Qualitative, Quantitative, and Mixed Methods Approaches PDF

Qualitative, Quantitative, and Mixed Methods Approaches Prepared by: Sir Leo Elona Research Methods and Statistics  Qualitative Research Qualitative researchers are interested in understanding how people interpret their experiences, how they construct their worlds, and...

Qualitative, Quantitative, and Mixed Methods Approaches Prepared by: Sir Leo Elona Research Methods and Statistics  Qualitative Research Qualitative researchers are interested in understanding how people interpret their experiences, how they construct their worlds, and what meaning they attribute to their experiences. For example, rather than finding out how many retired folks take on part - time jobs after retirement, which could be done through a survey, we might be more interested in how people adjust to retirement, how they think about this phase of their lives, the process they engaged in when moving from full - time work to retirement, and so on. These questions are about understanding their experiences and would call for a qualitative design.  Definition Denzin and Lincoln (2005), for example, begin their paragraph - long definition by saying “ qualitative research is a situated activity that locates the observer in the world ” After several sentences on the practice of qualitative research, they conclude with “ qualitative researchers study things in their natural settings, attempting to make sense of, or interpret, phenomena in terms of the meanings people bring to them ”  Definition A more concise though several years older definition that I particularly like is by Van Maanen (1979): Qualitative research is “ an umbrella term covering an array of interpretive techniques which seek to describe, decode, translate, and otherwise come to terms with the meaning, not the frequency, of certain more or less naturally occurring phenomena in the social world ”  Definition Basically, qualitative researchers are interested in understanding the meaning people have constructed, that is, how people make sense of their world and the experiences they have in the world.  A second characteristic of all forms of qualitative research is that the researcher is the primary instrument for data collection and analysis. Since understanding is the goal of this research, the human instrument, which is able to be immediately responsive and adaptive, would seem to be the ideal means of collecting and analyzing data.  Rich Description Finally, the product of a qualitative inquiry is richly descriptive. Words and pictures rather than numbers are used to convey what the researcher has learned about a phenomenon. There are likely to be descriptions of the context, the participants involved, and the activities of interest. In addition, data in the form of quotes from documents, field notes, and participant interviews, excerpts from videotapes, electronic communication, or a combination of these are always included in support of the findings of the study. These quotes and excerpts contribute to the descriptive nature of qualitative research. Three elemental methods: Approaches to Coding (1) Descriptive, (2) In Vivo, and (3) Process coding.  Descriptive Coding A descriptive code assigns labels to data to summarize in a word or short phrase—most often a noun—the basic topic of a passage of qualitative data. These eventually provide an inventory of topics for indexing and categorizing, which is especially helpful for ethnographies and studies with a wide variety of data forms (field notes, interview transcripts, documents, etc.). Descriptive codes are perhaps more appropriate for social environments than social action.  Descriptive Coding 1As I walked toward the school, there was a 7-11 convenience store 1 block away, next to a small professional office building: an optometrist, podiatrist, and other medical/health-related clinics. Directly across the street was an empty lot, but next to that stood a Burger King restaurant. 1BUSINESSES An analyst would extract all passages coded BUSINESSES from various field notes to compose a more detailed inventory of the case and to construct a narrative describing the business climate in the area.  In Vivo Coding This is one of the most well-known qualitative coding methods. In Vivo coding uses words or short phrases from the participant’s own language in the data record as codes. It may include folk or indigenous terms of a particular culture, subculture, or microculture to suggest the existence of the group’s cultural categories (e.g., in a hospital, you may hear unique terms such as “code blue,” “sharps,” and “scripts”).  In Vivo Coding In Vivo coding is appropriate for virtually all qualitative studies but particularly for beginning qualitative researchers learning how to code data, and studies that prioritize and honor the participant’s voice. Phrases that are used repeatedly by participants are good leads; they often point to regularities or patterns in the setting. In Vivo codes are placed in quotation marks to differentiate them from researcher-generated codes.  