Research in Nursing & Health, 2000, 23, 246255

Focus on Research Methods

Combining Qualitative
and Quantitative Sampling,
Data Collection,
and Analysis Techniques
in Mixed-Method Studies
Margarete Sandelowski*

University of North Carolina at Chapel Hill, School of Nursing,

7460 Carrington Hall, Chapel Hill, NC 27599
Received 12 July 1999; accepted 13 January 2000
Abstract: Researchers have increasingly turned to mixed-method techniques to expand the
scope and improve the analytic power of their studies. Yet there is still relatively little direction on
and much confusion about how to combine qualitative and quantitative techniques. These techniques are neither paradigm- nor method-linked; researchers orientations to inquiry and their
methodological commitments will influence how they use them. Examples of sampling combinations include criterion sampling from instrument scores, random purposeful sampling, and stratified purposeful sampling. Examples of data collection combinations include the use of instruments for fuller qualitative description, for validation, as guides for purposeful sampling, and as
elicitation devices in interviews. Examples of data analysis combinations include interpretively
linking qualitative and quantitative data sets and the transformation processes of qualitizing and
quantitizing. 2000 John Wiley & Sons, Inc. Res Nurs & Health 23:246255, 2000

Keywords: mixed-method studies; combination studies; qualitative and quantitative research;

sampling; data collection; data analysis; triangulation

The idea of mixing qualitative and quantitative

methods has stimulated much interest and debate
(e.g., Greene & Caracelli, 1997a; Sandelowski,
1995; Swanson, 1992; Tashakkori & Teddlie,
1998). Researchers increasingly have used mixedmethod techniques to expand the scope of, and
deepen their insights from, their studies. As advocates of mixed-method research have argued, the
complexity of human phenomena mandates more
complex research designs to capture them. Despite this interest, there is still relatively little direction on and much confusion about how to accomplish mixed-method studies. In this paper, I
discuss the where, what, why, and especially, the

how of studies combining qualitative and quantitative techniques.

THE WHERE, WHY, AND WHAT

OF COMBINATIONS
Combination or mixed-method studies are concretely operationalized at the technique level, or
the shop floor, of research: that is, at the level of
sampling, data collection, and data analysis.
Mixed-method studies are not mixtures of paradigms of inquiry per se, but rather paradigms are

*Professor.

246

2000 John Wiley & Sons, Inc.

COMBINING / SANDELOWSKI

reflected in what techniques researchers choose to

combine, and how and why they desire to combine
them.
At the Paradigm Level
As many scholars (e.g., Guba & Lincoln, 1994;
Heron & Reason, 1997) have variously named and
described them, paradigms of inquiry are worldviews that signal distinctive ontological (view of
reality), epistemological (view of knowing and the
relationship between knower and to-be-known),
methodological (view of mode of inquiry), and axiological (view of what is valuable) positions. Indeed, paradigms of inquiry are best understood as
viewing positions: ways, and places from which,
to see. Because different paradigms, such as (neoor post-) positivism, constructivism, critical theory, and participatory inquiry, entail contradictory
viewing positions, combinations at the paradigm
level are not true combinations, mergers, or reconciliations of worldviews, but rather the explicit
framing of inquiry in two or more worldviews,
each of which remains distinct from the other. That
is, for example, it is not possible to combine, merge,
or reconcile a view of reality as singular and objective (positivist) with views of it as multiple and
individually or culturally constructed (constructivist), or as historically contingent (critical theory). A critical theorist will frame the issues around
hormone replacement for midlife women differently from a neopositivist.1 That is, they will each
see different things and, therefore, ask different
questions that will, in turn, require the use of different methods and techniques to answer them.
Such a paradigm combination may be used to elicit two or more perspectives on hormone therapy,
or for the purpose of initiation (Greene, Caracelli, & Graham, 1989, p. 259), to surface paradoxes and contradictions that surround hormone
therapy. Accordingly, two or more paradigms of
inquiry can be used to frame the same target phenomenon (in this case, hormone replacement).
Yet, arguably, the positivist and the critical theorist may not really be studying the same phenomenon, because to see a phenomenon in a certain
way is to change that phenomenon. As Pearce
(1971, p. 2) observed, a change of world view can
change the world viewed. Moreover, it is debatable whether one researcher can hold two different viewing positions, albeit at different times.
1I

use terms like positivist and positivism to quickly communicate complex and admittedly controversial ideas for the
purposes of this paper, not to stereotype researchers or viewing positions.

