The data
This paper combines data from several complementary sources.
The (Rwandan) Household Living Conditions Survey (EICV) was
executed between July 2000 and June 2001 and relied on a nationally-representative
sample of 5280 rural households. The Food
Security Research Project (FSRP)2 gathered agricultural production
and land use data over 3 years (i.e. between 2000 and 2002), each
time for both seasons (season A: September–February, season B:
March–August). This survey drew from a subsample of the EICV survey
covering 1584 households.
For the principal component and cluster analysis done for this
paper, we chose overlapping samples from EICV 2001 and FSRP
2001A representing 1433 cases. We use weights that reflect the
probability of being sampled. Multivariate outliers are identified
based on the Mahalanobis distance3 (critical value of 37.697;
p 6 0.001). Based upon this criterion, a total of 55 identified outliers
are excluded. Further, the remaining sample contains 158 cases for
which only incomplete information is available. They are also excluded
leaving 1220 cases in the analysis. The combined EICV-FSRP
sample is representative of the rural inhabitants of Rwanda.
Through the selection of variables, we have tried to identify those
elements necessary to capture each type of the five capitals identi-
fied within the livelihood framework (i.e. natural, physical, human,
social and financial capital), in addition to variables accounting for
regional differences and aggregate incomes and expenditures.
The dataThis paper combines data from several complementary sources.The (Rwandan) Household Living Conditions Survey (EICV) wasexecuted between July 2000 and June 2001 and relied on a nationally-representativesample of 5280 rural households. The FoodSecurity Research Project (FSRP)2 gathered agricultural productionand land use data over 3 years (i.e. between 2000 and 2002), eachtime for both seasons (season A: September–February, season B:March–August). This survey drew from a subsample of the EICV surveycovering 1584 households.For the principal component and cluster analysis done for thispaper, we chose overlapping samples from EICV 2001 and FSRP2001A representing 1433 cases. We use weights that reflect theprobability of being sampled. Multivariate outliers are identifiedbased on the Mahalanobis distance3 (critical value of 37.697;p 6 0.001). Based upon this criterion, a total of 55 identified outliersare excluded. Further, the remaining sample contains 158 cases forwhich only incomplete information is available. They are also excludedleaving 1220 cases in the analysis. The combined EICV-FSRPsample is representative of the rural inhabitants of Rwanda.Through the selection of variables, we have tried to identify thoseelements necessary to capture each type of the five capitals identi-fied within the livelihood framework (i.e. natural, physical, human,social and financial capital), in addition to variables accounting forregional differences and aggregate incomes and expenditures.
การแปล กรุณารอสักครู่..