Examples are taken from a coded interview transcript about an adolescent girl’s experiences with school: I 1 hated school last year. Freshman year, it was awful, I hated it. And 2this year’s a lot better actually I, um, don’t know why. I guess, over the summer I kind of 3stopped caring about what other people thought and cared more about, just, I don’t know. 1“HATED SCHOOL” 2 “THIS YEAR’S BETTER” 3 “STOPPED CARING”  Process Coding This coding method uses gerunds (“-ing” words) exclusively to connote observable and conceptual action in the data. Processes also imply actions intertwined with the dynamics of time, such as things that emerge, change, occur in particular sequences, or become strategically implemented. Process coding is appropriate for virtually all qualitative studies, but particularly for grounded theory research that extracts participant action/interaction and consequences.  Here is an example from an interview transcript about an adolescent girl explaining how rumors get spread: Well, that’s one problem, that [my school is] pretty small, so 1 if you say one thing to one person, and then they decide to tell two people, then those two people tell two people, and in one period everybody else knows. 2 Everybody in the entire school knows that you said whatever it was. So.... 1SPREADING RUMORS 2 KNOWING WHAT YOU SAID  Three affective methods (1) Emotion, (2) Values, and (3) Evaluation coding.  Emotion Coding Perhaps obviously, this method labels the emotions recalled and/or experienced by the participant or inferred by the researcher about the participant. Emotion coding is particularly appropriate for studies that explore intrapersonal and interpersonal participant experiences and actions. It also provides insight into the participants’ perspectives, worldviews, and life conditions. Note that a participant himself or herself may sometimes label the emotion, and thus, it should be In Vivo coded in quotation marks.  The following example is taken from an interview transcript about a middle-aged man complaining about one of his work colleagues: 1I just hated it when he got awarded with the honor. 2 I mean, we’re praising mediocrity now. Never mind that what you’ve accomplished isn’t worth squat, it’s all about who you know in the good ol’ boys network. 1“HATED IT” 2”BITTERNESS”  Values Coding This is the application of three different types of related codes onto qualitative data that reflect a participant’s values, attitudes, and beliefs, representing his or her perspectives or worldview. A value (V:) is the importance we attribute to ourselves, another person, thing, or idea. An attitude (A:) is the way we think and feel about oneself, another person, thing, or idea.  Values Coding A belief (B:) is part of a system that includes values and attitudes, plus personal knowledge, experiences, opinions, prejudices, morals, and other interpretive perceptions of the social world. Values coding is appropriate for studies that explore cultural values, identity, intrapersonal and interpersonal participant experiences and actions in case studies, appreciative inquiry, oral history, and critical ethnography.  Here is an example from an interview transcript about a female university student discussing her political beliefs: 1 Government regulation of women’s health issues has gotten out of hand. It’s not about “protecting” us, it’s about their need to control and dominate women 2 through covert religious ideology. White Christian men are deciding what’s law and what’s moral and what’s, how it’s supposed to be. 3 They can say, “It’s not a war on women” all they want, but trust me—it’s a war on women. 1B: GOVERNMENTAL CONTROL 2 B: COVERT RELIGIOUS MOTIVE 3 A: MISOGYNIST  From Codes to Patterns Your initial or First Cycle coding of data generates an array of individual codes associated with their respective data chunks. Let’s take a look at a fictional and extended example of how First Cycle codes transform into Second Cycle Pattern codes and then get inserted into matrices and networks.  