Dr. Victor Lofgreen

Part 1: Ph.D. Mixed-Methods Research

The primary reason to use a mixed research method or model is that a single approach falls short of meeting the goals of the research project. By using a mix of both quantitative and qualitative techniques, the researcher is able to gain a clearer view of the variables or activities that compose the phenomena under study.

Triangulation
The concept of triangulation represents the use of two or more techniques of observation, measurement, or description of a phenomenon to answer a research question. The term triangulation comes from the geometric procedure used to discover the location of an unknown point of a triangle when the length of one side and two angles of the triangle are known. Symbolically, it represents using more than one perspective or point of view to understand the features of the research question (Tashakkori & Teddlie, 1998, p. 18).

Forms of Data Collection
Data collection can take on a wide variety of forms. At the quantitative end of the spectrum, data is most often collected using a highly structured questionnaire or a prepared instrument. At the qualitative end of the spectrum, the data may come from an unplanned interview. Between the two extremes lie a number of options that favor either the quantitative or qualitative form of data, from descriptive statistics through univariate and multivariate analysis on the quantitative side to content analysis and thematic analysis from narrative data on the qualitative side (pp. 112–124).

From a mixed methods or model perspective, the researcher constructs a procedure that includes both quantitative and qualitative assessments that will enlarge or enrich the overall view of the phenomena. This process can take on a number of different formats. Within the study, the researcher may decide to use a qualitative or quantitative primary approach and add a data observation from the opposite approach, either at the same time or in a sequential fashion.

Examples of This Approach

1. A community measure per 1,000 population is compared over time to monitor the difference in a pretest/posttest design, and then an open-ended survey of community members or public officials is added, to gain a qualitative description of the community’s experience.
   
   
2. Conversely, a descriptive case study of the community based on qualitative reports might be further enhanced by the inclusion of some data that more specifically compares or measures some key features of the community’s experience.

Tashakkori and Teddlie (1998)identify three categories of mixed data analysis:

1. Concurrent Mixed Data Analysis
   This category involves the parallel analysis of both quantitative and qualitative data at the same time. The same data may be treated as both qualitative and quantitative by transforming them from one form to the other and then conducting the analysis. An example might be classifying  quantitative data to reveal specific categories of subjects and then using the data to develop the descriptions of the categories (p. 128).
   
   
2. Sequential Qualitative to Quantitative Analysis
   Here, subjects are divided according to some qualitative characteristic. Quantitative data are then collected and the groups analyzed and statistically compared for variations of variables within groups. Tashakkori and Teddlie refer to this method as typology development.
   
   Another example of this category is the identification of groups of subjects through content analysis of qualitative data, (e.g., interviews or discussions with focus groups). Using these categories, the researcher can develop an instrument and use factor analysis to determine the validity of the groups in representing the phenomena (p. 133).
   
   
3. Sequential Quantitative to Qualitative Analysis
   Here, the quantitative measures are used to create the categories and the qualitative analysis to describe the differences between the groups. A common application of this technique is to develop categories from the scores obtained from an objective measure and classifying subjects into these classes. Then an in-depth follow-up interview is conducted to gather more descriptive data from each subject. The analysis of the follow-up data is used to confirm the categories created from the quantitative data (p. 135).

Closing Note
From this brief review it becomes apparent that it is possible to mix types of data and analysis in a variety of ways to achieve a better understanding of the phenomena under investigation. It is important that the variations chosen by the researcher are logically consistent with the research question and the data available. The mixed methodology allows us to broaden the scope of the analysis and to present a more robust view of the phenomena.

Reference
Tashakkori, A., & Teddlie, C. (1998). Mixed methodology combining qualitative and quantitative approaches. Thousand Oaks, CA: Sage Publications.