Hello to everyone,
I prepared ‘Agro-Climatic Map of Ahmednagar district with three zones’ using Rainfall, Temperature, Soil, DEM and Agricultural productivity layers in ArcGIS. I want conduct the fieldwork to study ‘farmer’s perception on climate and its impact on agricultural productivity’. There are 14 tahsils covering 1581 villages, therefore selection of sample villages is needed.
Can anyone suggest me.....
1. Appropriate ‘sampling method’ for selection of villages?
2. How many villages will be selected as sample (sample size)?
3. For data collection how many respondents (farmers) will be select from each sample village?
I hope getting some advices and answers from you.
Thank you in advance
Hi Sandip
You can try multistage stratified random sampling based on your zones.Broadly each zone may be considered as a strata.
Dear Sandip Baban Sasane,
If the 1581 villages are homogeneous, use simple random sampling without replacement. If the villages are not homogeneous, use stratified random sampling.
You can select 10% (160) villages from the 1581 villages in order to obtain reasonable accuracy in estimates. Don't select beloow 5% (80) villages.
Considering that you have identified some ways to stratify data already, and you desire to conduct high quality research, my advice would include involving or at least some project review by a professional statistician. If none available, a well published professor or researcher review would be helpful. In addition, the type of farmer, crops, economic and educational background, methods of farming, size of farm, etc. could some some added factors to consider in a survey, but some are not going to be willing to share personal or business information so easily. And you may want to identify if some areas have greater climate concerns than others. Do farmers use weather almanac, real time or forecast weather, or other indicators to help make decisions. If desiring to address climate change concerns, I would suggest leaving that out unless the vicinity is being impacted by ongoing sea level rise or something that research has specifically identified for your area. You did not mention flooding, but this and wetlands (hydric soils) can be an attribute to soil types. If you find that you just have too much to sample, select the climate zone that is probably most affected by climate or where farmers are having the greatest losses, such as loosing farms, unable to feed family, severe erosion or barren lands, etc.
Selection of sample size depends upon several factors like accuracy of estimate wanted, the cost involved, the population size, strata sizes, time allotted for survey etc. Thanks to Ette Etuk.
Thanks to all respected scholars for suggesting me appropriate sampling method and valuable guidance.
I delineate my study area in three agro-climatic zones, therefore with applying 'stratified random sampling' methods, I will select sample villages from each agro-climatic zones.
All,1581 villages are not homogeneous, some of them are irrigated (27 %) and most of them are' Rainfed' (73 %). In irrigated areas impact of rainfall on crop productivity is not significant factor (r= 0.30 or below) but in rainfed areas, rainfall is significant factors (r= 0.80 or more). Can I select all sample villages only from Rainfed areas?
Can anyone work on the same topic, so please, send me same references.
Thanks once again.
Dear Sandip Baban Sasane,
Welcome.
Thank you for your acknowledgement.
God bless you.
I wish you success in life.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size