184 Designing Reseorch

in the proposal that convey a sen se of the overall activities of qualitative

data analysis, su eh as the following drawn from my own thaughts
(Creswell, 20(7) and those af Rossman and Rallis (1998):
lt is an ongoing process involving continual reflection about
the data, asking analytic qucstions, and writing mClllOS throughout the
study. 1 say that qualitative data analysis is conducted coneurrently with
gathering data, making interpretations, and writing reports. While inter-
vicws are going on, ror exanlple, the rescarcher 111ay be analyzing an
interview collected carlier, writing memos that may ultimately be
included as a narrative in the final report, and organizing the structure of
the final reporl.
Data analysis involves coUecting open-ended data, based on asking
general questi0118 and developing un analysis froID the information sup-
plied by participants.
orten we see qualitative data analysis reportcd in journal articles and
books that is a gcneric fornl 01' analysis. [n this approach, the researcher
colIects qualitalive data, analyzes it for themes or perspectives, ancl reports
4-5 themes.! consider this approach to be basic qualitative analysis; today
111any qualitative researchers go beyond ihis generic analysis to add a pro-
cedure within one of the qualitative strategies of inquiry. For example,
grounded theory has systematic steps (Corbin & Strauss, 2007; Strauss &
Corbin, 1990, 1998). These involve generating categories of inl"ormation
(open coding), selecting onc of the categories and positioning it within a
theoretical model (axial coding), and then explicating a story from the
interconnection of these categories (selective coding). Case stw/!f ancl et/1110-
gmplzie researe/ involve a detailed description of the setting or individuals,
fallowed by analysis of the data for themes or issues (see Stake, 1995;
Wolcott, 1994). Plzellonzellologieal researel, uses the analysis oi" significant
statclnents, the generation o' lneaning units. and the dcvelopn1cnt of
what Moustakas (1994) calls an essence description. Narrative researell
employs restorying the participants' stories using strue!ural devices, such
as plot, setting, activities, climax, and clenouement (Clandinin & Connelly,
20(0). As these examples illustrate, the processes as well as the terms dif-
fer from one analytic strategy to another.
! Despite these analytie differences depending on the type of strategy
used, qualitative inquirers oftcn use a general procedure and convey in the
proposal the steps in data analysis. An ideal situation is to blend the
general steps with the specific research strategy steps. An overview of
the dala analysis process is seen in Figure 9.1. As a it'cea:rch ti!!" 1 urge
researchers to look at qualitative data analysis as following steps from the
specillc to the general and as involving multiple levels of analysis.
Valldating the
Accuracy of the
Information
Qualitative Procedures
Interpreting the Meaning of
Themes/Descriptions
Interrelating Themes/Description
(e.g., grounded theory, case study)
Coding the Data
(hand or computer)
Reading Through AH Data
Organizing and Preparing
Data for Analysis
Raw Data (transcripts,
fieldnotes, images, etc.)
Figure 9.1 Data Analysis in Qualitative Research
This fignre suggests a linear, hierarchical approach building li'om the
bottom to the top, but 1 see it as more interactive in practice; the various
stages are interrclated and not always visited in the arder pl'esented. These
levels are emphasized in the foUowing steps:
Slepl, Organize am! prepare the data for analysis. This involves tran-
scl'ibng intel'views, optically scanning 111ateral, typing up field notes, ol'
sorting and arranging the data into different types depending 011 the
sources of infarmation.
Slcp 2. Read through aU the data. A first step is to obtain a general sense
of the information ami to reflee! on its overal! meaning. What general
ideas are participants saying? What is the tone 01" the ideas? What is the
impression 01" the overall depth, credibility, and use 01" the information?
Sometimes qualitative reseal'chers write notes in margins or start record-
ing general thoughts about the data at this stage.
185
186 Designing Research
Stcp 3. Begin deLailed analysis with a coding process. Cuding is the
process of organizing the material into chunks or segments of text belre
bringing meaning to inf,mnation (Rossman & Rallis, 1998, p, 171), It
involves taking text data or pietures gathered dnring data collection, seg-
l11cnting sentences (or paragraphs) Of in1ages inLo categories, and labeling
those categories with a term, ofren a term based in the actnallangnage al'
the participant (called an in vivo term),
Befare proceeding to Step 4, eonsider some remarks that will provide
detailed guidance f<lr the coding proeess, Tesch (l990, pp, 142-145) pro-
vides a useful analysis oi' the proeess in eight steps:
1, Gel a sense of the whole, Read all the transcriptions earefully,
Perhaps iot down some ideas as they come lo rnind,
2, Piek one doeument (i.e" one interview)-the most interesting one,
the shortest, the one on the top of the pile. Ca through it, asking
yourself, "Whal is this about?" Do not think aboul the substanee of
the information but its underlying meaning. Wrile thoughts in the
margin,
l. When you have completed this task for several participants, make a
list of all tapies. Cluster together similar topies. Form these tapies
inLo colUlnns, perhaps arrayed as major topics, unique topics, and
lelrovers.
-l. Now [ake this list anel go back to your data. Abbreviate the topics as
codes and write the codes next to the appropriate segments al' the
text. Try this preliminary organizing schen1e to see if new categories
and codes cn1erge.
'). Find the most descriptive wording for your tapies and turn them into
calegories. Look fr ways of reducing your totallist 01' eategories by
grouping tapies that relate to each ather. Perhaps draw lines between
your categories to show interrelationships.
h> Make a final decison on the abbreviation for each category and
alphabelize lhese codeso
Assemble the data material bclonging to each category in onc place
and perforro a preliminary analysis.
g. If necessary, recode your existing data.
These eight steps engage a researcher in a systematic process of analyz-
ing textual data. Variations exist in this process. As a reseril.E',ch Op,
1 encouragc qualitative rcsearchers to analyze thcir data for material that
can address the Ic)lIowing:
@; Codes on tapies thal reaclers would expect to flnd, based on the past
literaturc and common scnse
Qualllullve Procedures
@; Codes that are surprising and that were not antieipated at the begin-
ning of the study
@; Codes that are unusual. and that are, in and al' themselves, of con-
ceptual interest to readers (e.g., in Asmussen and Creswell, 1995, we iden-
lified retriggering as one of the codeslthemes in the analysis that suggested
a new dimension for us to a gunman incident on campus and that seelued
to conneet with experienees of others 011 campus)
@; Codes that address a larger theorelieal perspective in the researeh
As an alternative conceptualization, consider the Iist by Bogdan and
Bilden (1992, pp. 166-172) of the types of codes that they look for in a
qualitative database:
@ Setting and eontext codes
@; Perspectives held by subjects
@; Subjects' ways 01' thinking about people and obiects
@; Process eodes
Aetivity codes
@ Strategy codes
@; Relationship and social stmeture codes
Preassigned eoding schemes
One further issue about coding is whether the researcher should
(a) develop codes o/JI!1 on the basis of the emerging infonnation collecled
frorn participants, (b) use predetermined eodes and then fit the data to
them, or (c) use some combination of predetermined and emerging eodes.
