Research Methods Literature ReviewPrior to beginning work on this assignment, review the qualitative and quantitative research designs encountered so far in this course.For your literature review, you

GUEST EDITORIAL Qualitative and Mixed Methods in Social Work Knowledge Development Deborah K. Padgett P roviding guidelines on qualitative and mixed methods in social work knowledge develop- ment is a daunting task.

Quantitative methods also require careful consideration, but they rarely entail the degree of epistemological self-searching and ongoing consequential decision making that qualitative methods demand. As a reviewer of qualitative studies for academic journals and federal funders, and as the recipient of many such reviews (some quite negative), I have learned some lessons along the way.

This editorial offers a few sugges- tions arising from these experiences that I hope will be of assistance to those interested in conducting qualitative research.

Qualitative methods have been contributing to knowledge development for a very long time— ethnography and field observation were around a century before the 20th century rise of quan- tification, with its emphasis on measurement and statistical analysis (Padgett, 2008).

Nevertheless, the codification of qualitative methods is a relatively recent development, beginning in the late 1970s and growing by leaps and bounds ever since.

Their embrace in social work came somewhat later than in education and nursing, but qualitative studies have since become commonplace in social work research, as evidenced by pubhcation of such studies in social work journals and by numerous presenta- tions at the annual conferences of the Society for SocialWork and Research and the Council on Social Work Education.

Without revisiting the paradigm wars that have consumed much time and energy, suffice it to say that disagreements about epistemology contribute to (but are not entirely responsible for) the lack of consensus regarding what is "good" versus what is "bad" qualitative research.

At the more construction- ist end of the epistemological continuum, standards tend to conform more to the humanities than to the sciences. At the other (postpositivist) end of the continuum, standards are not formulaic but are more concretely specifiable. This editorial hews closer to the postpositivist end but will hopefully resonate with social work researchers all along the continuum who wish to make their own contribu- tions to knowledge.

I will make seven points—both exhortations and recommendations:

The burden of proof is heavier but doable.

Choose an approach and stick with it.

Theories and concepts matter.

Social justice values do not have to be sidelined.

Research designs should be detailed and specific.

6. Writing the report: balancing description and interpretation.

Mixed methods require multiple inputs of expertise and effort.

Paying attention to these will not guarantee indi- vidual success in doing, publishing, and disseminating qualitative research, but it will likely help to raise standards (and the professional profile) of social work research in a broader sense. The distinction between what one does in a study and what one reports having done needs to be taken seriously, for there is too often a disconnect between these in qualitative research.

A superbly conducted study that is inadequately written up will not make its right- ful contribution and will hkely run into problems getting pubhshed.

THE BURDEN OF PROOF IS HEAVIER BUT DOABLE We live in a quantitative world, and adaptation to this reahty requires anticipation of skepticism (however unfair).

Elsewhere, I have promoted using one or more of six strategies as a means of strengthening a quahtative study's rigor (Padgett, 2008). These strategies are triangulation of data, peer debriefing and support, negative case analysis, maintaining an CCC Code: 0037-8046/09 $3.00 O2009 National Association of Social Workers 101 audit trail, prolonged engagement, and member checking.

The choice of which of these to deploy will depend on a study's goals and design, but, in general, the more used the better.

Rigorous qualitative research is accountable even if it follows flexibly applied guidelines.

Criteria used in quantitative research do not apply. Thus, most types of validity—internal, external, and measure- ment related—are not appropriate, but cultural or ecological validity may be on-target. Similarly, reliability and replicability are not suitable criteria because they imply fidelity and repetition rather than fluidity and uniqueness.

Lincoln and Guba's (1985) concept of trustwor- thiness is the most widely used global standard for adjudging qualitative studies.

A trustworthy study is one that is carried out ethically and whose findings represent as closely as possible the experiences of the participants. Because trustworthiness is not a matter of blind faith or glib assurances, the burden of proof is on the qualitative researcher to carry out the study as rigorously as possible and to faithfully give an account of what happened.

In one of the best published examples I have found. Morrow and Smith (1995) conducted a grounded theory study of women who survived childhood sexual abuse in which they used all six strategies for rigor and reported exactly how that was done.They also argued that their study demonstrated evidentiary adequacy by reporting on the breadth of the data: 220 hours of audio- and videotapes, 165 hours of interviews, 24 hours of group sessions, and 25 hours of follow-up interactions over a period of 16 months. Data for analysis exceeded 2,000 pages of transcriptions, field notes, and documents (Morrow & Smith, 1995).

It is uncommon to see a quahta- tive study that involves such an expansive effort.

