Category Archives: research design

methodology as recipe

A challenging idea for many graduate students learning how to do interpretive and critical research is that one must make a commitment to a research methodology. The two most common errors/scenarios are that students want to begin with research methods (ways of making and analyzing data) or research design features (the most common is to say they want to do a case study, a topic I discuss in some detail here.)

Screen Shot 2014-03-27 at 4.34.21 PMI find Michael Crotty’s definitions and relationships among the elements that frame research helpful. See this image and here, for a short note on this.

A methodology is a framework that contains core ideas reflecting more fundamental epistemological and ontological groundings, but it also provides guidance on the focus of our inquiry, key concepts, values, and often a hint at methods (although this is a separate matter). A methodology contains core definitional foci and cue us to what to expect in a research project. Methodologies also give us a sense of what we might expect the outcome/product of the research to be, even though there is tremendous variation in representational forms within any methodology.

Methodology as Recipeimgres

To help students understand what a methodology is and why we need one, I use the metaphor of methodology as recipe. A recipe is a framework for preparing food: it has a name that reflects what the outcome will be; a list of ingredients, procedural guidelines; tools and techniques; and often a color photo to show us what the food should look like when we are finished. And, different recipes illustrate fundamental differences. For example, a recipe for a chocolate cake and one for beef stew vary on all of these features and are therefore about something quite different. One would never follow a recipe for a chocolate cake and expect to end up with beef stew.

Chocolate Cake                                                            Beef Stew

flour, sugar, cocoa, eggs, butter                            beef, broth, carrots, potatoes, onions

whisking, stirring, mixing                                         chopping, browning, deglazing

mixer, cake pan, whisk                                              knives, heavy pot

baking                                                                               braising, stewing

Screen Shot 2016-05-04 at 4.47.08 PM Screen Shot 2016-05-04 at 4.47.34 PM

A methodology is a framework for doing research: it contains a name that reflects the outcome; what is needed to identify the research as being within that framework; procedures and tools; and we often have a general notion of what the outcome will look like.

Ethnography

culture, language, rituals, artifacts

field work: prolonged engagement, participant observation, interviewing, kinship charts, mapping, SNA

field notes, photographs, transcripts

thick description; analysis with cultural categories; emic perspective

Screen Shot 2016-05-04 at 5.03.37 PM

 

 

 

Novice cooks and researchers are more likely to follow recipes closely, developing the knowledge and skills that will in time free them from a specific recipe while still working within the recipe framework. A good cook might substitute Guinness beer for some broth in the beef stew, but she is still making beef stew. An ethnographer might substitute live field note taking using social media for more traditional field notes, but she is still taking field notes.

Without a methodology a research project is ungrounded, drifting and has a high probability of being atheoretical. With a methodology, with a recipe, the researcher plans on making an ethnography or a narrative analysis or a hermeneutic investigation because the core ideas, the ingredients, the tools are valued and indeed reflect deeper senses of the nature of both the world and knowledge about it.

learning by example

If you think you will do interpretive or critical research it will be helpful to see what this kind of research looks like. Reading other studies is a window into the research process as well as the ways researchers represent knowledge from their studies. While many genres of research trade in the peer reviewed journal article, to get the most pedagogical benefit from reading qualitative research look more to book and monograph length works. First, good studies are complex and so it just takes more than 25 ms pages to communicate the findings. Second, increasingly interpretive and critical researchers include a confessional methodological tale in an appendix, a rich source of learning from others.

This is not really a part of the literature review for your study (although it could be) and to avoid conflating reading studies related to your research topic with learning
about research methodology and methods, I require my students and encourage others to read book length works that are not in your area of expertise. You might learn something about which you know little, but more importantly can focus on the research process more easily.

So what do I recommend as teachable/learnable texts, remembering that you are not looking for a perfect study but rather a really well done study that has flaws and features that permit you to see how the research process plays out in real research life.

