New American Foundation

FATA: Inside Pakistan's Tribal Regions

Data and in-depth analysis from all seven tribal agencies of FATA-pakistan

Polling Methodology

This survey was jointly conducted by the New America Foundation (NAF) and Terror Free Tomorrow (TFT), with field work in the Federally Administered Tribal Areas of Pakistan (FATA) by the Community Appraisal & Motivation Programme (CAMP), a Pakistani NGO operating in FATA and Khyber Pakhtunkhwa Province. Interviews were conducted face-to-face with 1,000 FATA residents age 18 or older across 120 villages/sampling points in all seven tribal Agencies of FATA-Pakistan. The fieldwork was conducted from June 30 to July 20, 2010. NAF-TFT-CAMP also conducted 200 separate interviews with local maliks and tribal elders. This study is held for future release. The conclusions in this report are based solely on the 1,000- resident sample. The questionnaire consisted of 43 substantive questions, 19 demographic questions, and 24 quality control questions. Respondents were selected using a multi-stage random stratified sampling methodology. During the course of fieldwork, there were 1,294 contacts attempts made. There were 294 refusals giving the study a net response rate of 77.3 percent. The poll has a +/- 3 percent margin of error at the 95 percent confidence interval.

TFT, NAF and CAMP used face-to-face research techniques in FATA. Interviews were conducted by 22 trained interviewers (male and female) and 6 field supervisors, all of whom are Pakistanis and residents of FATA. Before sending the interviewers and supervisors to the field, extensive project orientation and training workshops were conducted in Peshawar. During the training, interviewers and supervisors were briefed on a number of topics including, but not limited to, the objective of the survey and its details, selection of respondents, the questionnaire (both asking of questions and recording of responses), timing and control issues, and usage of the questionnaire. Mock interviews and field testing were also conducted. Eleven survey teams were deployed, under the supervision of six supervisors. Each team was made responsible for interviewing respondents in their respective areas. Detailed field work-plan and list of villages were shared with the field teams before field deployment.

Interviews were subjected to numerous quality control procedures including 54.1 percent direct supervision of interviews, 27.8 percent back-check in person by the supervisor, and back-check by telephonic call by supervisor or the central office.

The survey was conducted in Federally Administered Tribal Areas (FATA). The target sample was a random selection of Pakistani nationals having FATA residence, both male and female, above the age of 18. The sample covered all seven tribal agencies, namely Khyber, Mohmand, Bajaur, Kurram. Orakzai, North Waziristan and South Waziristan, with the number of sampling points chosen in proportion to the size of each Agency’s population. Following is the sample plan of the survey.

Proportion Stratified Sample Size

Agency Population Sample Leaders Sample Total
Bajaur 595,227 202 40 242
Mohmand 334,453 114 23 137
Khyber 546,730 186 37 223
Orkzia 225,441 77 15 92
Kurram 448,310 152 31 183
North Waziristan 361,246 123 25 148
South Waziristan 429,841 146 29 175
Total 2,941,248 1,000 200 1,200

The National Census data from 1998 are used as the universe for this sample. The national census classifies villages as the primary unit in the rural areas. We used a multi-stage random stratified sampling method, selected over 120 primary sampling units, comprising villages in all the seven Agencies of FATA. The completed primary sample contains 120 sampling points; in which approximately 8 interviews were carried out in households selected in each village using a random walk. A total of 1,000 men and women from a scientific cross-section of FATA in terms of gender, age and other socio-economic characteristics are represented in the sample. For South and North Waziristan, however, local conditions only permitted CAMP personnel to sample men.

After selecting the villages, each village was divided into four hypothetical quarters. A starting point was selected in each quarter. Two starting points were selected for male respondents and two for female respondents. Three interviews were conducted around each starting point. Skipping of three households in Stratum I villages and skipping of three households in Stratum II villages were made after each successful interview. Male and female surveyors were used for male and females respondents respectively. For the selection of a household around a particular starting point the Right Hand Rule (RHR) was used for female interviewees and the Left Hand Rule (LHR) was used for male respondents.

TFT’s previous nationwide surveys of Pakistan can be accessed in the Polls section of www.terrorfreetomorrow.org. Additional findings, particularly the survey of local leaders, have been held for future release.