In situations where administrative divisions can be
divided into only two groups based on the intensity of
conflict, use low intensity and high intensity. If there is no
clear geographical variation in the intensity of conflict (all
regions were affected more or less the same), you may consider
a different stratification variable. Bear in mind that
the purpose of stratification is to ensure that minorities,
which may be small in number and concentrated in certain
areas of the country, are not missed by random selection
of administrative division. For example, if in a religiously
diverse country the majority of Christians live in one province
or region, it may be more important to stratify by religion
than by conflict intensity to ensure that Christians
are included in your sample. There are two considerations
when choosing a stratification variable: (1) there should be
reason to believe that members of different groups have divergent
views or experiences; and (2) there should be some
geographic clustering of the groups. Following this example,
if your variable is religion instead of conflict intensity,
you may be able to divide a country’s administrative areas
into: (a) predominantly Muslim; (b) predominantly Christian;
and (c) similar numbers of Muslims and Christians.
In situations where administrative divisions can bedivided into only two groups based on the intensity ofconflict, use low intensity and high intensity. If there is noclear geographical variation in the intensity of conflict (allregions were affected more or less the same), you may considera different stratification variable. Bear in mind thatthe purpose of stratification is to ensure that minorities,which may be small in number and concentrated in certainareas of the country, are not missed by random selectionof administrative division. For example, if in a religiouslydiverse country the majority of Christians live in one provinceor region, it may be more important to stratify by religionthan by conflict intensity to ensure that Christiansare included in your sample. There are two considerationswhen choosing a stratification variable: (1) there should bereason to believe that members of different groups have divergentviews or experiences; and (2) there should be somegeographic clustering of the groups. Following this example,if your variable is religion instead of conflict intensity,you may be able to divide a country’s administrative areasinto: (a) predominantly Muslim; (b) predominantly Christian;and (c) similar numbers of Muslims and Christians.
การแปล กรุณารอสักครู่..