The traditional approach in the social sciences is lo allow the codes to
emerge during the data analysis. In the health sciences, a popular
approaeh is to use predetermined codes based on the theory being cxam-
ined. In this case, the researchers might develop a """mav" n"l"h""k,
atable or record that contains a list of predetermined eodes that researchers
use for coding the data. This eodebook might be composed with the names
of eodes in one colurnn, a deflnition al' codes in another eolumn, and then
specillc instances (e.g., line numbers) in which the code was iound in the
transcripts. Having sueh a codebook is invaluable when multiple
researchers are coding the data from different transeripts. This eodebook
can evolve and change during a study based on clase analysis of the data,
even when the researcher is not starting from an emerging eode perspec-
tive. Por researchers who have a distinct theory they want to test in their
proieets, 1 would reeommend that a preliminary eodebook be developed for
coding the data and permit the codebook to develop and change based on
187
188 Designing Research
the information learned during the data analysis. The use of a codebook is
especiaUy helpful for ficids in which f/lIantitative research dominales am! a
n10re systematic approach to qualitative research is needed.
Returning lo lhe general coding process, some researchers have found
it use fuI to hand eocle qualitative transcripts ar nformation, sometimes
using color codc schemes and lo cut and paste lext segments onlo note
canls. This is a labarious am! time-consuming appraach. Others tene! to
use qualitative COlllputer software progrmns to hclp code, organize, and
sort informalion lhal will be useful in writing the qualitative study. Several
excellent computer software programs are available, and they have similar
features: gaod tulorials and demonstration CDs, abilily to incorporate bolh
texl and image (e.g., pholographs) dala, the feature of sloring am! orga-
nizing e!ala, lhe search capacity of locating aH texl associatee! wilh specific
coe!es, inlerrdaled codes for making queries of the relationship among
codes, am! the imporl and export of qualilative data to '1l/(mtitative pro-
grams, such as spreadsheets ar data analysis programs.
The basic idea behind these programs is lhat using lhe computer is an
efficient means for storing ami locating qnalitative data. Although lhe
researcher still needs to go through each line of text (as in transcriptions)
and assign codes, this proccss may be raster and nlOre efficient than hand
coding. Also, in large dalabases, lhe researcher can qnickly locate al! pas-
sages (or lexl segments) coded the same am! determine whether partici-
pants are rcspnding lo acode idca in sil11ilar or different ways. Beyond
lhis, lhe computer program canlcilitale comparing e!iI'ferenl codes (e.g.,
How do males and females-lhe tlrsl code 01' gender-differ in lerms of
their altitudes to smoking-a second code) These are just a few I'ealnres of
the software progranls that 111ake them a logical choice for qualitative dala
analysis over hand coding. As with any software program, qualitative
soflware programs require time ane! skill to learn am! employ efl'ectively,
although books for learning the programs are widely available (e.g.,
Weilzman & Miles, 1995).
Mosl 01' lhe programs are available only on the PC plattClrm. The com-
puter software prognnl1s that lUy sLa1'1' and [ use in my research office are
lhese:
@ MAXqda (http://www.maxqda.com/).This is an excellenl PC-based
program ihnn Germany that helps rescarchers systenlatically evaluale
and interpret qualitative texls. It has aH of the features mentioned earlier.
@ Atlas.ti (http://www.atlastLcom).This is anolher PC-based pro-
gnllll fro111 Gernlany that enables a rescarcher to organize text, graphic,
audio, and visual data 111es, along with coding, men10S and findings, into
a project.
iII QSR NVivo (http://www.qsrinternational.com/) This program, from
Australia, featurcs lhe popular soflware program N6 (or Nud.ist) and NVivo
concepl mapping in eombination. It is available only for Windows PC.
Qualitative Procedures
@ HyperRESEARCH (http://www.researchware.com/).This is a pro-
gram availahle for either lhe MAC or PC. It is an easy-to-use qualitative
software package enabling users to code, retrieve, build lheories, and con-
ducl analyses of lhe dala.
Slcp 4. Use the coding process lo generale a deseriptioIl 01' the setting or
peoplc as well as categories or themes for analysis. Descriptiol1 involves a
detailed rendering 01' information about people, places, or events in a set-
ting. Researchers can generate codes for this description. This analysis is
useful in designing delailed descriptions fm case studies, ethnographies,
and narrative research projects. Then use the cocling to gcnerate l s111a11
nunlber of themes Of categories, perhaps five to seven categorics for a
research study. These thelnes are the ones that appear as 111ajor findings in
qualitative studies and are ol'len lIsed lo create headings in the findings
sections 01' sludies. They should display multiple perspeetives fmm individ-
uals am! be supported by diverse qnotations am! specific evidence.
Beyond identifying lhe themes during the coding process, qualitative
researchers can do much wilh themes to build additionallayers of com-
plex analysis. Por examplc, researchers interconnect thclnes into a story
line (as in narratives) or develop them inlo a thcorctical model (as in
grounded lheory). Thernes are analyzed for each individual case and
across different cases (as in case studies) or shaped into a general descrip-
tion (as in phenornenology). Sophistcated qualitative sludies go beyond
deseription and theme identification and inlo complex theme eonnections.
Stcp 5. Advanee how the description ami themes will be representa! in
the qualitative narrative. The most popular approach is to use l narrative
passage to convey the findings 01' the analysis. This might be a discussion
lhal mentons a chronology of cvenls, the e!etailed discussion 01' several
themes (complete wilh subthemes, specific illustrations. multiple perspec-
tives fro111 individuals. and quotations) or l discussion with interconnect-
ing themcs. Many qualitative researchers also use visuals, figures, or tables
as adjuncts lo the discussions. They present a process mode! (as in
grounded theory), ae!vance a drawing 01' lhe specillc research sile (as in
ethnography), or convey descriptive inforrnation ahout each participanl
in a table (as in case studies and ethnographies).
Stcp 6. A final step in data analysis involves making an interpl'ctation
or meaning 01' lhe data. Asking, "What were the lessons learned?" cap-
lures the essence of this ielea (Lincoln & Cuba, 1985). These !esson" could
be the rescarcher's personal inlerpretation, couched in lhe understanding
lhat the inquirer brings lo lhe study from her or his own culture, history,
and experiences. It couId aIso be a meaning derived fro111 a comparison 01'
the llndings with information gleaned lrom lhe litemtllre or t"eories. In lhis
way, authors suggest thal lhe findings confirm pasl information or diverge
from it. It can also suggesl new 'lllestions thal need to be asked-questions
189
j
190 Designing Reseorch
raiscd by thc data and analysis that the inquirer had 110t foreseen earlier in
lhe sludy. One way elhnographers can end a sludy. says Wolcotl (1994).
is to ask furthcr questions. 1'he questioning approach is also used in
advocacy and participatory approaches to qualtative rescarch. Moreover.
when qualitative researchers use a theoreticallens, they can form inter-
pretations that call ror action agendas ror reform and change. Thus, inter-
pretation in qllalitative research can take lTIlny forms, be adapted rOl"
differenl types 01' designs. and be flexible lo convey personal, research-
based, and action meanings.
