In order to avoid selection bias, several aspects were considered for the selection of the interview and survey participants. First, purpose sampling was applied for the selection of the provinces Punjab, Sindh, Khyber Pakhtunkhwa as well as for Islamabad Capital Territory, in which interviews and surveying were conducted. Balochistan could not be included in the analysed sample due to the highly volatile security context. The selection criterion for the four provinces was geopolitical diversity: Punjab and Sindh border India, Khyber Pakhtunkhwa borders Afghanistan while Islamabad is located quite central and epitomises an urban melting pot of different Pakistani ethnic groups and organisations. Pakistan is a diverse country and this selection was anticipated to enable a great variation among the independent variables and ensure a greater representativity of the results. As the data revealed, most of the research participants conduct operations and activities in more than one province. This is assumed to be reflected in their answers and inputs.
Applying randomisation for the sample selection was not possible because the size of the studied population is unknown or difficult to estimate. The respondents were selected from a database which I established ahead of the field research. To establish the database, initial desk research was conducted. The database was complemented during the field research. Geographical representativity, but also the background of the respondents, i.e. military ranking, religious background, domestic/international statute or gender, were considered in the selection procedure. This ensured that military officers from both strategic command and operational level (mostly retired), but also with different religious background (Shia, Sunni, Christian) are represented in the sample. In addition, it was ensured that the opinion of both domestic and international organisations operating in different provinces was captured. The interviewees were contacted in person. In most cases, the contact to the respondents was made in advance of the field research. The respondents were informed about the safe conditions of participating in the interview under the provisions of the ethical approval obtained from Dublin City University. They were also informed about the benefits and impact of this research. Interviews and surveys were carried out in Islamabad, Rawalpindi, Karachi, Lahore and Peshawar.