247

That is, it is uncertain whether a researcher can

conduct a project framed first, for example, in positivism and then later, in critical theory. Such
worldviews, like all strongly held belief systems,
are not easily exchanged.
As data collection and analysis techniques are
not linked to paradigms (Berman, Ford-Gilboe, &
Campbell, 1998; Sandelowski 1995), both the researcher in a positivist viewing position and the researcher in a critical theory viewing position may
use interviews and even the very same standardized measures to answer their questions, but they
will employ these techniques and, more importantly, analytically treat their results differently. Indeed, arguably the key difference between qualitative and quantitative (when these words are
used to signal different inquiry paradigms or research traditions) researchers is in their attitude toward and treatment of data. Accordingly, although
techniques can be mixed, the resulting mix will reveal the researchers viewing position (or, in cases
of mixed-up research, the researchers futile effort
to mix the research equivalent of oil and water).
At the Method Level
Combinations at the method level can be used to expand the scope of a study as researchers seek to capture method-linked dimensions of a target phenomenon (Greene et al., 1989, p. 259). As Wolfer (1993)
proposed, different aspects of reality lend themselves to different methods of inquiry. For example,
the physiology of hormone replacement therapy for
women lends itself to study via physiologic measures, while the debate around the benefits and liabilities of hormone replacement therapy lends itself
to discourse, semiotic, and other cultural analyses.
In the study I conducted with my colleagues of the
transition to parenthood of infertile couples (Sandelowski, Holditch-Davis, & Harris, 1992), we used
grounded theory methodology with naturalistic,
ethological observation to capture different features
of this phenomenon, including couples understandings of their struggle to become parents and
their interaction with their babies. Grounded theory and ethological observation share a common naturalist imperative (that is, not to manipulate, or impose a priori conceptualizations on, the target
phenomenon) that make them compatible for use in
one study. But grounded theory is particularly wellsuited to theorizing human understandings, while
ethological observation is especially well-suited to
theorizing human behavior.
Although the various dimensions of phenomena may be method-linked in that different dimensions may be best captured by different methods,

248

RESEARCH IN NURSING & HEALTH

methods, like paradigms, are not specifically

linked to techniques. Grounded theory may be
generated using an array of qualitative and quantitative data collection techniques and sources.
Moreover, methods are not uniformly linked to
paradigms (Greene & Caracelli, 1997b). Grounded
theory may be conducted in a neopositivist or constructivist paradigm (Annells, 1996). A neopositivist researcher conducting grounded theory believes in an external and objectively verifiable
reality. In contrast, a constructivist conducting
grounded theory believes in multiple, experientially based, and socially constructed realities. For
the neopositivist, concepts emerge or are discovered, as if they were there to be found. The act of
discovery is separate from that which is discovered. For the constructivist, concepts are made,
fashioned, or invented from data. What constructivists find is what they made. For the constructivist, all human discovery is creation.
At the Technique Level
The technique level of research is the site where
combinations actually occur and is what is most
often referred to in discussions of mixed-method
research. Such combinations entail the use of sampling, data collection, and data analysis techniques
commonly (although not necessarily) conceived
as qualitative or quantitative. Because techniques
are tied neither to paradigms nor to methods, combinations at the technique level permit innovative
uses of a range of techniques for a variety of purposes. Three purposes include (a) triangulation, to
achieve or ensure corroboration of data, or convergent validation; (b) complementarity, to clarify, explain, or otherwise more fully elaborate the
results of analyses; and (c) development, to guide
the use of additional sampling, and data collection
and analysis techniques (Greene et al., 1989, p.
259). The rest of this paper is devoted to illustrating how such purposes can be achieved.

THE HOW OF COMBINATIONS

Mixed-method studies entail concrete operations
at the technique level of research by which qualitative and quantitative techniques are used together and either remain distinct design components, or are explicitly integrated (Caracelli &
Greene, 1997). As shown in Figure 1, either qualitative or quantitative approaches to sampling,
data collection, and data analysis may prevail or
have equal priority in a study, and both qualitative