RH.IA!m.HV, V.i,,'UL1
Although validation 01' findings occurs throughout the steps in the process
01' rescarch (as shown in Pigure 9.1). this discussion focuses on it to enable
a researcher to write a passage into l proposal on the procedures ror vali-
dating lhe findings thal will be underlaken in a sludy. Proposal developers
need to convey lhe steps they wiU take in lheir sludies lo check for the
accuraey and eredibility 01' their findings.
Validity does not carry the same connotations in qualitative research as
it cIoes in quantitative research, nor is it a companion of reliability (cxam-
ining stability or consistency 01' responses) or generalizability (lhe external
validity of applying results to new settings. people. or samples; both are dis-
cussccl in Chapter 8). means lhat lhe researcher
checks I"r the aeeuraey 01' the Ilndings by employing certain procedures.
while indica tes that the rcseareher's approach is
consistent across differcnt researchcrs and dil'ferent projects (Gibbs, 20(7).
How do qualitative researchers check to determine if their approaches
are consistent or reliable? Yin (2003) suggesls thal qualilative researchers
neccl to doeumcnt the proeedures 01' lheir case sludies and lo document as
many 01' the steps 01' lhe procedures as possible. He also recommends set-
ting up a detailed casc study protocol anel database. Gibbs (2007) suggesls
several
@ Check lranscripls to make sure that they clo not eontain obvious mis-
takes macIe during transcription.
, Make sure lhat there is not a drift in lhe definition 01' cocles. a shift in
the meaning 01' tbe codes during the process 01' coding. This ean be accom-
plished by conslantly comparing dala with the codes anel by writing
memos aboul the codes ancl lheir definitions (see the discussion on a qual-
itative codebook).
@ For tcam research. coordnate the cOlnmunication among the cocIers
by regular clocumentecl meetings and by sharing the analysis.
@ Cross-check cocles developed by ditTerent researehers by eomparing
results that are inclependently clerivecl.
Quolitotive Procedures
Proposal writers need lo inclucle several 01' these procedures as evidence
lhal they will have consistent results in their proposed sludy. 1 recommend
that several procedures be menlioned in a proposal and that single
researehers find another person who can cross-check their codes, for what
r call ag"(,e"",,,1 (or cross-checking). Such an agreemenl
mighl be based on whether 11'.'0 or more eoders agree on codes used for the
same passages in the lext (it is not thal they code the same passage 01' texl,
but whether another codel' would code it wilh the same or a similar code).
Slatistical procedures or reliabilily subprograms in qualitative computer
software packages can lhen be used to determine the level 01' consistency
of coding. Miles and Huberman (1994) recommend that lhe consistency
01' the coding be in agreement at least 80% of the time for good qualitative
reliability.
Validity. on lhe other hand, is one 01' the strengths of ql1alitative
research. ancl il is based on delermining whether the findings are accurale
from the standpoint 01' the researcher. lhe participant. 01' the readers of an
account (Creswell & Miller. 2000). Terms abound in the qualitative lilera-
ture lhat speak lo lhis idea. sueh as trustworthiness, autllentieit!!. and ered-
ibility (Creswell & Miller, 2(00). ancl it is a ml1eh-discussed topie (Lineoln &
Guba. 2(00).
A procedural perspective lhal r recommend I"r research proposals is lo
identify amI discuss one or more strategies available to check lhe accuracy
of the findings. The researcher actvely incorporates vaUidH:y t.rratcgies
,. inlo their propasa!. 1 recommend the use 01' mulliple strategies. and these
should enhance lhe researcher's abilily to assess lhe accuracy of findings
as well as convince readers 01' that accuracy. There are eight primary
strategies. organized from thosc most frequently used and easy lo imple-
menl to those oeeasionally used ancl more diftleult to implement:
@ TrianlluZate different data sourees of information by examining evi-
dence from the sources ancl using it to build a coherent justification for
themes. If thernes are established based on converging several sources of
data or perspectives from participanls. then this process can be claimed as
adding to the validity of lhe study.
@ Use memIJer chec1dng to determine the accuracy of the qualitative
Ilndings through taking lhe Ilnal report or speeific descriptions or lhemes
baek to participants and determining whether these participants feel thal
they are aecurale. This does not mean taking back lhe raw transcripts lo
check for accuraey; instead. the researcher takes baek parls of the polished
producl, sueh as the themes. the ease analysis. the grounded theory. the
cullural clescription. amI so forth. This procedure can involve eondueting
a follow-up interview with participants in the sludy amI providing an
opporlunily fol' lhem to comment on the fmdings.
@ Use rieh. tlliek description to convey the findings. 1'his description may
lransport readers to the setting and give the discussion an element 01'
191
192 Designing Research
shared experiences; When q ualitative researchers providc detailed descrip-
tions 01' lhe setting, for example, or provide many perspectives about a
lheme, the results become more realistic and rieher. This procedure can
add to the validity of the findings.
i!> Clarify the bias the researeher brings lo lhe study. This
creates an open and honest narrative that will resonate well with readers.
Refleetivity has been mentioned as a core characteristic 01' qualilative
rescarch. Goad qualitative rcsearch contains COl1Ullcnts by the rcscarchers
about how lheir inlerprelation 01' the nndings is shaped by their back-
ground' su eh as their gender, culture, history, and socioeconomic origino
i!> Abo presenl negative or discrepan! inJ(mnation lhal runs eounter to
the themes. Because reallife is composed of differenl perspeetives lhat do
not always coalesce, discussing contrary information adds to the creclibil-
ity 01' an account. A researcher can accon1plish this in discussing evidence
about a theme. Most evidence will build a case for the theme; researchers
can aIso prcsent infonnation that contradicts the general perspective of
lhe lheme. By presenling lhis contradictory evidenee, the aecount beco mes
more realstic and hence valido
i!> Spend pmlollged time in the field. In lhis way, the researeher develops
an in-depth underslanding of the phenomenon uuder sludy aud can eon-
vey detail about the site and the people tbat lends credibilily to lhe narra-
tive account. The 1110re experience that a researcher has with participants
in their actual setting, tbe more accurate or valid will be the findings.
i!> Use peer delriejillfl lo enhanee the aeeuraey 01' the aeeount. This
process involvcs locating a person (a peer debriefer) who reviews and asks
questions aboul the qualilativc sludy so thal lhe aeeount will resonate
with people other than lhe researcher. This strategy-involving an inter-
pretation beyond the researcher and invested in anolher person-adds
validity to an account.