Certainly, breadth should not be mistaken for depth, and this sort of evidentiary accounting cannot fuUy convey the intellectual and emotional engagement with the data that distinguishes qualitative research.

But it does help in enhancing trustworthiness.

CHOOSE AN APPROACH AND STICK WITH IT While conducting a meta-synthesis of 62 qualita- tive studies of women with AIDS, Sandelowski and Barroso (2003) received a rude awakening:

The studies were virtually unclassifiable.

For example, a "phenomenological" study used coding from grounded theory, and a "case study" was nothing more than a lengthy case description. Given the diversification in qualitative methods, one may choose to do ethnography, grounded theory, nar- rative analysis, phenomenological analysis, or case study analysis along with other, less-known ap- proaches. Each of these has, to varying degrees, a codified methodology that distinguishes it from the others.

The researcher is wise to adhere to one of these methods and to cite leading texts describing it.

One may mix qualitative approaches—for ex- ample, an ethnographer might carry out a grounded theory analysis of audiotaped interviews, and this could lead to separate studies or to mixing within the confines of a single study.

But any qualitative study should reveal a consistency and integrity of approach that is easily recognized by the reader and the reviewer.

3. THEORIES AND CONCEPTS MATTER Qualitative studies do not take place in an intel- lectual vacuum; theories and concepts are used to inform but not constrain. Grounded theorists, for example, refer to "sensitizing concepts" and the requirement that they earn their way (Charmaz, 2006) into the findings.

Some research topics tap into a deep reservoir of available knowledge, and others represent virtually uncharted territory.

Patricia Attia, a doctoral student I advised, chose to study Orthodox Jewish runaways, a problem unknown and unacknowledged when she began to develop her dissertation proposal.

Yet even this new area of interest could be linked to previous literature on runaway youths and eth- noreligious identity maintenance.

Ben Henwood, another doctoral advisee of mine, is interested in how case managers work with homeless people with serious mental illness.

His theoretical and conceptual foundations range from organizational theories to psychotherapeutic concepts such as the working alliance.

Qualitative studies may draw on several theo- retical frameworks at once.

They may also draw in new theories during analysis, and they may produce midlevel theories as part of their findings.

In our National Institute of Mental Health (NIMH)- funded study of homeless people with serious mental illness (Padgett, Hawkins, Abrams, & Davis, 2006; Padgett, Henwood, Abrams, & Davis, 2008), a priori theoretical lenses included empowerment (Friere, 1973), social ecology (Bronfenbrenner, 1979), and capabilities (Nussbaum, 1997) theories.

102 Social Work VOLUME 54, NUMBER 2 APRIL 2009 During analyses of the interview data, two additional theories were drawn in as a natural fit: feminist theory (Padgett et al., 2006) and Giddens' (1990) theory of ontological security (Padgett, 2007).

We also developed a grounded theory to explain a key outcome of interest—engagement and retention in services (Padgett et al., 2008).

Of course, there is the very real danger of theory and conceptual overkill crowding out the inductive thinking that makes qualitative studies uniquely valuable. It takes time and experience to get the balance right. Inductive thinking ensures that data are approached from a fresh perspective and that theoretical concepts are held lightly and discarded if not found to be relevant to the data.

4. SOCIAL JUSTICE VALUES DO NOT HAVE TO BE SIDELINED One of the primary dividing lines between quan- titative and qualitative methods has been the unapologetic embrace of social justice values by practitioners of the latter. Social work researchers are conversant in the language of empowerment and share a commitment to social welfare policies and practices that are equitable and humane. In public health, the rise in popularity of community-based participatory research (Israel, Eng, Schulz, & Parker, 2005) attests to the embrace of empowerment values by other professions.

Yet there persists a not-unfounded belief that socially conscious values are incompatible with rigorous research. Scientific review committees, concerned about bias, are prone to look askance at studies that appear to tilt more toward ideology than methodology.

This does not have to be an either-or situation.

In the NIMH study mentioned earlier, we drew on Freiré s (1973) empowerment theory and built the study around foregrounding consumer input and egalitarian relationships between researchers and study participants (Padgett & Henwood, in press).

The caveat, probably obvious by now, is that rigor- ous methods are vital even when social values are brought in to infuse a study with larger meaning.

After all, advocacy without empirical support is a far less credible stance.

RESEARCH DESIGNS SHOULD BE DETAILED AND SPECIFIC Flexibility, a hallmark of qualitative inquiry, does not mean that a study is haphazard or unsystematic Qualitative designs can be seen as road maps, with allowances made for detours and nonlinear progress.

That said, their development and implementation is an exercise in specification both before and after the fact. At the planning stage, several questions are addressed and answered. At the write-up stage, one describes what was done and why (with the understanding that detours were warranted and defensible).