and quantitative approaches may be used sequentially, concurrently and iteratively, or in a sandwich pattern.
Combining Sampling Strategies
One of the most important features distinguishing
what is commonly referred to as qualitative from
quantitative inquiry is the kind of sampling used.
While qualitative research typically involves purposeful sampling to enhance understanding of the
information-rich case (Patton, 1990), quantitative
research ideally involves probability sampling to
permit statistical inferences to be made. Although
purposeful sampling is oriented toward the development of idiographic knowledgefrom generalizations from and about individual cases
probability sampling is oriented toward the development of nomothetic knowledge, from generalizations from samples to populations. Notwithstanding these key differences, purposeful and
probability sampling techniques can be combined
usefully.
Criterion sampling. For example, in design
template 2 shown in Figure 1, in which the use of
quantitative techniques precede the use of qualitative techniques, research participants scores on
the instruments used to collect data in the quantitative portion of the study can be used to initiate a
criterion sampling strategy. Criterion sampling is
a kind of purposeful sampling of cases on preconceived criteria, such as scores on an instrument.
Cases may be chosen because they typify the average score; this kind of sampling may also be referred to as typical case sampling. Cases may be
chosen because they exemplify extreme scores;
this kind of sampling may also be called extreme
or deviant case sampling. (The term deviant here
refers to any departure from a specified norm.)
Such cases are highly unusual. Or, cases may be
chosen because they show a variable intensely, but
not extremely; this kind of sampling may also be
referred to as intensity sampling (Patton, 1990, pp.
182183).
Researchers using scores on instruments as the
criterion for purposeful sampling may wish to collect more data (e.g., via interviews or observations) from the chosen participants for the purpose
of triangulation: that is, to discern whether a typical, extreme, or intense case of something on a
standardized test is also a typical, extreme, or intense case using other data collection techniques.
Or, researchers may use the criterion of scores
(typical, extreme, or intense scores) for the purpose of complementarity: that is, to find out more
about what makes a case typical, extreme, or in-

COMBINING / SANDELOWSKI

249

FIGURE 1. Hybrid, combination, or mixed-method design templates.

tense. In the process of sampling for complementarity, researchers will inevitably also obtain information on convergent validity and, thereby,
also achieve the purpose of triangulation. That is,
in the process of obtaining fuller information on
why persons scored as they did, they will also obtain information on whether persons look the same
on interview or observation as they did on the
quantitative measure of a target phenomenon. Researchers will sample participants in scoring categories until the point of informational redundancy; that is, until they have collected information
from enough cases in each scoring category to allow them to draw conclusions about the validity of
the result (if they are seeking convergent validity),
or to elaborate on and clarify the result. Sampling
on the basis of scores is an especially useful strategy in clinical trials of interventions to validate or
clarify their effects on different participants.

Random purposeful sampling. Another example of the combined use of probability and purposeful sampling is random purposeful sampling,
which may also be used in design template 2 or 3.
This sampling strategy is employed when there is
a very large pool of potentially information-rich
cases and no obvious reason to choose one case
over another. For example, in a clinical trial of an
intervention to reduce pain with 500 people, 300
of them scored as having less pain on a standardized measure of pain, 150 scored as having as
much pain as they had before the treatment, and 50
scored as having more pain. These numbers are
too large for any purposeful sampling strategy oriented toward the intensive study of the particulars
of each case. Accordingly, cases can initially be
chosen from each of these three scoring groups
(the criterion for sampling here is again the scores)
by assigning all the cases in each group a number

250

RESEARCH IN NURSING & HEALTH

from a random number table and then drawing

them in turn. Each case drawn must meet the minimum criterion in all purposeful sampling: namely, that it is an information-rich case. And, only so
many cases are drawn in each scoring group that
will permit researchers to make the kinds of inferences they wish to make: for example, concerning
convergent validity, or fuller description or explanation of cases.
Stratified purposeful sampling. Another kind
of combination of sampling techniques is stratified
purposeful sampling, where the researcher wants
to ensure that certain cases varying on preselected
parameters are included. This strategy can potentially be used in any of the design templates shown
in Figure 1. Although this kind of sampling is
from a probability sampling standpointstatistically nonrepresentative (Trost, 1986), it is, from a
purposeful sampling standpoint, informationally
representative. Each case represents a prespecified
combination of variables, the distinctive confluence of which is the focus of study. The researcher
wants to explain how these variables come together to make a case the case that it is.
For example, a researcher studying caregiving
may wish to ensure that s/he has 1-2 cases that
minimally to maximally vary on four parameters:
type of caregiving dyad, role of each member of
dyad, mortality of illness, and transmissibility of
illness. Caregiving dyads include parents caring
for their children, adult or minor children caring
for their parents, husbands caring for their wives,
wives caring for their husbands, heterosexual partners caring for each other, male and female homosexual partners caring for each other, brothers
caring for sisters, sisters caring for brothers,
grandparents caring for grandchildren, grandchildren caring for grandparents, friends caring for
friends, strangers caring for strangers, paid caregivers caring for patients, and so on. In each of
these dyads, one or the other member can be the
caregiver or the cared-for; accordingly, wives care
for or are cared for by husbands. Finally, illnesses
can be classified as mortal or nonmortal, and
transmissible or nontransmissible.
Accordingly, a researcher may want to have the
most variation s/he can achieve on each of these
four parameters, or limit variation on one or more
of these parameters. In a maximally varied form of
purposeful stratified sampling, the researcher might
want to have 12 cases of every combination of
variations s/he can find. The researcher may want
to have 12 cases of a husband caring for his wife,
who suffers from a mortal and nontransmissible
illness; 12 cases of a husband caring for his wife,
who suffers from a nonmortal and nontransmissible

