I have a data set of 4100 households of Teknaf Peninsula, the southern part of Bangladesh. My variables are as follows
Focus Variable: Living inside/outside the forest (data collected by GPS loggers)
Others
> Age
> Education Level
> Gender (considering the family head only)
> Dependency ratio
> Occupation Category (i.e. fisherman, farmer, business, labor, and others)
> Annual Income (sum of income from different sources)
> Length of residence
> Fuelwood collection (yes/no)
> Fuelwood consumption (amount in kg)
I want to develop a statistical model showing the effect of the determinants on the forest encroachment.
I will be highly pleased to get advice on which type of statistical test should I go for.
Thank you in advance.
First of all, I would like to recommend you to revisit previous literature and statistical methods that have been used preciously. Perhaps, that would be the best solution.
As per my understanding from your explanation, your data has been collected through observation and survey. Your dataset also contains both primary and secondary sources and have a dummy variable (though I don't know your standpoint).
If I fall into your situation, I would apply SEM once. However, I'm not confident that SEM would be able to answer the research question, yet I can see understand some pros and cons.
S M Asik Ullah
Congrats you have completed huge data collection,
I suggest that Non-wood forest prodcuts (NWFPs) are also output...............if they are collecting than you s/d take care this aspect also..............................Thanks!
If yes, you will need the list, i will provide you!
Mr. Iftekhar, Thanks for your advice. I am planning to do Binary logistic regression.
Thank you very much, Mr. Raj K Yogi. My research is also concerned about NWFPs. I am planning to write a research article based on NWFP collection.