A qualitative design typically entails description of the following: sample size, types of data to be collected, sampling and recruitment techniques (including inclusion/exclusion criteria and proce- dures for obtaining voluntary informed consent), data collection procedures, data management and analysis plans, and what strategies for rigor will be used.Virtually all qualitative studies use some form of purposive sampling, but under this rubric are a number of options—maximum variation sampling, criterion sampling, intensity sampling, and so on— that can be used. Qualitative data collection may be retrospective (as in life history interviews), or it may be prospective (multiple points of data col- lection proceeding over a period of time).

A study may revolve around group comparisons, or it may zero in on a specific population, entity, or event.

Describing one's design in detail does not preclude the qualitative caveat that flexibility will prevail over rigidity if the study's goals can thereby be better met (hence the permissible detours).

6. WRITING THE REPORT: BALANCING DESCRIPTION AND INTERPRETATION Qualitative researchers originally preferred book- length monographs but have long since come to rec- ognize that peer-reviewed journals are the preferred outlet in academia. A few journals—Qualitative Health Research, Qualitative Social Work, Qualitative Sociology—are dedicated to qualitative inquiry, but the vast majority are predominantly quantitative and thus must grapple with how to conduct fair reviews of qualitative submissions in the absence of prescriptive standards for quality.

Decisions about how to frame and present qualitative findings can make or break a study's publication prospects.

As alluded to in point 2, the ideal scenario involves maintaining a clear align- ment between one's choice of method and the explicit terminology used to describe the study and its procedures (if invoked, epistemology should also be compatible).The method section is criti- PADGETT / Qualitative and Mixed Methods in Sociai Work Knowledge Development 103 cal.

Here, crisp, factual, and thorough specification pays off, including description of procedures for recruitment and consent, sample size, number of interviews per participant, length of data collection period, training and supervision of interviewers, and so on.

Acknowledgment of a study's strengths and limita- tions is an area rarely discussed in qualitative research.

In most instances, one need not apologize for the sample size, and a qualitative study is not suspect because it "lacks generalizability,"At the same time, the depth and intensity of data may leave something to be desired (for example, if there was only one interview per participant or some strategies for rigor were not used even though they would have been appropriate). Explanation of what is meant by "saturation" may be necessary to familiarize readers (and reviewers) with this concept as the guidepost for knowing when to end data collec- tion and analysis.

Presentation of a study's results varies by approach but typically involves conceptual fmdings along with direct quotations,This mix of description and interpretation is a delicate balance; too much of the former makes the study appear simplistic, and too much of the latter makes it appear contrived (Creswell,2007, suggested ratios of 70/30 or 60/40, favoring description).

Balancing interpretation and description entails considerations of "voice," As my colleague Deb- bie Gioia noted in a recent e-mail exchange, "it's about voice (researcher and participant) and not just quotes," Thus, too heavy reliance on direct quotation does not necessarily honor what participants said, even as it compromises a study's ultimate contribu- tion to knowledge.

To be sure, qualitative methods offer a degree of interpretive latitude that is daunting, especially for the novice researcher. Finding one's voice requires self-discipline and constant referenc- ing of the data, 7. MIXED METHODS REQUIRE MULTIPLE INPUTS OF EXPERTISE AND EFFORT Mixed methods are rising in popularity, yet their design and conduct require careful consideration (see Creswell, 2003, for guidance on designing mixed-methods studies). The required expertise in quantitative and qualitative methods does not necessarily rule out the solo investigator, but this approach is far more plausible for larger resourced studies and teams of investigators.

Decisions need to be made and specified regard- ing whether the mixing is to be done sequentially or concurrently and whether it will be qualitative- dominant or quantitative-dominant. The choice of which methods to mix depends on compatibility and portability.

Thus, while focus groups are a popular choice among quantitative-dominant researchers, ethnography is far less commonly adopted (or feasible). Similarly, quahtative-dominant research- ers may use scaled measures as a small component of in-depth interviews, but they are not likely to incorporate a large-scale survey into their mixed- methods study.

As might be expected, there are serious constraints attendant upon the writing and publishing of mixed- methods research—-journals and grant funders do not allocate extra space for such efforts. Perhaps the most daunting challenge is integrating findings from the two "sides"; it is far easier when the two sets of findings corroborate or complement each other than when they conflict (Padgett, 2008), Despite their demands, mixed-methods studies present unique possibilities for synergy and knowledge growth that mono-method studies cannot match, CONCLUSION This editorial represents a contribution to what is already a lively dialogue in social work research.