illness; 12 cases of a husband caring for his wife,

who suffers from a mortal and transmissible illness; and, 12 cases of a husband caring for his
wife, who suffers from a nonmortal and transmissible illness. The researcher may also want to have
cases of wives caring for husbands, which show
the variations just itemized on illness mortality
and transmissibility. In addition, the researcher
may want to include these variations for as many
of the other kinds of caregiving dyads previously
listed as s/he can. Or, the researcher may want to
restrict sampling only to one kind of caregiving
dyad (eg., parentchild caregivers) varying on the
parameters of member role and type of illness.
In short, in stratified purposeful sampling, researchers want to fill each sampling cell with 1-2
cases that exemplify the kinds and degree of variation they surmise are relevant to understanding a
target phenomenon, such as caregiving. This kind
of sampling typically involves empirical, as opposed to theoretical, cases that represent combinations of demographic (e.g., age, sex, income, and
education) and other low-inference variables. Lowinference variables are variables concerning which
there is likely to be consensus about what they are.
That is, there is likely to be no disagreement on
whether a case represents a wife caring for her sick
husband. In contrast, an empirical case of Mrs.
Jones caring for her husband, Mr. Jones, who is
dying from cancer, might also represent a theoretical case of reluctant caregiving, as opposed to
engaged caregiving. Although researchers can
use a stratified purposeful sampling strategy to fill
sampling cells with such theoretical cases, they
must first have made the case for such theoretical
cases. Figure 2 illustrates a stratified purposeful
sampling plan restricted to one kind of caregiving
dyad (parentchild), but varying on member role
and type of illness. A total sample size of 16 would
be required to fill each sampling cell with one
case. Figure 2 also illustrates why certain sampling cells may be difficult or impossible to fill;
few parentchild dyads will be comprised of a minor child caring for her/his own minor ill child.
Combining Data Collection Techniques
Another set of concrete operations at the technique
level of research entail the combined use of data
collection techniques that are commonly (but not
necessarily) associated with either qualitative or
quantitative research, such as open-ended and unstructured interviewing and structured questionnaires, respectively. (Some would argue that all
data collection techniques in human subjects research, including instruments, are qualitative in

COMBINING / SANDELOWSKI

251

FIGURE 2. Illustration of stratified purposeful sampling plan.

that they involve verbal data, which are only later

transformed into numbers.) Researchers viewing
positions will influence how they use these techniques. For example, for many researchers in a
positivist viewing position, data collection techniques vary in the degree to which they yield objective data. Observations of behavior are generally thought to be more objective than self-reports
of behavior. When a target phenomenon can be observed, observation is often the criterion measure
against which self-report is judged. Accordingly,
researchers in a positivist viewing position will often seek more objective measures to evaluate
the validity of more subjective measures. Moreover, whenever there is a discrepancy between
what participants do and say they do, what observers see participants doing is generally considered
a more accurate reflection of reality than self-report.
The self-report is typically called into question.
In contrast, from a constructivist viewing position, there is no hierarchy of data collection techniques whereby one technique is judged to yield
more objective (or more accurate or truer) data
than another. If the results from two data collection techniques do not converge, these results are
treated as interpretive opportunities: either to
show that no true discrepancy exists or to suggest
the phenomenon that accounts for the apparent
discrepancy. For example, Silverman (1993) demonstrated how a discrepancy between parents accounts of their behavior and observations of their
behavior in an examining room could be accounted for without resorting to judgments about these
parents as unreliable informants. Indeed, he advised against the kind of simple-minded triangulation whereby one kind of data is used simply to corroborate or refute the results of another
without attention to the embedded, situated nature of accounts (p. 200). Helitzer-Allen and
Kendall (1992) described how discrepancies between data derived from surveys and laboratory

tests were resolved by the analysis of ethnographic

interviews, which provided them the opportunity
to deepen their understanding of the use of antimalarial chemoprophylaxis during pregnancy. This
study also shows the inherently multi-technique
nature of ethnographic studies.
The Varied Uses of Instruments
Instruments can be used in combination studies to
fulfill a variety of objectives. They can be used to
provide fuller description of cases in areas suggested by interview or observation data, as, for example, when interview data suggest that participants are depressed and the researcher decides to
administer a depression inventory. Each participants score may be compared to the normative
score for that instrument or to the scores of other
participants in the study. Instruments are used here
to make case-bound or idiographic generalizations
that is, generalizations about each participant in
the studynot to make nomothetic generalizations, or generalizations from the study sample to
populations. This use of instruments exemplifies
design template 1 shown in Figure 1, where the decision to use an instrument is made on analytic
grounds developed from data collected in a qualitative study. This use also exemplifies the development purpose for mixed-method research, whereby the results of using one kind of data collection
technique informs or guides the use of another
kind. Researchers may also use instruments for
description (as opposed to statistical inference)
concurrently with interviews or observations for
the purpose of complementarity, as exemplified in
design template 1a, or as a development bridge between the qualitative exploration of a target phenomenon and the further explanation of this phenomenon, as shown in design template 6. The
development bridge herethe instrumentdirects researchers toward whom they will sample

