如何开展混合研究方法混合研究方法的新近趋势.docVIP

How to Conduct a Mixed Methods Study: Recent Trends in a Rapidly Growing Literature Mario Luis Small Keywords qualitative versus quantitative, multimethod research, research design, methodology, formal models Abstract The present article selectively reviews the large number of recent studies that have been described as based on mixed methods. I begin by discussing a body of work that has emerged to promote mixed methods research across the social sciences. I then review and critique empirical studies in each of two general approaches to mixed methods:mixed data? collection studies, which combine two or more kinds of data; andmixed data?analysis studies, which combine two or more analytical strategies, examine qualitative data with quantitative methods, or explore quantitative data with qualitative techniques. I argue that, although mixed methods research is by no means new, empirical studies today combine methods in more diverse and, at times, innovative ways. Nevertheless, important methodological tensions will likely surface as the research becomes more self-reflexive. INTRODUCTION The twenty-first century has seen a remarkable number of publications that identify themselves as based on mixed methods. It is a surprising turn of events. As recently as the 1980s, many researchers were still debating whether quantitative or qualitative methods were superior, reflecting an un-catholic approach to social science methods that threatened to splinter sociology (see Collins 1984). While such attitudes have not disappeared, alternatives have certainly flourished. Today, methodologists of one stripe routinely promote the value of methods different from their own, as when demographers openly call for more “qualitative research” to answer the questions that their own methods cannot address (Goergin 2003, p. 89). In fact, social scientists appear increasingly willing to embrace mixed methods in a number of different forms, as demographers collaborate with interviewers