Use an externaI auditor to review the entire project. As distinct from tI
peer debrieler, tbis auditor is not familiar with the researeher or the projeet
amI can provide an objective assessment 01' the project throughollt the
process of research or at thc conclusion af the study. The role is similar to
that of a fiscal auditor, ami specinc questions exist that auditors might ask
(Lineoln & Guba, 1985). The proeedure of having an independent investi-
galor look over mauy aspeets 01' the project (e.g., aeeuraey af transcrip-
tion. the relationship betweeu the research questions and the data, the
level of data analysis from the raw data through interpretation) enhances
the overall validily 01' a qualilative study.
QuaHvaUve generaib:aHon is a tenn that is used in a limited way in
qualitative research, sinee the inlenl 01' this form of inquiry is nol to
QUCllilatlvo ProCoclurOli
generalize findings to individuals, sites, or plaees oulside 01' those under
sludy (see Gibbs, 2007, for his cautionary note aboul qualitative general-
izability). In fact, the value of qualitative researeh lies in lhe particular
deseription and themes developed ill cOlltext of a specinc site. Particularity
rather than generalizabWty (Greene & Caracelli, 1997) is the hallmark of
qualitative research. However, there are a 1'ew discussions in the qualita-
tive literature about generalizability, especially as applied to ease study
researeh in whieh the inquirer studies several eases. Yin (2003), for exam-
pIe, feels that qllalitative case study results can be generalized to some
broader theary. 'fhe generalization occurs when qualitative rcsearchers
study additional cases amI generalize findings lo the new cases. It is the
smne as the replicllton JaBie used in experimental rescarch. Howcver, to
repeat a case study's findings in a ncw case setting rcquires good docu-
rnentation 01' qualitative proccdures, such as a protocol 1'or dOCU111enting
the problem in detail amI tbe development 01' a thorough case study data-
base (Vin, 20(3).
THE QUAUTATlVE WRmA,lp
A plan for a qllalitative procedure should end wilh some comments about
the narrative that emerges from the data analysis. NumerOllS varietics of
, narratives exist, amI examples from scholarly jOllrnals illustrate models.In
a plan ror a study, consider advancing several points about the narrative.
The basic procedure in reporting the results 01' a qualitative study are to
develop e!escriptions and themes fram the data (see Figure 9.1), to present
these descriptions amI lhemes that convey multiple perspectives 1'rom par-
ticipants amI detailed descriptions 01' the setting or individuals. Using a
qualitative strategy of inquiry, these results may also provide a chronolog-
ical narrative ol' an individual's lite (narrative research), a detailed descrp-
tion 01' their experienees (phenomenology), a theory generated I'rom lhe
data (grollnded theory), a detailed portrait 01' a culture-sharing group
(ethnography), al' an in-deplh analysis 01' one or more cases (case sludy).
Civen these different strategies, the findings ane! interpretation sections
of a plan Illr a study might discuss how lhe seelions will be presented: as
objective accounts, fieldwork experiences (Vau Maanen, 1988), a chrouol-
ogy, a process' lnodel, an extended story, an analysis by cases or across
cases, or a detailed descriptive portrait (Creswell, 2(07).
At the specifie level, so me wrHing ,trteg;cs might be as follows:
i!> Use quotes amI vary theil' length from short to long embedded passages.
$) Script cOl1versation and report the conversation in dif1'erent langl1ages
lo reneet cultural sensitivity.
194 Designing Research
Present text information in tabular [orrn (e.g" matrices, comparison
tables of different codes).
'i!l Use the wording from participants to form codes and theme labels.
i!! lntertwine quotations with (the author's) interpretalions.
i!! Use indents 01' other special fi::>rmatting of the manuseript to eaU
attention to quotations from participants.
i!! Use the first person ''l'' 01' coUective "we" in the narrative formo
i!! Use metaphors amI analogies (see. for example. Richardson. 1990.
who discusses sorne of these ()rms).
i!! Use the narrative approaeh typicalIy used within a qualitative strat-
cgy 01' inquiry (e.g., dcscription in case sludics and cthnographics, l
detailed story in narrativc research),
~ Describe hol'! the narrative outcome will be compared with theories
and lhe general lileralure on the topie. In Illlny qualitative arUcles,
researchers disc:uss the lileratllre at the end of the stlldy (see the discussion
in Chapler 2).
Qualitative Procedures
The following is an example 01' l qualitative procedure written as part of a
doctoral proposal (Miller. 1992). Miller's project was an ethnographic study
of first-year experienees of the president of a 4-year eoUege. As 1 present this
diseussion. 1 refer baek to the seclions addressed in this chapter and highlight
them in boldfaced type. Also. 1 have maintained Miller's use of the term in/or-
Inant. although today. the more appropriale termo participilIlt. should be used.
The Qualitative Research Paradigm
The qualitative research paradigm has its roots in cultural anthropology and
American sociology (Kirk & Miller. 1986). It has only recently been adopted
by educational researchers (Borg & Gal!. 1989). The intent 01 qualitative
research is to understand a particular social situation. event. role. group. or
interaction (Locke. Spirduso, & Silverman, 1987). It is largely an investigative
process where the researcher gradually makes sense of a social phenome-
non by contrasting. comparing. repllcating. cataloguing and classifying the
object of study (Miles & Huberman. 1984). Marshall and Rossman (1989) sug-
gest that this entails immersion in the everyday lile 01 the selting chosen lor
the study; the researcher enters the inlormants' world and through ongoing
interaction. seeks the inlormants' perspectives and meanings. (Qualitative
assumptions are mentoned,J
Qualitative Procedures
Scholars contend that qualitative research can be distinguished from quan-
titative methodology by numerous unique characteristics hat are inherent
in the designo The following is a synthesis of commonly articulated assump-
tions regarding characteristics presented by various researchers.
1. Qualitative research occurs in natural seltings. where human behavior
and events occur.
2. Qualitative research is based on assumptions that are very different Irom
quantitative designs. Theory or hypotheses are not established a priori.
3. The researcher is the primary instrument in data collection rather than
some inanimate mechanism (Eisner, 1991; Frankel & Wallen. 1990; Lincoln
& Guba. 1985; Merriam. 1988).
4. The data that emerge from a qualitative study are descriptive. That is.
data are reported in words (primarily the participan!'s words) or pic-
tures. rather than in numbers (Fraenkel & Wallen. 1990; Locke et al..
1987; Marshall & Rossman. 1989; Merriam. 1988).
5. The locus 01 qualitative research is on partlcipants' perceptions and
experiences. and the way they make sense 01 their lives (Fraenkel &
Wallen. 1990; Lockeetal .. 1987; Merriam. 1988). Thealtemptistherefore
to understand not one. but multiple realities (Lincoln & Guba. 1985).