252

RESEARCH IN NURSING & HEALTH

and what data they will collect in the second qualitative portion of a study.
Another use of instruments that fulfills the
purpose of development is to guide purposeful
sampling. The results of instruments can direct researchers more precisely to the kinds of participants they may wish to recruit and the nature of information they will want to obtain from them. As
described previously, instrument scores can be the
basis for sampling in a follow-up qualitative study.
Results from a survey can direct researchers toward the most fruitful variables and associations
to examine further both qualitatively, as shown in
design template 3 in Figure 1, or first qualitatively
and then again, quantitatively, as shown in design
template 7 in Figure 1. Ornstein and his colleagues
(1993) initiated focus groups after learning from
the results of a telephone survey that there were
three groups of nonresponders to a reminder
letter for cholesterol screening. The focus groups
were organized to elicit more information concerning the lack of response. This information could
then serve as the basis for revising the reminder
protocol and then qualitatively or quantitatively
evaluating its effectiveness in reducing the numbers of nonresponders. For example, the new reminder protocol could be tested in a clinical trial
and the scores on the instruments used to appraise
the protocol could then be used as the basis for criterion sampling to further explain those scores.
This combined example illustrates how a quan !
Qual study (survey>focus group) can lead to a
Quan ! qual (experiment ! criterion sampling "
interviews) study. Together, these studies can be
represented as quan!Qual! Quan!qual.
Instruments can also be used as elicitation devices in interviews concerning both the target phenomenon and the instrument itself. For example, a
researcher may use participants responses on a
depression inventory to trigger more thoughts and
feelings about depression in individual interview
sessions. Participants often require some assistance to articulate inchoate thoughts or to speak
the unspeakable. The use of an instrument as an
elicitation device can serve this purpose, just as
the use of projective techniques (e.g., Ornstein et
al., 1993) can help participants language, and focus on, a target experience.
Researchers may also ask participants to comment specifically on their view of each item on an
instrument in order to appraise its content and construct validity. In the infertility study mentioned
previously (Sandelowski et al., 1992), we used a
symptom inventory to appraise the type and intensity of symptoms women and men were experiencing as they awaited birth or adoption. One of

the symptoms listed was headache. One waiting

father spontaneously mentioned to me that he had
experienced a headache from hitting his head on a
pipe doing a home repair, but, on the symptom inventory, did not mark headache as one of his
symptoms. As this man explained it, he surmised
that we were not interested in symptoms that derived from home repairs or other events apparently unconnected to waiting for a child. What this
man clarified for me was all of the judgments participants make when they encounter any item on
an instrument of which researchers are completely unaware. Accordingly, researchers may want to
explicitly use instruments as data elicitation devices to determine exactly what participants see
and, therefore, respond to in each item. Because
participants may also offer reasons for responding
as they did, in addition to explanations for how
they and persons like them are likely or ought to
respond, researchers may obtain information not
only on the content validity of an instrument, but
also on its construct validity.
Instrument scores can also be used for qualitative profiling. Qualitative profiling, which is discussed further in the section on qualitizing, entails theoretically grouping or typing participants
according to their scores on two or more instruments. These typologies can be used to guide theoretical sampling in grounded theory studies designed to develop theory.
Combining Data Analysis Techniques
Qualitative and quantitative data sets can be linked,
preserving the numbers and words in each data set.
Or, these data can be transformed to create one data
set, with qualitative data converted into quantitative data, or quantitative data converted into qualitative data (Caracelli & Greene, 1993).
Linking the results of qualitative and quantitative analysis techniques is accomplished by treating each data set with the techniques usually used
with that data; that is, qualitative techniques are
used to analyze qualitative data and quantitative
techniques are used to analyze quantitative data.
For example, constant comparison, qualitative content, and narrative analysis techniques are used to
analyze interview data, whereas one or more statistical techniques are used to analyze data from instruments. The results of the qualitative analysis
of qualitative data and of the quantitative analysis
of quantitative data are then combined at the interpretive level of research, but each data set remains
analytically separate from the other.
In contrast to the process of linking data are
treatments of data that transform one kind of data