From Codes to Patterns A selected series of codes related to the first month of withdrawal symptoms described by a participant voluntarily participating in a smoking cessation treatment program, in random order and with their First Cycle code types indicated, are as follows:  From Codes to Patterns 1. ANXIETY [Emotion code] 2. NERVOUSNESS [Emotion code] 3. “HURT SOMEONE BAD” [In Vivo code/Emotion code] 4. RESTLESSNESS [Emotion code] 5. DEEP BREATHING [Process code]  From Codes to Patterns 6. THROAT BURNING [Process code] 7. “FELT LIKE CRYING” [In Vivo code/Emotion code/Process code] 8. ANGRY [Emotion code] 9. “EATING A LOT MORE” [In Vivo code/Process code]  From Codes to Patterns 10. WANDERING AROUND [Process code] 11. HABITUAL MOVEMENTS [Descriptive code] 12. MEMORIES OF SMOKING [Descriptive code] 13. SMELLING NEW THINGS [Process code] There are several ways to approach the categorizing or patterning of these 13 codes. One possible way is to pattern them by code type: EMOTIONS (ANXIETY, NERVOUSNESS, “HURT SOMEONE BAD,” RESTLESSNESS, “FELT LIKE CRYING,” ANGRY) PROCESSES (DEEP BREATHING, THROAT BURNING, “FELT LIKE CRYING,” “EATING A LOT MORE,” WANDERING AROUND, SMELLING NEW THINGS) DESCRIPTORS (HABITUAL MOVEMENTS, MEMORIES OF SMOKING)  EMOTIONS (ANXIETY, NERVOUSNESS, “HURT SOMEONE BAD,” RESTLESSNESS, “FELT LIKE CRYING,” ANGRY) Since negative and strong emotions seem to play a critical role in withdrawal symptoms from smoking, EMOTIONS as a Pattern code choice makes sense. One can even enhance the code further with the adjective NEGATIVE EMOTIONS.  Clustering Cluster 1: DEEP BREATHING, THROAT BURNING, “EATING A LOT MORE,” SMELLING NEW THINGS Cluster 2: WANDERING AROUND, HABITUAL MOVEMENTS Cluster 3: “FELT LIKE CRYING,” MEMORIES OF SMOKING  Clustering Cluster 1: DEEP BREATHING, THROAT BURNING, “EATING A LOT MORE,” SMELLING NEW THINGS First, what do the four codes in Cluster 1 have in common? They seem to be all upper-body functions: respiratory, sensory, and digestive. The analyst reflects on what the four codes have in common; they seem to have a PHYSICAL CHANGES theme that unifies them, and thus get that Pattern code assigned to them.  Clustering Cluster 2: (WANDERING AROUND, HABITUAL MOVEMENTS) The codes of Cluster 2 (WANDERING AROUND, HABITUAL MOVEMENTS) seem to evoke a metaphoric RESTLESS JOURNEY of some sort.  Clustering Cluster 3: “FELT LIKE CRYING,” MEMORIES OF SMOKING Cluster 3’s codes (“FELT LIKE CRYING,” MEMORIES OF SMOKING) suggest a conceptual Pattern code of REGRETFUL LOSS. Where did the Pattern code labels of RESTLESS JOURNEY and REGRETFUL LOSS come from? They came from the researcher’s reflection on what their constituent codes seemed to have in common.  Notice that these four Pattern codes—(1) NEGATIVE EMOTIONS, (2) PHYSICAL CHANGES, (3) RESTLESS JOURNEY , and (4) REGRETFUL LOSS—are one person’s analytic proposals. Other researchers reflecting on and clustering the First Cycle codes might develop different Pattern codes altogether. Thus, an important principle to note here is that Pattern coding is not always a precise science—it’s primarily an interpretive act.  Analyzing The researcher can now use these four Pattern codes in various ways, according to the needs of the study. Basic narrative description is one approach; and visual displays are another primary way of analyzing data in fresh perspectives.  Narrative Description The researcher can compose a section that identifies and elaborates on the Pattern code, weaving its component First Cycle codes into the narrative and supporting it with field note data: “Smoking withdrawal symptoms during Month 1 include a restless journey for the individual: “I found myself just wandering around the house, just walking from room to room because I couldn’t smoke, so I didn’t know what to do with myself.” The ex- smoker also continues to replicate habitual movements related to smoking, such as reaching for a cigarette pack in a shirt pocket, or leaving an indoor office to go outside to smoke. These physical actions interrelate with, and may even be caused by, several of the negative emotions induced by nicotine withdrawal: anxiety, nervousness, and restlessness.”  Matrix Display Matrix displays chart or table the data —including codes—for analytic purposes. They organize the vast array of condensed material into an “at-a-glance” format for reflection, verification, conclusion drawing, and other analytic acts.  Matrix Display Matrix displays chart or table the data —including codes—for analytic purposes. They organize the vast array of condensed material into an “at-a-glance” format for reflection, verification, conclusion drawing, and other analytic acts.  Matrix Display Initiating Smoking Month 1 Month 6 Cessation Patterns NEGATIVE EMOTIONS Anxious, nervous, Occasionally anxious angry, aggressive PHYSICAL CHANGES Gained 5 pound, felt On weight loss program “burning” sensation in after gaining 20 pounds, throat and lungs heightened sense of smell RESTLESS JOURNEY Wandering and habitual Habitual movements movements REGRETFUL LOSS “Felt like crying”, hyper Nostalgic for smoking, conscious of cessation “hangs around” smokers  Matrix Display “FELT LIKE MILD CRYING” ANXIETY REGRETFUL COMFORT IN LOSS CAMARADEREI MEMORIES NOSTALGIA OF SMOKING  Reference First and Second Edition Copyright © 1994 by Matthew B. Miles and A. Michael Huberman Third Edition Copyright © 2014 SAGE Publications, Inc.  