6. Qualitative research locuses on the process that is occurring as well as
the product or outcome. Researchers are particularly interested in under-
standing how things occur (Fraenkel & Wallen. 1990; Merriam. 1988).
7. Idiographic interpretation is utilized. In other words. attention is paid to
particulars; and data Is interpreted in regard to the particulars 01 a
case rather than generalizations.
8. Quolitative research is an emergent design in its negotiated out-
comes. Meanings and interpretations are negotiated with human
data sources because it is the sUbjects' realities that the researcher
altempts to reconstruct (Lincoln & Guba. 1985; Merriam. 1988).
9. This research tradition relies on the utilizotion 01 tacit knowledge (intuitive
and lelt knowledge) because often the nuances 01 the multiple realities
can be appreciated most in this way (Lincoln & Guba. 1985). Therefore.
data are not quantifiable in the traditional sense 01 the word.
(Continued)
195
196 Designing Research
( Continucd)
10. Objectivity and truthfulness are critical to both research traditions.
However, the criterio lar judging a qualitative study differ Irom quan-
titative research. First and foremos!, the researcher seeks believabil-
ity, based on coherence, insight and instrumental utility (Eisner, 1991)
and trustworthiness (Lincoln & Guba, 1985) through a process of ver-
ification rather than through traditional validity and reliability mea-
sures. (Qualitotive choracferisfics are mentioned.)
The Elhnographic Research Design
This study will utilize the ethnographic research tradition. This design
emerged from the field of anthropology, primarily from the contributions of
Bronislaw Malinowski. Robert Park and Franz Boas (Jacob, 1987: Kirk & Miller,
1986). The intent of ethnographic research is to obtain a holistic picture 01
the subject 01 study wlth emphasis on portraying the everyday experiences
01 individuals by observing and interviewing them and relevant others
(Fraenkel & Wallen, 1990). The ethnographic study includes in-depth inter-
viewing and continual and ongoing participant observation of a situation
(Jacob, 1987) and in attempting to capture the whole picture reveals how
people describe and structure their warld (Fraenkel & Wallen, 1990). (The
author used the ethnographic approach.)
The Researcher's Role
Particularly in qualitative research, the role 01 the researcher as the primary
data collection instrument necessitates the identilication 01 personal values,
assumptions and biases at the outset 01 the study. The investigator's contribu-
tion to the research setting can be uselul and positive rather than detrimental
(Locke et al" 1987). My perceptions 01 higher educotion and the college pres-
idency hove been shaped by my personal experiences. From August 1980 to
May 1990 1 served as o college administrator on private campuses 01 600 to
5,000. Most recently (1987-1990), I served as the Dean lor Student Life at a small
college in the Midwest. As a member of the Presiden!'s cabine!, I was involved
with al! top level administrative cabinet activities and declsions and worked
closely with the faculty, cabinet officers, president and board of trustees. In
addition to reporting to the president, I worked with him through his first year
in office. 1 believe this understanding of the context and role enhances my
awareness, knowledge and sensitivity to many 01 the challenges, decisions and
issues encountered as a first year president and will assist me in working with the
informant in this study. 1 bring knowledge of both the structure of higher edu-
cation and of the role the college presidency. Particular attention will be paid
to the role of the new president in initiating change, relationship building, deci-
sien making, and providing leadership and visiono
Qualitative Procedures
Due to previous experiences working c10sely with a new college presiden!, 1
bring certain biases to this study. Although every effort will be made to ensure
objectivity, these biases may shape the way 1 view and understand the data
1 collect and the way I interpret my experiences. 1 commence this study with
the perspective that the college presidency is a diverse and often difficult
position. Though expectations are immense, I question how much power the
president has to initiate change and provide leadership and visiono I view the
first year as critical: filled with adjustments, frustrations, unanticipated surprises
and challenges, (Author reflected on her roJe in the study.]
Bounding Ihe Study
Salling
This study will be conducted on the compus 01 a stote college in the Midwest.
The college is situated in a rural Midwestern community. The institution's 1,700
students nearly triple the town's population of 1,000 when c1asses are in session.
The institution awards associate, bachelor and master's degrees in 51 majors.
Aclors
The informant in this study is the new President of a state college in the Midwest.
The primary inlormant in this study is the President. However, 1 will be observing
him in the context of administrative cabinet meetings. The president's cabinet
inciudes three Vice Presidents (Academic Aftairs, Administration, Student
Aftairs) and two Deans (Graduate Studies and Continuing Education).
Evan!s
Using ethnographic research methodology, the locus 01 this study will be
the everyday experiences and events of the new college presiden!' and the
perceptions and meaning attached to those experiences as expressed by
the informant. This includes the assimilation of surprising events or informo-
tien, and making sense of critical events and issues that orise,
Processes
Particular attention will be po id to the role 01 the new president in initiating
change, relationship building, decision making, and providing leadership
and visiono (Aufhor mentioned data collection boundaries.)
E!hical Consideralions
Most authars who discuss qualitative research design address the impor-
tance of ethical considerations (Locke et al" 1982: Marshall & Rossman,
1989: Merriam, 1988: Spradley, 1980). First and laremos!' the researcher has
(Col1tirHled)
197
198 Designing Research
( Continued)
an obligation to respect the rights, needs, values, and desires of the infor-
mant(s). To an extenl, ethnographic research is always obtrusive. Participant
observation invades the life of the informant (Spradley, 1980) and sensitive
information is frequently revealed. This is of particular concern in this study
where the inforrnan!'s position and institution are highly visible. The following
safeguards will be employed to protect the informan!'s rights: 1) the
research objectives will be articulated verbally and in writing so that they are
clearly understood by the informant (including a description 01 how data will
be used). 2) written permission to proceed with the study as articulated will
be received from the inlormanl, 3) a research exemption lorm will be liled
with the Institutional Review Board (Appendixes B 1 and B2). 4) the inlormant
will be inlormed 01 all data collection devices and activities, 5) verbatlm
transcriptions and written interpretations and reports will be made available
to the inlormant, 6) the inlorrnan!'s rights, interests and wishes will be con-
sidered first when choices are made regarding reporting the data, and
7) the linal decision regarding inlormant anonymity will rest with the informant.
(Aulhor addressed elhical issues and IRB review.)
Dala Colleclion Slralegies
Data will be collected Irom February through May, 1992. This will include a
minimum of bi-monthly, 45 minute recorded interviews with the informant
(initial interview questions, Appendix C), bimonthly two hour observations 01
administrative cabinet meetings, bi-monthly two hour observations 01 daily
activities and bi-monthly analysis of the presiden!'s calendar and docu-
ments (meeting minutes, memos, publications). In addition, the inlormant
has agreed ta record impressions 01 his experiences, thoughts and leelings
in a taped diary (guidelines for recorded reflection, Appendix D). Two follow-
up interviews will be scheduled for the end of May 1992 (See Appendix E for
proposed timeline and activity schedule). (fhe author proposed lo use face-
to-face interviews, participate as observe!, ond obtoin privote documents.)