The lists below hews mostly to education and schooling, but not exclusively.imgres

The books I have most often used in my research classes are:

God’s Choice by Alan Peshkin
Dude You’re a Fag by A. J. Pascoe
Ain’t No Makin’ It by Jay MacLeod
Working Class Without Work by Lois Weiss
Home Advantage by Annette Lareau

But, it’s good to go back to what I consider classic texts:

Boys in White by Blanche Geer, Everett C Hughes, Anselm Strauss, Howard Becker
Street Corner Society by William Foote Whyte
The Man in the Principal’s Office by Harry Wolcott
Deep Play: Notes on a Balinese Cockfight by Clifford Geertz (this is more a monograph, but too good not to list)
Learning to Labour by Paul Willis
Living and Dying in Murray Manor by Jaber F. Gubrium
Contradictions of Control by Linda McNeil
Life in Schools by Peter McLaren
The Polish Peasant in Europe and America by Florian Znaniecki and W. I. Thomas
Asylums by Erving Goffman

And, I can’t resist adding what has to be the most controversial piece of research in quite some time…

On the Run: Fugitive Life in an American City by Alice Goffman

 

case study ~ research design, methodology, method?

Since the early 80s when I did my MA degree I have been steeped in “case study research,” and I have the deepest respect for those who have articulated the importance of focusing on the particular in social science research (Robert Yin, Robert Stake, Sharon Merriam, for example). The work of these individuals is valuable ~ Yin provides a valuable foundation for why we should look at cases, and Stake has added detail about different motivations for looking at cases.

With a background in sociology and cultural anthropology my early exposure to case studies created little confusion ~ I was and remain interested in the particular. So researching and understanding a case make conceptual sense. Being schooled in ethnography as a methodology meant that using the language of case study provided a way to engage the tools of ethnography in a flexible way, focusing on pretty much anything that can be identified as a case, that is, a bounded system. Indeed, Creswell defines case study as “an in-depth exploration of a bounded system (e.g., an activity, event, process, or individuals) based on extensive data collection” and he goes on to clarify what a bounded system is: “the case is separated out for research in terms of time, place, or some physical boundaries.” The coupling of a bounded case in naturalistic settings aligns these methodological ideas with interpretivist and critical perspectives on research, but a case can also be investigated within a post-positivist perspective.

In his 2000 book, The Art of Case Study Research, Bob Stake characterized case studies as intrinsic, instrumental or collective. Intrinsic cases are those that are inherently interesting to a researcher, perhaps because of their uniqueness or peculiarity. Instrumental cases are those that researchers study because they have features connected to bigger concepts and that provide an empirical instance to study a bigger idea. Collective case study is looking at multiple cases often with a desire to compare and understand variation, in other words it is a collection of instrumental cases.

So far, we have case study as the investigation of bounded systems that we are motivated to investigate with three possible intentions. Sounds like a very important feature of research design.

But is it a methodology?

Research methodology is a framework that guides research practice ~ it is the theoretical frame that pulls epistemology forward into a discourse that further articulates the nature of knowledge and that guides our choice of methods. Crotty describes it as: “the strategy, plan of action, process or design lying behind the choice and use of particular methods and linking the choice and methods to the desired [research] outcomes.” Note that methodology is a theoretical framework. An important feature of methodologies is that they have substantive content, notions about what the focus of the research will be. If one does ethnography, for example, some notion of culture (even if adapted substantially from cultural anthropology) is central to the investigation. If one does critical research, some notion of power (and likely inequity) is central to the investigation. If one does narrative research, some notion of storying is central. So methodologies bring together salient, foundational social constructs with features of doing research. One could argue that certain methodologies logically entail the investigation of cases, in which case, case study could be an element of a methodology, but that it is only one among a number of elements.

So where does this leave case study as a methodology? What are the foundational social constructs that are central to it as a methodology? Here is where the logic breaks down. Looking at a case, in a natural setting, doing extensive data collection in natural settings doesn’t begin to hint at any particular foundational social constructs… inevitably researchers must draw on some other methodology for those. There is considerable variation in methodologies that inform what one does when investigating a case, and even when researchers do not articulate their methodology it lurks in the articulation of what those central social constructs are and the means by which we investigate them (reflection on pre-reflected experience in phenomenology; story telling in narrative analysis; culture in ethnography; and so on).

Terminology is inconsistent in discussions of research ~ there is a bewildering, often rolling sea of ideas, concepts, and practices to navigate in learning about research. There is no single ship of understanding, but thoughtful (re)articulation of the ideas underlying the theoretical and practical aspects of research is part of being in a community of social science researchers.

PS.