COMBINING / SANDELOWSKI

into another kind to create one data set. Tashakkori

and Teddlie (1998, p. 126) referred to the conversion of qualitative data into quantitative data as
quantitizing, and to the conversion of quantitative data into qualitative data as qualitizing.
Quantitizing. Quantitizing refers to a process
by which qualitative data are treated with quantitative techniques to transform them into quantitative
data. The researcher must first reduce verbal or visual data (e.g., from interviews, observations, artifacts, or documents) into items, constructs, or variables that are intended to mean only one thing and
that can, therefore, be represented numerically. One
of the most commonly used examples of this process (and of design template 4 in Figure 1) is the creation of items for an instrument from interview
data. Although the researchers intention is to preserve qualitative meaning in the development of
instruments (Fleury, 1993), any one item can only
have one meaning to function well psychometrically. A valid item in an instrument cannot be meaningful. That is, it cannot be full of meaning; it cannot mean different things at the same or different
times to the same or different people.
Another example of a quantitative transformation aimed at producing a one-meaning unit is the
reduction of narrative data to a variable that can be
correlated with other variables. In a study exemplifying mixed-method design template 4 in
Figure 1, in which quantitative methods prevail
but follow and depend on the results of qualitative
methods, Borkan, Quirk, and Sullivan (1991) elicited narrative data from elders concerning how
they viewed the hip fractures they had suffered.
Using a form of narrative analysis, they determined
that there were two major narrative emplotments
of hip fracture: the mechanical and the organic.
They conducted a series of reliability tests to ensure that each narrative was consistently classified
as mechanical or organic. They then grouped elders according to whether they had told a largely
mechanical or organic story about their hip fracture, and then conducted a correlation study to determine whether and in what direction the emplotments-turned-into-variablesthat is, mechanical
narrative and organic narrativepredicted functional outcomes. They found that the organic story predicted poorer outcomes than the mechanical
story.
This study is an especially good illustration of
the axiom that methods are not uniformly linked
to paradigms. Borkan and his colleagues used a
narrative methodology for positivist ends. They
sought to create a predictor variable by reducing
the interview data they had obtained to units with
only one meaning and then appropriately used re-

253

liability measures to validate them. In contrast, a

constructivist using narrative methodology would
emphasize the meaningfulness of the data and
would treat those data as inherently revisionist
and, thus not amenable to reliability testing.
Quantitative treatments of qualitative data can
also be used to extract more information from
qualitative data, and to confirm researchers impressions from these data. For example, from interview data with infertile couples, we suspected an
association between physicians encouragement
to have an amniocentesis performed and couples
decision to undergo the procedure (Sandelowski,
Harris, & Holditch-Davis, 1991). Accordingly, in
order to validate our impression, we created a data
display to discern the congruence between decision and advice by counting the numbers of couples who were: (a) encouraged to have the procedure and had it (n # 7); (b) encouraged to have the
procedure but did not have it (n # 3); (c) not encouraged or discouraged to have the procedure
and did not have it (n # 7); and, (d) not encouraged or discouraged to have the procedure but did
have it (n # 3). The display visually confirmed our
impression of an association between the decision
to have the procedure and the physicians advice.
We then analyzed these data using Fishers exact
probability test, which showed a nonsignificant
statistical relationship (N # 20; p # .078), which
might reach significance in another study with a
larger sample size.
Qualitizing. Qualitizing refers to a process
by which quantitative data are transformed into
qualitative data. As in quantitizing, qualitizing can
be used to extract more information from quantitative data, or to confirm interpretations of it.
An example of this process is the use of scores
on instruments to profile participantsto create
verbal portraits or typologies of themaround
target phenomena. Tashakkori and Teddlie (1998,
pp. 130133) described five kinds of narrative or
qualitative profiling: modal, average, comparative,
normative, and holistic. A modal profile is a verbal
description of a group of participants around the
most frequently occurring attributes. For example,
if most of the participants in a group are in their
50s, the group can be described as middle-aged. If
most of the participants in a group score in a certain range on a depression inventory, they can be
described as mildly or severely depressed. Similarly, an average profile is a verbal description of
a group of participants around the mean of an attribute. Although not usually identified in research
reports as qualitative profiling, this kind of qualitizing is commonly done to describe samples and
to interpret research results.