Reference Qualitative data analysis: a methods sourcebook / Matthew B. Miles, A. Michael Huberman, Johnny Saldaña, Arizona State University. The Three Types of Designs Three types Qualitative research Quantitative research Mixed methods research Research design Plan or proposal to conduct research Intersection of: Philosophical worldviews Strategies of inquiry Research methods  Strategies of Inquiry Quantitative Qualitative Mixed Methods Experimental Narrative Sequential designs research Concurrent Non- Phenomenology Transformative experimental Ethnographies designs, such as Grounded surveys theory studies Case study  Strategies of Inquiry Quantitative Qualitative Mixed Methods Experimental Narrative Sequential designs research Concurrent Non- Phenomenology Transformative experimental Ethnographies designs, such as Grounded surveys theory studies Case study  Strategies of Inquiry Quantitative Qualitative Mixed Methods Experimental Narrative Sequential designs research Concurrent Non- Phenomenology Transformative experimental Ethnographies designs, such as Grounded surveys theory studies Case study Research Methods Quantitative Mixed Qualitative Methods Methods Methods Pre-determined Both pre-determined Emerging methods Instrument-based and emerging methods Open-ended questions Both open- and questions Performance, attitude, closed-ended questions Interview, observation, observational, and Multiple forms of data document, and audio- census data drawing on all visual data Statistical analyses possibilities Text and image Statistical Statistical and text analyses interpretation analyses Themes, patterns Across databases interpretation interpretation Research Designs as Worldviews, Strategies, and Methods Tend to or Qualitative Quantitative Mixed Methods typically... Approaches Approaches Approaches Use these Constructivist/ Post-positivist Pragmatic philosophical Advocacy/ knowledge claims knowledge claims assumptions Participatory knowledge claims Employ these Phenomenology, Surveys & Sequential, strategies of inquiry grounded theory, experiments concurrent, & ethnography, case transformative study, & narrative Employ these Open-ended Closed-ended Both open- and methods questions, emerging questions, pre- closed-ended approaches, text or determined questions, both image data approaches, numeric emerging and data predetermined approaches, & both quantitative and qualitative data and analysis Research Designs as Worldviews, Strategies, and Methods (cont.) Tend to or Qualitative Quantitative Mixed Methods typically... Approaches Approaches Approaches Use these Positions him- or herself Tests of verifies theories Collects both Collects participant or explanations quantitative and practices of meanings Identifies variables to qualitative data research, as the Focuses on a single study Develops a rationale for researcher concept or phenomenon Relates variables in mixing Brings personal values questions or hypotheses Integrates the data at into the study Uses standards of different stages of inquiry Studies the context or validity and reliability Presents visual pictures setting of participants Observes and measures of the procedures in the Validates the accuracy information numerically study of findings Uses unbiased Employs the practices of Makes interpretations of approaches both qualitative and the data Employs statistical quantitative research Creates an agenda for procedures change or reform Collaborates with the participants Criteria for Selecting a Research Design The Research Problem An issue or concern that needs to be addressed If the problem calls for Explanation or theory testing: Quantitative Exploration or understanding: Qualitative One approach alone is inadequate: Mixed methods Personal Experiences Training, preferences, time, resources Audience Advisors, journal editors, graduate committees, etc. The Research Proposal Introduces the Research Problem or Question Reviews Existing Literature on Topic Outlines Proposed Methodology or Procedure Explains How the Acquired Data will be Analyzed or Interpreted Shows the Importance and Significance of the Proposed Research Addresses Anticipated Outcomes and Relevance to the Larger Community Appendices Should include Examples of Letters of Inquiry used to Recruit Sample Population Should include Examples of Letters of Informed Consent for Participants Should include Examples of Instruments or Data Gathering Tools Should include Examples of Any Other Materials that will be Used to Conduct Research May also Include Estimates for Budget, Staffing, and Equipment Needs  Why Mixed Methods? Quantitative data can reveal generalizable information for a large group of people These data often fail to provide specific answers, reasons, explanations or examples Qualitative research provides data about meaning and context regarding the people and environments of study Findings are often not generalizable because of the small numbers & narrow range of participants Both methods have