8
9
10 Appendix 2: Methodology ...........................................................................................85 11 Appendix 3: Interview Discussion Guide .................................................................89
Afghanistan in 2006 1
Preface
This report presents the findings of the single-largest, comprehensive public opinion poll conducted in Afghanistan. The survey was conducted by The Asia Foundation between June and August, 2006 and consisted of a random, representative sample of 6226 in-person interviews with Afghan citizens 18 years of age and above resident in Afghanistan (see Appendix 1). The survey sample was divided first according to urban and rural characteristics of Afghanistan.1 The universe is divided into seven geographical regions consisting of 34 provinces out of which 32 provinces were covered in the survey. Uruzgan, representing 1.1 percent of the population of Afghanistan, and Zabul, representing 1.2 percent, were excluded from the sampling plan due to extreme security conditions during the fieldwork period of the survey. The margin of sampling error is 2.5 percent. The survey report groups the country's 34 provinces into the following seven regions: Central-Kabul (Kabul, Kapisa, Parwan, Panjshir); Eastern (Nangarhar, Kunar, Laghman, Nuristan); South Central (Wardak, Logar, Ghazni, Paktika, Paktia, Khost); South Western (Helmand, Kandahar, Zabul); Western (Badghis, Herat, Farah, Nimroz); Northern (Badakhshan, Takhar, Baghlan, Kunduz, Samangan, Balkh, Jawzjan, Sar-i-Pul, Faryab); Central-Hazarajat (Uruzgan, Ghor, Bamyan, Day Kundi). The sample was distributed proportional to geographical and residential characteristics of population per province. The interviews were completed using multi-stage random sampling. At the outset, the universe was divided into male and female sub-samples. Each region, province, and further strata was allocated an equal number of male and female respondents per sampling points. Field-force members of their respective gender covered the two sub-samples. The instability and frequent fighting in some provinces caused a few sampling points to be adjusted or replaced to keep interviewers out of areas with active violence. Full details on the methodology of the survey are available in Appendix 2 of this report. The survey questionnaire, with percentage of responses by question, can be found in Appendix 3. This survey builds on prior Asia Foundation surveys conducted in Afghanistan in 2003 and 2004, as well as in Bangladesh (2006), Cambodia (2000), and Indonesia (2003) (see http://www.asiafoundation.org/). The survey also uses some questions from the East and South Asia Barometers conducted in Asia as well as the public opinion survey conducted in 2004 on political culture of democracy in Mexico, Central America, and Colombia (see http://www.lapopsurveys.org/). These and other questions in the survey are looked at in greater and comparative detail by various experts to understand political and social progress, trust in state institutions, and human and development security in Afghanistan in a companion volume of five chapters (forthcoming December 2006).

上一页下一页