The Irish Social and Political Attitudes Survey was conducted in Autumn 2001, and was funded by the Programme for Research in Third Level Institutions of the Higher Education Authority. The study was commissioned by a group of researchers in Trinity College Dublin and University College Dublin. The purpose of the survey was to record details on social and political attitudes in Ireland. Number of cases: 2,493
Probability: Cluster. In selecting the sample for the ISPAS a three-stage clustered sample design based on the electoral register was used. In the first instance a random sample of Primary Sampling Units (PSU’s) was selected from the Electoral Register. [220 DEDs were selected as PSUs with an average of 19.9 cases in each]. In the second, a random sample of households was selected from within each PSU. In the third stage a random person within each household was selected. The sampling frame used for this study was the most up-to-date national Electoral Register. Electors are recorded in the electoral list in so-called Polling Books. For sample selection purposes these Polling Books were reconstituted into areal units known as District Electoral Divisions. There is a total of 3,400 District Electoral Divisions (DEDs) in Ireland. These DEDs are the most spatially disaggregated areal units in Ireland for which census data are available and are the standard PSU building block for random sample selection. Once the Electoral Register has been re-built into the District Electoral Division structure a random sample of PSUs was selected. Each PSU was made up of the District Electoral Division or aggregate thereof using a minimum population threshold criteria. The PSU's were selected with a probability proportionate to size (on a PPS basis). When the first stage of PSU selection was completed we then selected a sample of households from within each. This was implemented using a random start and selecting a systematic sample at an appropriate interval to yield a constant number of households per cluster. At the third stage of sample selection we then randomly selected a target respondent from within each. This was done using a so-called 'next birthday' rule as a randomising criterion. In other words, we selected the person from the household who had the next birthday from among all household members who were aged 18 years or over. To do this the interviewer simply listed all members of the household aged 18 or more and identified the one who had the next birthday. This 'next birthday' person was the target respondent. No substitution was allowed in cases where that target respondent was unavailable or refused to participate etc.
Face-to-face interview: PAPI
Self-administered questionnaire: Paper