Further reading: This recently published comparative analysis of Yin, Stake & Merriam‘s take on what case study research is explicates their positions well, and I think still leaves unanswered the fundamental question of what case study is. My view of case study as a feature of research design is unchanged.

Learning to be a better researcher… by being a research participant

section_volunteerResearchers are egocentric… the research they do stems from their interests and motivations. That’s a given. Within interpretive research methodologies this egocentric position is tempered by the interpersonal intersection of researchers’ interests with those of their research participants. Mostly I am a researcher, but I have also been a research participant and these different vantage points provide useful lessons for thinking about how we conceptualize research participants’ engagement with us and how we treat those who participate in our research.

In general, the interests of the researcher are more important, and procedures to protect research participants are institutionalized through research review boards. For example, research participants’ identities are anonymized to encourage them to participate and protect them from harm, embarrassment, possibly even legal sanctions should the details of their lives become known. (Whether this makes sense is debatable ~  for a good discussion see Jan Nespor’s article Anonymity and place in qualitative research.)

But, equally important is the shield anonymity provides the researcher, shielding them from research participants’ challenges if and when they see how their experiences, thoughts, and emotions are used as data, as well as shielding them from professional critique since data sources (and often data) are held in secret (this is, in part, the current criticism leveled against Alice Goffman). I was able to read the research report of the study I participated in because I know how to access research, cultural capital that researchers might safely assume (hope?) most research participants don’t have.

Narratives of Research Participant Reaction

Here are three sketches of research participant reaction to a published account of the research in which they were involved. All are real. There are a number of lessons to be learned, but I will draw out just a couple.

1. In an introductory doctoral level interpretive research course students read Lois Weis’ Working Class Without Work, a study of an upstate NY high school. A student in the course had been a student in that high school and felt Weiss had understood well some of the gender related occupational aspirational conflict, but demonstrated misunderstandings of the role of the school in students’ future aspirations.

2. I read the dissertation, the report of the research I participated in, and found conversations between myself and the researcher that were not about the research focus had been included as data to corroborate evidence of my behavior and motivations as a mother. I was unaware of this until I read the completed dissertation as neither the analysis nor final dissertation were shared with the research participants.

3. A doctoral student used narrative inquiry to investigate parental experiences dealing with adult children with mental illness. His decision to use a first person story telling strategy was seen as misrepresentation by one parental participant who withdrew consent for including their data in the study.

Involving Research Participants to Get It ‘Right’

As researchers we want to get it as ‘right’ as we can and there are a number of strategies often invoked to do so. The most common are developing a habit of self-reflexivity (often manifest in journal writing), member-checking (an unfortunate term introduced by Guba & Lincoln), using key informants, and peer debriefing. These strategies are of value only if we as researchers practice them seriously.

Member-checking or informant feedback (the term I’ll use) is when researchers share research data, analysis or reports with participants to ensure categories, constructs, explanations and interpretations “ring true” and to explore what might be missing. This is a much talked about and seldom used strategy.

I suspect most researchers avoid informant feedback because it entails: the possibility of creating more, new data (like in sketch #1 and #2); invites potential disagreements and conflict with research participants (like in all three sketches); and invites research participants to opt out of the research study should they not like what they see or how they are represented (like happened in sketch #3). I also suspect most researchers do not see the value in the iterative process of multiple engagements with research participants in the space of data making and interpretation, seeing them as worthy of one but not multiple viewings (like in sketch #1 and #2). Within any interpretivist or critical research methodology this is a contradictory stance, and in establishing respectful relationships with research participants this is a disrespectful stance.

I suspect most researchers worry about the purpose, conditions and consequences of seeking informant feedback ~ can research participants give informed, useful feedback? should I change what I’ve written? does the research participant have veto power? will I destroy the rapport I have if research participants don’t agree with my analysis? Informant feedback is more likely to lead to confusion and contradiction than confirmation and clarity ~ a situation, which I have argued in delineating the process of triangulation, that ought to be invited not avoided by researchers. This is critically important.  If researchers use informant feedback they should think in advance how feedback will be used and to be transparent with research participants about that.

It is also critical to care about whether you are getting it as ‘right’ as possible, and that means not avoiding informant feedback because it may challenge your ideas. Regrettably, this avoidance is safeguarded by the conventions of anonymity and confidentiality that shield researchers from engagement with research participants who cede their right to engage publicly with the researcher by agreeing to those conventions ~ a Gordian knot.