strengths and weaknesses When used together, these methods can be complimentary  Criteria for Choosing a Strategies Theoretical Impleme Integration Priority Perspective ntation No Sequence At data Equal Concurren collection Explicit t At data Sequential analysis - At data Qualitative Qualitative interpretation first Sequential Implicit With some Quantitati - combination ve Qualitative  Criteria for Choosing a Strategies What is implementation sequence of the quantitative and qualitative data collection in the proposed study? What priority will be given to the quantitative and qualitative data collection and analysis? At what stage in the research project will the quantitative and qualitative data and finding be integrated? Will an overall theoretical perspective (e.g., gender, race/ ethnicity, lifestyle, class) be used in the study? Alternative Strategies and Visual Models Sequential Explanatory Design Quan Data Collection QUAN QUAL Quan Data Analysis Qual Data Collection Qual Data Analysis Interpretation of Entire Analysis Alternative Strategies and Visual Models Sequential Exploratory Design Qual Data Collection QUAL QUAN Qual Data Analysis Quan Data Collection Quan Data Analysis Interpretation of Entire Analysis Alternative Strategies and Visual Models Sequential Transformative Design QUAL QUAN Vision, Advocacy, Ideology, Framework QUAN QUAL Vision, Advocacy, Ideology, Framework Alternative Strategies and Visual Models Concurrent Triangulation Strategy QUAN QUAL QUAN Data QUAL Data Collection Collection QUAN Data QUAL Data Analysis Analysis Data Results Compared Alternative Strategies and Visual Models Concurrent Nested Strategy QUAL QUAN QUAN QUAL Analysis of Analysis of Findings Findings  Alternative Strategies and Visual Models Concurrent Transformative Strategy QUAN + QUAL QUAN Vision, Advocacy, QUAL Ideology, Framework Vision, Advocacy, Ideology, Framework  Data Collection Procedures  Identify and be specific about the type of data. Some forms of data such as interviews and observations can be either quantitative or qualitative. Although reduction information to numbers is the approach used in quantitative research, it is also used in qual. Research.  Recognize that quantitative data often involve random sampling, so that each individual has no equal probability of being selected and the sample can be generalized to the larger population. In qualitative data collection, purposeful sampling is used to that individuals are selected because they  Data Collection Procedures Relate the procedures specifically to the visual model. For e.g, in a sequential explanatory model, the general procedures can be detailed even further. A discussion of this approach might include describing the use of survey data collection followed by both descriptive and infertial data analysis in the first phase. Then qualitative observations and coding and thematic analysis within an ethnographic design  Data Analysis and Validation Procedures It’s related to the type of research strategy chosen for the procedures. Some of the more popular approaches:  Data transformation: In the concurrent strategies involve creating codes and themes qualitatively, then counting the number of times they occur in the text data. This quantification of qualitative data enables a researcher to compare quantitative results with the qualitative data. For instance, in a factor analysis of data from a scale on an instrument, the researcher may create factors or themes that then can be compared with themes from the qualitative  Data Analysis and Validation Procedures Explore outliers: In a sequential model, an analysis of quantitative data in the first phase can yield extreme or outlier cases. Follow-up qualitative interviews with these outlier cases can provide insight about why they diverged from the quantitative sample. Instrument development: In a sequential approach, obtain themes and specific statements from participants in an initial qualitative data collection. In the next phase, use these statements as specific items and the themes for scales to create a survey instrument that is grounded in the views of the participants. A third, final phase might be to validate the instrument with large sample representative of a population.  Data Analysis and Validation Procedures Examine multiple levels: in a concurrent nested model, conduct a survey at one level (e.g. with families) to gather quantitative results about a sample. At the same time, collect qualitative interviews (e.g., with individuals) to explore the phenomenon with specific individuals in families.  Data Analysis and Validation Procedures  It is necessary the validation of both qualitative and quantitative phases of study. Each of methods has the specific ways, for the qualitative data, the strategies that will be used to check the accuracy of the findings  Report Presentation Structure  For a sequential study, mixed method researcher typically organize the report of procedures into quantitative data collection and qualitative data analysis followed by qualitative data and collection and analysis.  