254

RESEARCH IN NURSING & HEALTH

A comparative profile is a verbal description

based on the comparison of participants to each
other on one or more sets of scores. In contrast, a
normative profile is a verbal description based on
the comparison of participants scores to the normative scores for one or more instruments. Both
of these kinds of qualitative profiling depend on
the results of quantitative cluster analyses, which
entail a set of statistical techniques to identify homogeneous groups of subjects, or distinctive subgroups in which subjects may be placed because
they are more similar to each other than to subjects
in other subgroups. For example, Rothert and her
colleagues (1990) identified four groups of women
on the basis of their responses to eight scenarios
concerning hormone replacement therapy. That is,
they created a typology of information use that
showed how differently people can respond to the
same sets of information, and that these differences
could be clustered in ways that distinguished
women from each other. After they quantitatively
identified these groups, they interviewed three
women from each of these groups for further clarification and to confirm their typology. Their work
is an example of design template 2 in Figure 1.
Qualitizing entails further studying these theoretical groups to explicate the features of members
in each group that make them like each other and of
the groups themselves that make them different
from each other, and to choose a name that will capture these features. This kind of investigation lends
itself especially well to subsequent grounded theory
study, as the clusters can provide the basis for theoretical sampling and for typology development.
Researchers can then validate these groupings with
further qualitative study or quantitative study, as
exemplified in design templates 5 or 7.
Finally, a holistic profile is a verbal description
based on impressions rather than specific attributes or scores. Holistic profiles may also be comprised of various combinations of modal, mean,
comparative, and normative profiles.
CONCLUSION
Mixed-method research is a dynamic option for
expanding the scope and improving the analytic
power of studies. When done well, mixed-method
studies dramatize the artfulness and versatility of
research design. Mixed-method research operationally includes an almost limitless array of combinations of sampling, and data collection and
analysis techniques, only a few of which could be
described here. Indeed, closer examination of research reports will show researchers have long

been using combinations of qualitative and quantitative techniques, but the qualitative techniques
have not been featured, have been inappropriately
or inadequately used or explained, or have been
explicitly minimized. Sutton (1997) lauded the
virtues of closet qualitative research whereby
researchers conceal or downplay (p. 97) the use
of qualitative techniques in order to have their research reports published!
The combined use of qualitative and quantitative techniques will inevitably be informed by the
researchers viewing position, which shapes what
techniques will be combined, and how and why
they are combined. Accordingly, if researchers
want to combine stories and numbers. . . without
compromise (Ford-Gilboe, Campbell, & Berman,
1995), use numbers and words in a shamelessly
eclectic manner (Rossman & Wilson, 1994), and
ease the uneasy alliance (Buchanan, 1992) between qualitative and quantitative methods, they
must have a clear view of their viewing positions
and what dynamic mixes they suggest or permit.
Researchers must also resist the mix and
match syndrome (Leininger, 1994, p. 103) that
can result in a qualitative quagmire (Barbour,
1998) and that mandates that all research be
mixed-method research. As Chen (1997) suggested, mixed-method research is in danger of becoming the new ideology. Mixed-method research
should never be used because of the misguided assumptions that more is better, that it is the fashionable thing to do, or, most importantly, that qualitative research is incomplete without quantitative
research (Morse, 1996). Indeed, qualitative techniques have been used to salvage quantitative
studies (Weinholtz, Kacer, & Rocklin, 1995). The
completeness of any individual study, no matter
what kind it is, must be judged without resorting
to methodological fads or fetishes.
REFERENCES
Annells, M. (1996). Grounded theory method: Philosophical perspectives, paradigm of inquiry, and postmodernism. Qualitative Health Research, 6, 379393.
Barbour, R.S. (1998). Mixing qualitative methods:
Quality assurance or qualitative quagmire? Qualitative Health Research, 8, 352361.
Berman, H., Ford-Gilboe, M., & Campbell, J.C. (1998).
Combining stories and numbers: A methodologic approach for a critical nursing science. ANS: Advances
in Nursing Science, 21(1), 115.
Borkan, J.M., Quirk, M., & Sullivan, M. (1991). Finding meaning after the fall: Injury narratives from elderly hip fracture patients. Social Science & Medicine, 33, 947957.