Presentation of Self as Researcher

I suggest novice researchers develop a 30 second or so introduction of themselves as researchers to be used in the research context. A reminder to all that you are here as a researcher and what you are interested in… over the course of a research study it will be needed less and less but it should also be a reminder to self. Screen Shot 2015-07-07 at 12.49.20 PM If we have extended interactions with research participants it is easy to forget that we are researchers and not friends, neighbours, confidants, compatriots, lovers, and so on. We may become those things, but then we are no longer just researchers and the challenge of distinguishing what is within the parameters of the research becomes cloudy, and we may enter into a murky moral ambiguity of ethics and intimacy. (The classic example of this is Harry Wolcott’s relationship with Brad, summarized from his perspective in his book Sneaky Kid and It’s Aftermath.)

Researchers bear greater responsibility than research participants to maintain clarity about their role and their legitimate access to or defining what is data, what is within the boundaries of the research project. In sketch #2, some of my conversations with the researcher had nothing to do with the research topic. I gave this no thought at the time but implicitly assumed they were interactions between if not friends then perhaps a professor and doctoral student, the other most significant roles we played in relation to one another. Even as an experienced researcher, I was naive as a research participant and let the researcher in on too much of my life ~ even being ‘friends’ on Facebook during the research to then be unfriended upon completion of the research project. I mistook this gesture as ‘friendship’ when in hindsight it was data collection, albeit never negotiated as such with me.

Being clear that we are researchers should be ever present in our minds, but as researchers we need to insure it is ever present in research participants’ minds as well. “Anything you say [or do] can and will be used as data by me” might be more useful at protecting research participants than many of the promises institutionalized by research review boards.

We should remind ourselves…

narcissism should not prevail

and being a researcher is simply not a license to tell all.

 

 

Grounded Theory ~ a brief note

grounded-theory-new-4-638Few research methodologies have been so beguiling as grounded theory. Even when researchers are not working squarely within this methodology, many strategies for data analysis are liberally borrowed from the work of grounded theory methodologists. For those who claim to be doing grounded theory, the ground on which they stand has and continues to shift from the original positions proffered by Glaser & Strauss in the mid 60s. Most especially Glaser’s realist ontology has been left behind.

Certainly, Juliet Corbin pushed that process along in her 9780857029140collaborations with Anselm Strauss and more recently Kathy Charmaz has offered what she calls a constructivist grounded theory approach. On the latter, you can watch this interview with Charmaz that clarifies her view, one that I think captures the emerging contemporary stance on grounded theory.

The use of prior empirical and theoretical literature is a key idea in contemporary grounded theory… the old school notion of avoiding all prior knowledge has been largely set aside in favor of abductive reasoning that sees the researcher traveling back and forth between their empirical data and relevant literature.

Grounded theory has also taken a more narrative turn, emphasizing more literary writing and more acknowledgement of the centrality of the researcher to the analysis. Nonetheless, grounded theory remains unique in the search for a core category, that theoretical idea that integrates the various aspects of the theoretical explanation. (See, for example, this recent article The San Miguel Project.)

 

The role of empathy in research

Our ability to empathize is key to our success as researchers. To understand the human conditions, social relations, cultural meanings we must first understand and only after that to engage in the conversations of what if.imgres I recently came across a FB post by Kirstie Elgersma. I’ve extracted some parts of her post to illustrate what empathy is and, more importantly, is not:

A monologue on Empathy:
If someone said to me today,’Holy crap, Kristie. Life freakin’ sucks. I’m so broke. And I feel so scared. I’ve never been this far down before. I don’t know how I’m going to make it…’

– I would rarely say:
‘Well where do you live? I can help you find a job….’
That’s fixing it. And no one needs to be ‘fixed’.

– I would rarely IF EVER say:
‘I think you should figure out your life and get your shit together!’
That’s advising and unless they specifically asked for it, they aren’t needing that from me.

– I would rarely IF EVER say:
‘Ohhh shit! How could you let that happen?!?’
That’s interrogating… And they didn’t come looking for the Spanish Inquisition.

– I would rarely IF EVER say:
‘It’s only that way because you aren’t being positive or you aren’t trying hard enough…. You did this to yourself.’
They aren’t looking for an explanation.