The stakeholders are many: students, experienced researchers, journal editors, and reviewers, to name only a few. There are other topics on the horizon deserving of attention. Secondary analysis of quali- tative data, for example, is becoming increasingly popular and deserves attention as distinct from quantitative secondary analysis (Thorne, 1998),The rise of evidence-based practice, with its emphasis on systematic reviews, presents unprecedented chal- lenges to those who seek to synthesize knowledge through aggregate reviews of qualitative studies.

Last but not least, qualitative social work researchers continue to have epistemological differences that undedie questions about the borderland between practice and research.

In conclusion, the development and expansion of the knowledge base in social work research is a dynamic enterprise that depends on contributions from diverse empirical methods. Shared commit- ment to transparency and rigor unites quantitative and qualitative methods even as their respective strengths are complementary and necessarily distinct, BUH 104 SocialWork VOLUME 54, NUMBER Z APRIL 2009 REFERENCES Bronfenbrenner, U. (1979). The ecology of human develop- ment:

Experiments hy nature and design.

Cambridge, MA: Harvard University Press.

Charmaz, K. (2006).

Constructing grounded theory.

Thousand Oaks, CA: Sage Publications.

Cresvvell,J.W. (2003).

Research design:

Qualitative, quantita- tive, and mixed metltods approaches (2nd ed.).Thousand Oaks, CA: Sage Publications.

Creswell,J.W. (2007).

Qualitative inquiry and research design (2nd ed.).Thousand Oaks, CA: Sage Publications.

Friere, P.

(1973).

Pedagogy of tlie oppressed (M.

13.

Ramos, Trans.), New York: Seabury Press.

Giddens,A. (1990). Vte consequences of modernity Oxford, England: Polity Press.

Israel, B.A.,Eng, E., Schulz, A.J., & Parker, B.A. (Eds).

(2005).

Methods in community-based participatory research for health.

San Francisco:Jossey-Bass.

Lincoln,Y. S., & Cuba, B. G. (1985).

Naturalistic inquiry.

Beverly Hills, CA: Sage Publications.

Morrow, S, L., & Smith, M. L. (1995). Constructions of survival and coping by women who have sur- vived childhood sexual abuse.JoiirHd/ of Counseling Psychology, 42, 24-33.

Nussbaum, M. (1997), Capabilities and human rights.

Fordliam Law Review, 66, 273—300, Padgett, D. K. (2007).There s no place like (a) home:

Ontological security in the third decade ofthe "homelessness crisis" in the United States.

Social Science & Medicine, 64, 1925-1936.

Padgett, D. K. (2008), Qualitative methods in social worb research (2nd ed,).Thousand Oaks, CA: Sage Publications.

Padgett, D. K., Hawkins, ft.

L., Abrams, C, & Davis, A.

(2006), In their own words:Trauma and substance abuse in the lives of formerly homeless women with serious mental illness. American Journal of Ortlwpsychiatry, 16, 461-467.

Padgett, D. K., Henwood, B., Abrams, C, & Davis, A.

(2008).

Engagement and retention in care among formerly homeless adults with serious mental illness:

Voices from the margins.

Psychiatric Rehabilitation Journal, 31, 226-233.

Padgett, D. K., & Henwood, B.

(in press). Obtaining large-scale funding for empowerment-oriented qualitative research:

A report from personal experi- ence.

Qualitative Health Research.

Sandelowski, M., & Barroso,J. (2003).Writing the proposal for a qualitative research methodology project.

Qtialitative Health Research, 13, 781-820, Thorne, S. (1998). Ethical and representational issues in qualitative secondary analysis.

Qualitative Heahh Research, 8, 547-555.

Deborah K. Padgett, PhD, MPH, is professor.

Silver School of Social Work, New York University, Ehrenkranz Center, 1 Washington Sqttare North, New York, NY 10003-6654; e-mail:

dehorah .padgett@nyu. edu.

NASW PRESS POLICY ON ETHICAL BEHAVIOR T he NASW Press expects authors to ad- here to ethical standards for scholarship as articulated in the NASW Code of Ethics and Writing for the NASW Press:

Information for Authors.

These standards include actions such as • taking responsibility and credit only for work they have actually performed • honestly acknowledging the work of others • submitting only original work to journals • fully documenting their own and others' related work.

If possible breaches of ethical standards have been identified at the review or publication process, the NASW Press may notify the au- thor and bring the ethics issue to the attention ofthe appropriate professional body or other authority.

Peer review confidentiality will not apply where there is evidence of plagiarism.

As reviewed and revised by NASW National Committee on Inquiry (NCOI), May 30, 1997 Approved by NASW Board of Directors, September 1997 PADGETT / Qualitative and Mixed Metboé in Social Work Knowledge Development 105