COMBINING / SANDELOWSKI

Buchanan, D.R. (1992). An uneasy alliance: Combining

qualitative and quantitative research methods. Health
Education Quarterly, 19, 117135.
Caracelli, V.J., & Greene, J.C. (1993). Data analysis
strategies for mixed-method evaluation designs. Educational Evaluation and Policy Analysis, 15, 195207.
Caracelli, V.J., & Greene, J.C. (1997). Crafting mixedmethod evaluation designs. In J.C. Greene & V.J. Caracelli (Eds.), Advances in mixed-method evaluation:
The challenges and benefits of integrating diverse paradigms (pp. 1932). San Francisco: Jossey-Bass.
Chen, H. (1997). Applying mixed methods under the
framework of theory-driven evaluations. In J.C.
Greene & V.J. Caracelli (Eds.), Advances in mixedmethod evaluation: The challenges and benefits of integrating diverse paradigms (pp. 6172). San Francisco: Jossey-Bass.
Fleury, J. (1993). Preserving qualitative meaning in instrument development. Journal of Nursing Measurement, 1, 135 144.
Ford-Gilboe, M., Campbell, J., & Berman, H. (1995).
Stories and numbers: Coexistence without compromise. ANS: Advances in Nursing Science, 18, 1426.
Greene, J.C., & Caracelli, V.J., (Eds.). (1997a). Advances in mixed-method evaluation: The challenges
and benefits of integrating diverse paradigms. San
Francisco: Jossey-Bass.
Greene, J.C., & Caracelli, V.J. (1997b). Defining and
describing the paradigm issue in mixed-method evaluation. In J.C. Greene & V.J. Caracelli (Eds.), Advances in mixed-method evaluation: The challenges
and benefits of integrating diverse paradigms (pp. 5
17). San Francisco: Jossey-Bass.
Greene, J.C., Caracelli, V.J., & Graham, W.F. (1989).
Toward a conceptual framework for mixed-method
evaluation designs. Educational Evaluation and Policy Analysis, 11, 255 274.
Guba, E.G., & Lincoln, Y.S. (1994). Competing paradigms in qualitative research. In N.K. Denzin & Y.S.
Lincoln (Eds.), Handbook of qualitative research
(pp. 105 117). Thousand Oaks, CA: Sage.
Helitzer-Allen, D.L., & Kendall, C. (1992). Explaining
differences between qualitative and quantitative data:
A study of chemoprophylaxis during pregnancy.
Health Education Quarterly 19, 41 54.
Heron, J., & Reason, P. (1997). A participatory inquiry
paradigm. Qualitative Inquiry, 3, 274 294.
Leininger, M. (1994). Evaluation criteria and critique in
qualitative research studies. In J.M. Morse (Ed.),
Critical issues in qualitative research methods
(pp. 95 115). Thousand Oaks, CA: Sage.
Miles, M.B., & Huberman, A.M. (1994). Qualitative
data analysis: An expanded sourcebook (2nd ed.).
Thousand Oaks, CA: Sage.
Morgan, D.L. (1998). Practical strategies for combining
qualitative and quantitative methods: Applications to
health research. Qualitative Health Research, 8, 362
376.

255

Morse, J.M. (1991). Approaches to qualitative-quantitative methodological triangulation. Nursing Research, 40, 120123.
Morse, J.M. (1996). Is qualitative research complete?
Qualitative Health Research, 6, 35.
Ornstein, S.M., Musham, C., Reid, A., Jenkins, R.G.,
Zemp, L.D., & Garr, D.R. (1993). Barriers to adherence to preventive services reminder letters: The patients perspective. Journal of Family Practice, 36,
195200.
Patton, M.Q. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage.
Pearce, J.C. (1971). The crack in the cosmic egg: Challenging constructs of mind and reality. New York:
Washington Square Press.
Rossman, G.B., & Wilson, B.L. (1994). Numbers and
words revisited: Being shamelessly eclectic. Quality & Quantity, 28, 315327.
Rothert, M., Rovner, D., Holmes, M., Schmitt, N., Talarczyk, G., Kroll, J., & Gogate, J. (1990). Womens
use of information regarding hormone replacement
therapy. Research in Nursing & Health, 13, 355366.
Sandelowski, M. (1995). Triangles and crystals: On the
geometry of qualitative research. Research in Nursing & Health, 18, 569574.
Sandelowski, M., Harris, B.G., & Holditch-Davis, D.
(1991). Amniocentesis in the context of infertility.
Health Care for Women International, 12, 167
178.
Sandelowski, M., Holditch-Davis, D., & Harris, B.G.
(1992). Using qualitative and quantitative methods:
The transition to parenthood of infertile couples. In
J.F. Gilgun, K. Daly, & G. Handel (Eds.), Qualitative
methods in family research (pp. 301322). Newbury
Park, CA: Sage.
Silverman, D. (1993). Interpreting qualitative data:
Methods for analyzing talk, text and interaction. London: Sage.
Sutton, R.I. (1997). The virtues of closet qualitative research. Organization Science, 8, 97106.
Swanson, S.C. (1992). A cross-disciplinary application
of Greene, Caracelli and Grahamss conceptual
framework for mixed method evaluation. Unpublished doctoral dissertation, Georgia State University, Atlanta, GA.
Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining qualitative and quantitative approaches. Thousand Oaks, CA: Sage.
Trost, J.E. (1986). Statistically non-representative stratified sampling: A sampling technique for qualitative
studies. Qualitative Sociology, 9, 5457.
Weinholtz, D., Kacer, B., & Rocklin, T. (1995). Salvaging quantitative research with qualitative data.
Qualitative Health Research, 5, 388397.
Wolfer, J. (1993). Aspects of reality and ways of
knowing in nursing: In search of an integrating paradigm. Image: Journal of Nursing Scholarship, 25,
141146.