– I would rarely IF EVER say:
‘Hey, you shouldn’t see it that way! There are many things in life to be thankful for.’
That would be me correcting them and they are not needing ‘correction’.

– I would rarely IF EVER say:
‘Well, there is much to be learned from this…’
They don’t need an education on what the fuck is happening to them.

– I would rarely say:
‘None of this is your fault.’
They don’t need me to console them.

– I would rarely IF EVER say:
‘Wellllll if you hadn’t quit that job you had then none if this shit might be happening right now…’
They don’t need me to evaluate their lives. And most people hate hearing,’I told you so’.

– I would rarely say:
‘Those FUCKERZ. The 1%. Fuck I wish they would die so we can all be free.’
They don’t need me to commiserate with them as much as I would want to. That serves no one.

– I would rarely IF EVER say:
‘Oh fuck you think you got it bad! You should hear what happened to this one guy I met… etc etc.’
That’s ‘one upping’ them. They don’t need to hear who has it worse than them. I could imagine that that would make people feel shitty for how they feel or feel shitty that they even shared anything with me in the first place.

– I would rarely say:
‘Hey, that reminds me of the time when I was in my early twenties….’
Telling tales distracts people from what is alive in them. And although that may be a great stall tactic it still serves nothing or no one but my furiously ‘telling tales’ thumbs.

– I would rarely say:
‘Omg! That’s so fucking awful!! I feel so bad for you!! What are you going to do?!?’
They don’t need sympathy from me either. Sympathy can dis-able a person further…..

empathy2

None of these responses is empathetic, although these responses are frequently confused with empathy. To use our capacity for empathy in research and in everyday life what is called for is to listen, to be curious, to try our very best to understand, and of course to care. Demonstrating empathy is never about telling anyone anything, it is about asking and seeking understanding. This is the first step researchers need to take before presuming they know both what the problem and its solution are.

Carl-Rogers-quotes-psychology-quotes-

Approaches to indepthness: ethnographic & longitudinal

As qualitative researchers there are different ways to think about how our research is indepth, seriously engaged with a research question and research participants. Two possibilities are immersing ourselves in a context (anthropological) or engaging with a context over a long time period (sociological) in order to deeply understand some social phenomenon.

Ethnography or ethnographic approaches are a methodology that draws us into a cultural context through participant observation over a sufficiently long enough period of time to formulate deep understandings of human experience. Living with, speaking with, and seeing from the inside out is the key to this indepth approach which is based on field work, participant observation, and interviewing. Other than the strategy of living within a cultural context, is the sociological strategy of longitudinal studies that continue engagement with research participants over a long period of time, sometimes referred to as a qualitative longitudinal study (QLS) which is based on indepth interviews with the same cohort of participants.

The same research question has been answered using both strategies. Two classic ethnographic studies that ask what happens to working class kids in schools are Paul Willis’ Learning to Labor and Jay MacLeod’s Ain’t No Makin’ It. Willis’ study looks at how British working class kids experience school so that class structures are perpetuated and MacLeod’s study looks at what happens to two groups of kids who differ by race but not class and their experiences of educational aspiration and schooling. MacLeod’s study is classic ethnography with a twist: twice after the initial research he went back to the Boston inner city neighbourhood to check in with his research participants and new editions of the book include appendices updating us about where the boys/men are at each point in time. MacLeod’s study might now be seen as a hybrid of the two approaches.

imgres

The Long Shadow is a good example of how understanding the same phenomenon, in this case framed as the transition of urban disadvantaged youth into adulthood, might be fostered by a long sustained relationship with research participants, i.e., knowing them for a longer time rather than intimately for a shorter time. What started out as a study to look at transitioning from home to grade one turned into a 25 yr longitudinal study of 800 low income inner city Baltimore kids.

These three studies are exemplars for researchers and implicitly offer lessons and advice on how to conduct indepth qualitative research in these two different ways.

imgres-1The QLS is useful for answering questions like:

  • what changes between T1 and T2 and Tn?
  • when does change occur?
  • are there epiphanies, tipping points, revelations?
  • what is missing over time? what is consistent over time?
  • what are the contextual factors related to change or stasis?

Click here for a short paper on the advantages/disadvantages, purposes and challenges of QLS and Johnny Saldaña’s Longitudinal Qualitative Research is another good resource.