There are many introductory books on social and educational research methods. Some of the most well known of these textbooks, such as those by Moser and Kalton (1982), Cohen and Manion (1980) and Hoinville et al. (1978) are widely used on undergraduate and postgraduate courses in universities and colleges and have been reprinted many times to serve successive cohorts of students. These books, and others like them, present research largely as an unproblematic process concerned with sampling, questionnaire design, interview procedures, response rates, observation schedules, and so on. They present an idealized conception of how social and educational research is designed and executed, where research is carefully planned in advance, predetermined methods and procedures followed, and ‘results’ are the inevitable conclusion. In essence, such books take what they perceive to be the methods used in the natural sciences as their model, and seek to present social and educational research as being equally ‘scientific’ in its methods.
In practice, however, it is now widely recognized that the careful, objective, step-by-step model of the research process is actually a fraud and that, within natural science as well as within social science, the standard way in which research methods are taught and real research is often written up for publication perpetuates what is in fact a myth of objectivity (Medawar, 1963). The reality is very different. There are now several autobiographical accounts by scientists themselves and academic studies by sociologists of science that show that natural science research is frequently not carefully planned in advance and conducted according to set procedures, but often centres around compromises, short-cuts, hunches, and serendipitous occurrences.