To assist in the data collection phase I will utilize a lield log, providing a
detailed account of ways I plan to spend my time when I am on-site, and in
the transcription and analysis phase (olso comparing this record to how time
is actually spent). I intend to record details reloted to my observations in a
lield notebook and keep a lield diary to chronicle my own thinking, leeling,
experiences and perceptions throughout the research process. (fhe outhor
recorded descriptive ond reffective informotion.)
Dala Analysis Procedures
Merriam (1988) ond Marshall and Rossman (1989) contend that data col-
lection and dota analysis must be o simultaneous process in qualitative
Qualilullve F'rococluros
research. Schatzman and Strauss (1973) claim that qualitative data analysls
primarily entails classifying things, persons, and events and the properties
which characterize them. Typically throughout the data analysis process
ethnographers index or code their data using as many categories as possi-
ble (Jacob, 1987). They seek to identify and describe patterns and themes
Irom the perspective al the participant(s). then allempt to understand and
explain these pallerns and themes (Agar, 1980). During data analysis the
data will be organized categorically and chronologically, reviewed repeat-
edly. and continually coded. A list 01 major ideas that surlace will be chron-
icled (as suggested by Merrlam, 1988). Taped interviews and the
participan!'s taped diary will be transcribed verbatim. Field notes and diary
entries will be regularly reviewed. (Author described sleps in data analysis.)
In addition, the data analysls process will be aided by the use 01 a qualitative
data analysis computer program called HyperQual. Raymond Padilla
(Arizona State University) designed HyperQual in 1987 lor use with the
Macintosh computer. HyperQual utilizes HyperCard software and lacilitates
the recording and analysis 01 textual and graphic data. Special stacks are
designated to hold and organize data. Using HyperQual the researcher can
directly "enter field data, including interview data, observations, researcher's
memos, and illustrations ... (and) tag (or code) all or part 01 the source data
so that chunks of data can be pulled out and then be reassembled in a new
and illuminating configuration" (Padilla, 1989, pp. 69-70). Meaningful data
chunks can be identified, retrieved, isolated, grouped and regrouped lor
analysis. Categories or code names can be entered initially or at a later date.
Codes can be added, changed or deleted with HyperQual editor and text
can be searched for key categories, themes, words or phrases. (Author men-
tions the proposed use of computer software for dota analysis.)
Verificolion
In ensuring internal validity, the lollowing strategies will be employed:
l. Trlangulation of data-Data will be collected through multiple sources
to include interviews, observations and document analysis;
2. Member checking-The informant will serve as a check throughout the
analysis process. An ongoing dialogue regarding my interpretations 01 the
inlorman!' s reality and meanings will ensure the truth value 01 the data;
3. Long terms and repeated observations at the research site-Regular
and repeated observations 01 similar phenomena and sellings will
occur on-site over o four month period of time;
( Contil1ued)
199
200 Designing Research
(Continued)
4. Peer examination-a doctoral student and graduate assistant in the
Educational Psychology Department will serve as a peer examiner;
6. Participatory modes of research-The informant will be involved in
most phases of this study, from the design of the project to checking
interpretations and conclusions; and
6. Clarification of researcher bias-At the outsel of this study researcher
bias will be articulated in writing in the dissertation proposal under the
heading, "The Researcher's Role."
The primary slralegy ulilized in Ihis projecl lo ensure exlernal validity will be
the provision of rich, thick, detailed descriptions so Ihat anyone interesled In
transferability will hove a solid framework for comparlson (Merriam. 1988).
Three techniques to ensure reliability will be employed In this study. Firsl. the
researcher will provide a detailed account of the focus of the study, the
researcher's role, the informant's position and basis ter selection, and
the context from which dala will be gathered (LeComple & Goetz, 1984).
Second, Iriangulation or multiple methods of data collection and analysis
will be used, which strengthens reliability as well as internal validity (Merriam,
1988). Finally, data collection and analysis strategies will be reported in detail
in order to provide a clear and accurate picture of the methods used in this
study. AII phases of Ihis projecl will be subjecl lo scruliny by an external audi-
tor who ls experienced in qualitative research methods. (Aufhor idenfified
strategies 01 validity to be used in the study.]
Reporling Ihe Findings
Lofland (1974) suggesls that although data collection and analysis strategies
are similar across qualitative methods, the way the findings are reported is
diverse. Miles and Huberman (1984) address the importance of creatlng a
data display and suggest that narrative text has been the most frequent
form of display for qualitative data. This is a naturalistic study. Therefore, the
results will be presented in descriptive, narrative form rather than as a scien-
tific reporto Thick description will be the vehicle for communicating a holistic
picture of the experiences of a new college president. The final project will
be a construction of the informan!'s experiences and the meanings he
attaches lo Ihem. This will allow readers lo vicariously experience the chal-
lenges he encounlers and provide a lens Ihrough which readers can view
the subjec!'s world. (Outcomes 01 the study were mentioned.]
SUMMARY
This chapler explored the steps that go into developing and wriUng a
qualitative procedure. Recognizing the variation that exists in qualitalivc
studics. the chapter advances a general guidcline 1'01' procedures. This
guidclinc includes a discussion about the general characteristics 01' qual-
itative research if audiences are not familiar with this approach to
research. These characteristics are that the rescarch lakes place in the
natural setting, relies on the rescarcher as the instrument [or data col!ec-
tion, employs multiplc melhods al' data collection, is inductive, is bascd
on participants' rneanings, is erncrgent, oftcn involves the use of a theo-
retical lens, is interpretive, and is holisUc. The guideline recommends
rnentioning a strategy of inquiry, such as lhe study 01' individuals (narra-
tive, phenomenology), the exploration al' processes, activities and events
(case study, grounded theory), or the examination of broad culture-sharing
behavior o[ individuals 01' groups (ethnography). The choice (J[ stralegy
neeels lo be presented ami defended. Further, lhe pro pos al needs to
address the role of lhe researcher: past experiences, personal connecUons
to the site, steps to gain entry, and sensitive ethical issues, Discussion of
data col!eelion should include lhe purposeful sampling approach amI the
[orms of dala to be collected (i.c., observations, interviews, documenls,
audiovisual materials). Tt is usefiJl to also indicate lhe types of data
recording protocols that will be used. Data analysis is an ongoing proccss
'Iduring research. It involves analyzing participant information, and
researchers typically employ general analysis sleps as wel! as those steps
found within a specific stralegy of inquiry. More general steps include
organizing and preparing the data, an initial reading through the infor-
malion, coding the dala, dcveloping from the codes a description and lhe-
rnatic analysis, using computer programs, reprcsenting the findings in
tables, graphs, and figures, amI interpreting the findings. Thcse interpre-
tations involve stating lessons learned, cOlllparing the findings with past
literaturc and theory, raising questions, and/or advancing an agenda for
reformo Thc proposal should also contain a seeUon on the expeeted oul-
comes for the study. Finally, an additional importanl step in planning a
proposal is to mentan the strategies that will be used to validate lhe aeen-
raey al' lhe findings, clcmonstrate the reliability uf procedures, and dis-
euss the role al' generalizability.