Report Presentation Structure  Then, in the conclusions or interpretation phase of the study, the researcher comments on how the qualitative findings helped to elaborate on or extend the quantitative results.  Report Presentation Structure  Alternatively, the qualitative data collection and analysis could come first followed by the quantitative data collection and analysis. In either structure, the writer typically will present the project as two distinct phases, with separate headings of each.  Report Presentation Structure  In concurrent study. The quantitative and qualitative data collection may be presented in separate section, but the analysis and interpretation combines the two forms of data to seek convergence among the results. The structure of this type of mixed methods study does not as clearly make a distinction between the quantitative and qualitative phases. Report Presentation Structure In a transformative study, the structure typically involves advancing the advocacy issue in the beginning of the study and then using either the sequential or concurrent structure as a means of organizing the content of the study. In the end of the study, a separate section may advance an agenda for change or reform that has developed as a result of the research.  Given the complexity of question formulation in MMR, here’s an example: Study: A mixed methods study of college students’ attitudes regarding gender roles (combining survey research and in-depth interviews) Quantitative hypothesis: Male undergraduates have more traditional attitudes about gender roles than female undergraduates. Qualitative question: How do male and female undergraduates describe their views on gender roles? Mixed methods question: How did the combination of survey research and in-depth interviews provide a more comprehensive understanding of male and female college students’ attitudes about gender roles?  There are two forms of data transformation: quantizing and qualitizing. Quantizing- is the process of transforming qualitative data into quantitative data (transforming qualitative codes into quantitative variables). Qualitizing- is the process of transforming quantitative data into qualitative data (transforming quantitative variables into qualitative codes) (Tashakkori & Teddlie, 1998).  Primary and Secondary Questions Research Question 1 (primary question): How do students describe the impact of the zero- tolerance bullying policy in their school? Research Question 1a (secondary): Do students feel safer because of the policy? Research Question 1b (secondary): Are students more likely to report bullying they experience or witness?  EXAMPLE 1: QUANTITATIVE The purpose of this study is to examine the prevalence of bullying in middle school with and without anti- bullying program x in order to determine the effect of program x on the rates of bullying in middle school. Hypothesis 1: The rates of bullying in middle school will be lower in middle schools that implement anti- bullying program x. Hypothesis 2: Students will be more likely to report bullying they experience or witness after participating in program x.  EXAMPLE 2: QUALITATIVE The aim of this study is to understand and describe middle school students’ perspectives on anti- bullying program x. Research question 1: What kinds of bullying did students experience or witness before program x? Research question 2: What did students think of program x and why? Research question 3: How, if at all, do students think program x has affected bullying in their school?  Mixed Method Questions The purpose of this study is to describe and evaluate the effect of anti- bullying program x on bullying in middle school using a mixed methods approach. Quantitative research question: What effect does anti- bullying program x have on the prevalence of bullying in middle school? Qualitative research question: How do students describe their experiences with bullying before and after anti- bullying program x? Mixed methods research question: How does the mixed methods design of the study contribute to our understanding of the effect of anti- bullying program x and the nature of that effect? Ethical Considerations  References Creswell, J. W. (2003). Research design. Sage publication. Four Worldviews for Research Post positivism Constructivism Determination Understanding Reductionism Multiple participant meanings Empirical observation and Social and historical construction measurement Theory verification Theory generation Advocacy/Participatory Pragmatism Political Consequences of actions Empowerment issue-oriented Problem-centered Collaborative Pluralistic Change-oriented Real-world practice oriented