Conclusion drawing and verification You start to draw conclusions about what things mean from the start of data collection, noting patterns and regularities, positing possible structures and mechanisms, etc. These are then firmed up during and after the data collection. Miles and Huberman stress that this should be accompanied throughout by a verification process: that is, you are testing their validity and reliability. Is an explanation plausible? Can you find evidence confirming it? Can a finding be replicated in another data set?
These three flows of activity, together with the activity of collecting the data itself, form a continuous iterative process. For example, coding a data set (data reduction) will lead to ideas of how the data may be displayed, which may help form a tentative conclusion about the operation of a mechanism, or for changing the display or coding system. Note that this has its counterpart in qualitative analysis. Data reduction is achieved through descriptive and summary statistics; data display through graphs and tables of correlations; conclusion drawing through the use of inferential statistics, test of significance, etc. ( The results of such quantitative analyses, of course, supplement the qualitative analysis if you have multiple methods in a case study yielding both quantitative and qualitative data.). The difference is that while the quantitative analysis techniques are fully codified and largely non-contentious, qualitative data analysis is more fluid and contested.
Conclusion drawing and verification You start to draw conclusions about what things mean from the start of data collection, noting patterns and regularities, positing possible structures and mechanisms, etc. These are then firmed up during and after the data collection. Miles and Huberman stress that this should be accompanied throughout by a verification process: that is, you are testing their validity and reliability. Is an explanation plausible? Can you find evidence confirming it? Can a finding be replicated in another data set?
These three flows of activity, together with the activity of collecting the data itself, form a continuous iterative process. For example, coding a data set (data reduction) will lead to ideas of how the data may be displayed, which may help form a tentative conclusion about the operation of a mechanism, or for changing the display or coding system. Note that this has its counterpart in qualitative analysis. Data reduction is achieved through descriptive and summary statistics; data display through graphs and tables of correlations; conclusion drawing through the use of inferential statistics, test of significance, etc. ( The results of such quantitative analyses, of course, supplement the qualitative analysis if you have multiple methods in a case study yielding both quantitative and qualitative data.). The difference is that while the quantitative analysis techniques are fully codified and largely non-contentious, qualitative data analysis is more fluid and contested.
การแปล กรุณารอสักครู่..