202 Designing Research
l. Write a plan fol' the procedure to be used in yaur qualitatve
study. Arter wriling lhe plan, use 'rabIe 9.1 as a checklist lo delermine
the comprehensiveness of your plan.
L Develop atable that lists, in a column on the left, the steps you
plan to takc to analyzc your data. In a column on thc right. indicate
the steps as they apply directly to your project, the research strategy
you plan to use, and data that you have collected.
ADDlTIONAL RfADiNGS
Marshall. e.o & Rossman, C. B. (20(6). IJes{fllin{f l![/{!!taLve research (4th ed.).
ThoLlsand Oaks, CA: Sage.
Cat,llCr,inc Marshall and Crctchcn Rqssman introduce the proccdurcs ror dcsigning
a quahtatlvc study ane! l qualitative proposaL The topies covcred are comprchensive.
'[.'hey building a conceptual framevvork around a st.udy; the logic and
tlOns 01 the overall design ami methods; methods of data coHection amI procedures ror
managing, recording, ami analyzing qualilative dala: ami tbe resources needed ror a
such as time, personncl, and funding. This is a comprehensive ami insightful text
trom which both beginners ami more experienced qualitative researehcrs learn.
Fliek. O. (1\eI.). (2007). '['he ,)'age Qllalitativc Hesearch Kit. London: Sage.
This is an cight-volume kit edited by Uwe Flick thal is authored by dil'l'ercnt world-
class qualitative researchers and was ereated to collectively address the eore issues that
arise vvhcn rcsearchcrs actuaHy do qualil.alive rescarch. It addrcsscs how 1.0 plan and
l stue!y, the collcetion ami production 01' qualitative data, tbe analy-
SIS ni. qu.ailtatwe dala (e.g .. visual data, discoursc analysis). ami tbe issucs 01' quality in
quahtatlve rcsearch. OveraU, it prescnts a recent, up-lo-date window n lo lhe ncld 01'
qualitative research.
Creswcll, J. W. (2007). QualitaUve jru{uiry wul researd! desi{jll: CllOosill{j (/frlOr/{j .live
(Ipproaches (2nd cd.). Thousand Oaks, CA: Sage.
Sometimes those \ovho write about qualitative research take a philosophical stance
loward the topic and rcaders are leJ't without an undcrstanding oI' thc procedures and
actually tlscd in designing ane! conducting a qualitative study. My book takes
hve approachcs Lo qualitalive research. phenomc!1o!ogy, groundcd
theory, ethnography, ane! case study-and discusscs how the proccdures I'or conduct-
ing these forms 01' inquiry are both similar and elilTerenL In the cnel, readers can more
easily choose \ovhich ot' thc Ove would best suit thcir rcscarch problems as weH as their
personal stylcs 01' rescarch.
Mixed Methods
Procedures
ith the development and perceived legitimacy ol both qual-
itative and quantitative research in the social and human
sciences, mixed methods research. employing the combina-
tion of quantitative and qualitative approaches. has gained popularity.
This popularity is beca use research methodology continues to evolve
and develop. and mixed methods is another step forward. utilizing
I
the strengths of both qualitative and quantitative research. Also, the
problems addressed by social and health science researchers are com-
1, plex. and the use of either quantitative or qualitative approaches by
themselves is inadequate to address this complexity. The interdisciplinary
nature of research, as well. contributes to the formation of research
teams with individuals with diverse methodological interests and
approaches. Finally. there is more insight to be gained from the combi-
nation ol both qualitative and quantitative research than either lorm by
itself. Their combined use provides an expanded understanding 01 research
problems.
This chapter brings together many 01 the threads inlroduced in
the earlier chapters: It extends the discussion about the philosophical
assumptions 01 a pragmatic philosophy. the combined use ol qualita-
tive and quantitative modes 01 inquiry. and the use of multiple methods
introduced in Chapter l. It also extends the discussion about research
problems that incorporate the need both to explore and explain
(Chapter 5). It lollows a purpose statement and research questions
locused on understanding a problem using both qualitative and quan-
titative methods (Chaplers 6 and 7). and it advances the reasons lor
using multiple forms 01 data collection and analysis (Chapters 8 and 9).
COMPONENTS Of MIXED MEl'HODS PROCEDURES
Mixed methods rescarch has evolved a set al' procedures that proposal dcvel-
opers can use in planning a mixcd methods study. In 200}, lhe Handbook 01'
203
204 Designing Research
Mixed Metilads in the Social & llehavim' Sciences (Tashakkori & Teddlie, 2003)
was published, providing the llrst comprehensive overview 01' this strategy of
inquiry. Now several journals emphasizc 111ixed lnethods research, such as
the ournal oI Mixed Metilads Researe/, Qualit!! "mi Quantit!!, and Fie/d Mel/lOds,
while numeraus others actively encourage this 10rm 01' inquiry (e.g" lntemalional
ollmal oI Social Researe/ MetllOdolayy, Qualitalive Healtil Researe/, Armals oI
Family Medicine). Numerous published research sludies have incorporated
mixed methods rescareh in the social and human sciences in diverse fields
such as oeeupational therapy (Lysaek & Krefting, 1994), interpersonal eom-
muniealion (Boneva, Kraut, & Frohlich, 20(1), AIDS prevention (Janz et al"
1996), demenlia carcgiving (Weitzmau & Levkoff, 2000), mental health
(Rogers, Day, Randall. & Bentall, 2(03), and in middle-sehool scienee (Hautz,
1995). New books arrive each year solely devoled lo mixed melhods research
(Bryman, 2006; Creswell & Plano Clark, 2007; Creene, 2007; Plano Clark &
Creswell, 2008; Tashakkori & Teddlie, 1998).
A ehecklist 01' questions for designing a mixed methods study appears
in Table 10.1. These components call1or advancing the nature 01' mixed
methods research and the type 01' strategy being proposed for the stndy.
They also inelude the need I'lr a visual model 01' this approaeh, the specifie
procedlll'es of data collection and analysis, the rescarcher's role, ami the
structure for presenting the final reporl. Following the discussion 01' each
01' these components, an example 01' a procedures section 1'ro111 a mixed
methods study is presented to show how to apply these ideas.
