Discuss (c) semi-structured interview

The survey = the systematic collection of standardized information from a sample of respondents — the breadth instrument of the non-doctrinal approach (U2.1). The survey's eight steps (the sequence answer): (1) the objective's definition (what exactly is to be measured?); (2) the population and the sampling design (below); (3) the instrument's design (the questionnaire — 5.2) and its pretest/pilot (the small-scale trial that catches the ambiguous questions); (4) the field organisation (the enumerators' training where used); (5) the collection (with the call-backs and the replacement rules); (6) the data entry and cleaning; (7) the analysis (U6); (8) the reporting (U7). In legal research: the victimisation survey (the unreported crime's measurement — the dark figure), the legal-needs-and-access survey (justice's demand side), the legal-awareness studies, the profession's surveys. Strengths: breadth, standardization (the groups' comparison), the quantification. Limits: the superficiality risk (the shallow breadth), the non-response bias, the self- report's validity questions. Census and sampling (the survey's base decisions): the census = the complete enumeration (every unit — totality without inference risk; the cost rarely feasible in legal research); sampling = the representative part for the whole. The probability samples (the inference-licensed): the simple random (the lottery), the systematic (every k-th unit), the stratified (the population's layers sampled proportionately — the judges by bench, the prisoners by offence), the cluster (the geographic units…