Willing (2009) argues that many qualitative researchers tend not to work with "variables" that are defined by the researcher prior to the research project because