Of MIXED MHHODS RESEARCH
Because luixed methods research is relatively new in the social and htnnan
sciences as a distinct research approach, it is useful to convey a basic clcll-
nition and deseription 01' the approach in a proposa!. This might include
the 1()lIowing:
Trace a brief history 01' its evolution. Several sourees identify its
inception in psychology ancl in the multitrait-multimethod matrix 01'
Campbell ami Fiske (1959) to interest in convcrging or triangulating d i l ~
ferent quantitative and qualitativc data sourees (lick, 1979) amI on to the
development of a distinct methodology of inquiry (see Creswell & Plano
Clark, 2007; Tashakkori & Teddlie, 1998).
liI Define mixed mcthods researeh by incorporating the definition in
Chapter 1 that locuses cm combining both quantitative and qualitative
research and methods in a rcsearch study (see a more expancled vicw 01'
defining mixed methods research in johnson, Onwuegbuzie, & Tumer,
20(7). Highlight lhe reasons why researchers employ a mixed methods
design (e.g., lo broaden understanding by incorporating both qualitative
Mixed Methods Procedures
Is a basic delinition 01 mixed methods research provided?
Is o reason given lar using both quantitative ond qualitative
approaches (ar dota)?
Does the reader hove a sense lar the potential use 01 a mixed
methods design?
Are the criteria identilied lar choosing a mixed methods strategy?
Is the strategy identilied, and are its criteria lar selection given?
Is a visual model presented that illustrates the research strategy?
Is the proper notation used in presenting the visual model?
Are procedures 01 data colleclion and analysis mentioned as they
relate to the model?
Are the sampling strategies lor both quantitotive and qualitative
data colleclion mentioned? Do they relate to the strategy?
Are specilic data analysis procedures indicated? Do they relate to
the strategy?
Are the procedures lor validating both the quantitative and
qualitative data discussed?
Is the narrative structure mentioned, and does it relate to the type
01 mixed methods strategy being used?
and quantitative rcsearch, or to use Dne approach to belter understand,
explain, or build on the results fl'Om the other approach). AIso note that
the mixing of the lwa might be within one sludy al' among several sludies
in a prognuTI of inquiry. Recognize that many diffcrent tern1S are used for
this approaeh, such as illtegmtillg, ''Ynthesis', ijllantitative "",1 qllalitative
metllOds, multimetllOd, and mixed metllOdalogy, but that reeent writings use
the term mixed lIlet/oi!s (Bryman, 2006; Tashakkori & Teddlie, 20(3).
liiI Briefly diseuss the growth 01' inlerest in mixed methods researeh as
expressed in books, journal articles, diverse disciplines, ami funded proj-
eets (see CresweIl & Plano Ciad" 2007 for a discussion about the many
initiatives that contribllle to mixed methads today).
liI Note the chaIlenges this I()rm of research poses for the inquirer.
These inelude the need lar extensive data coIlection, lhe time-intensive
naturc of analyzing both tcxt and numcric data, and the rcquirement for
the rescarcher to be familiar wilh both quantitativc ami qualitative forms
01' rescarch.
205
206 Designing Research
TYPES Of MIXEO METHOOS STRATEGIES ANO VISUAL MOOHS
There have been several typologies lr classifying and identifying types of
mixed methods strategies that proposal developers might use in their pro-
posed mixed methods study. Creswell and Plano Clark (2007) identify 12
classification systems drawn l'rom the fields 01' evaluation, nursing, public
health, education policy and research, and social and behavioral rcscarch.
[n these classifieations, authors use diverse terms I'or their types al' designs,
and a substantial amount 01' overlap exists in the typologies. Por purposes
01' this discussion 1 will identify and discuss the six types that my calle agues
and 1 advanced in 2003 (Creswell et al., 20(3).
Planning Mixed Methods PlOcedure
It is helpful, however, belre discussing the six types, to consider several
aspects that inlluence the design 01' procedures for a mixed methods study.
Four in1portant aspects are timing, weighting, mixing, and theorizing (as
shown in Figure lO.1).
Timng
Proposal developers need to consider the Urning af their qualitative and
quantilative dala collection, whelher it will be in phases (sequentially) or
gathered at lhe same lime (concurrently). When the data are collected in
phases. ether the qualitative 01' the quantitative data can come first. It
depends on the initial intent of the researcher. When qualitative data are
collected first, the intent is to explore the topic with participanls at sites.
Then the researcher expands the understanding through a seeond phase
in which data are collected from a large nnmber of people (typically a sam-
pie representative 01' a population). When data are collected concurrently,
both quantitative and qualitaUve data are gathered at the Saine time and
the implementution is simultaneous. In many projects it muy be unwork-
able to collecl data over an expanded time period (e.g .. in the health sci-
enees when busy medical personnel have limited time [or data collection in
the ficld). In this case, it is more manageable to eolleet both quantitalivc
and qualitative data at roughly the same time, when the researeher(s) is in
the field collecting data, rather than to revisit the field multiple times lr
data collection.
Weighling
A second factor that gaes into designing procedures is the 1/V'dght ar pri-
arity given 10 quantitative ar qualitative research in a particular study. In
some studies, the weight might be equal; in olher studies, it might emphasize
Mixed Melhods Procedures
Timing Weighting Mixing
No Sequence
Equal
lntegrating
concurrent
Sequential-
Oualitative first
Oualitative Connecting
Sequential-
Quantitative first
Quantitative
Embedding
figure '10.1 Aspects to Consider in Planning a Mixed Methods Design
SOURCE: Adapted from Creswell et al. (2003).
Theorizing
Explicit
lmplicit
one or the other. A priorily lr one lype dcpends on the interesls of the
researcher, the audience I'or the study (e.g., l'aculty committee, professional
association), and what the investigator seeks to emphasiz:e in the study.
.,ln practical terms, wcight OCCUl'S in a mixed methods sludy through such
strategies as whelher quantitative or qualitative inforn1ation lS C111phasized
Ilrst, the extent of trcatment 01' one type al' dala or the olher in the project,
or the use 01' primarily an inductive approach (Le., generating then1es in
qualitative) or a deductive approach (i.e .. testing a theory). Sometimes the
rescarcher intentianally uses one fornl 01' data in a supportive role to a larger
study, as is found in some experimental trials (see Rogers et al .. 2(03).
Mxng
Mixing the dala (and in a larger sense, mixing the research quesUons.
philosophy, the interprelation) iB difficult al best when one considers that
qualitative data consists 01' text and images and quantitalve data, num-
bcrs. There arc two di1'ferent quesUons here: Whc/l cines a researcher 111x
in a mixed melhods study? And !ww does mixing occur? The first questioIl
is mnch easier to answer than the second. Mixing of the two types 01' data
111ight occur at several stages: the data collection. the data analysis. inter-
pretation, or at al! three phases. For proposal developers using mixed
methods, il is important lo discuss amI present in a proposal when the mix-
ing will occur.
How the data are mixed has received considerable recent attcntion
(Creswell & Plano Clark, 2(07). Mbng means either t:hat lhe qualitative
.....
......
207