7/25/12 6:33 AM in reply to Asadullah Battar.

Thank you very much for your kind information,


Posted on 7/28/12 3:37 AM in reply to Marlam Kupar Kharmudai.

In Bangladesh ifad support project started RIMS and Household survey at a time. for RIMS it consider 900 household and 1200 household survey.but in HILIP project target household is 115000 directly and indirect 688000. in this situation you produce a separate survey form for identify target group. that is (LCS, Common Interest Group,Beel user group)

Posted on 10/15/13 6:59 AM in reply to Edward Mallorie.


The various components of HILIP have different target groups and different types of intervention, it therefore makes good sense for an impact survey to find out about the results of the project for each of these groups. A socio-economic household survey could therefore have a sub-sample for each of these groups (LCS, CIG, BUG) – which in M&E terms is called a “domain”. The sample for each domain needs to be large enough produce meaningful results. If we accept a 90% confidence interval (CI) and 10% precision (which means we can be 90% certain of differences / changes of 10% or greater), then a sample of 200 drawn from 20 randomly selected clusters (i.e. project groups) with 10 households being randomly selected in each cluster. This would be done for each domain, so for three domains the total sample is 600 (3 x 200). As changes will be happen to households in the absence of project interventions, a control group is needed to act as a basis of comparison. In this case, I would suggest three control groups, selected from villages where there are no project interventions but using criteria similar to the selection criteria used for LCS, CIG and BUG. These control groups should be the same size as the project groups, so we have a control sample of 3 x 200 = 600, and the survey covers 1,200 households in total.

While RIMS anchor indicator survey, including child anthropometrics, could be combined with the socio-economic household survey, this would significantly complicate data collection and analysis. For this project it seems to make sense to carry out a separate RIMS survey to cover a sample of 900 drawn from villages impacted by the project. This would therefore include households who are not in any of the project groups (LCS, CIG, BUG) but may have benefited from road and market infrastructure. A question should be added to find out in what way, if at all, sample households participated – i.e. were they members of LCS, CIG or BUG? If M&E resources allow, it would be good to have a control sample of 900 drawn from a sample of non-project villages where no infrastructure development took place.

I hope this is helpful.

Posted on 10/15/13 12:11 PM in reply to Hashibun Nahar Khanam.


Video Gallery

Image Gallery