The Explanatory Sequential Design

Instead of collecting data at the same time and merging the results, a mixed methods researcher might
collect quantitative and qualitative information sequentially in two phases, with one form of data
collection following and informing the other. This design, also shown in Figure 16.2, is an explanatory
mixed methods design; perhaps the most popular form of mixed methods design in educational
research. An explanatory sequential mixed methods design (also called a two-phase model; Creswell &
Plano Clark, 2011) consists of first collecting quantitative data and then collecting qualitative data to
help explain or elaborate on the quantitative results. The rationale for this approach is that the
quantitative data and results provide a general picture of the research problem; more analysis,
specifically through qualitative data collection, is needed to refine, extend, or explain the general
picture. Referring back to Figure 16.2, you can see that in this design,

◆ The mixed methods researcher places a priority on quantitative data (QUAN) col lection and analysis.
This is done by introducing it first in the study and having it represent a major aspect of data collection.
A small qualitative (qual) component typically follows in the second phase of the research.

◆ The mixed methods researcher collects quantitative data first in the sequence. This is followed by the
secondary qualitative data collection. Researchers often present these studies in two phases, with each
phase clearly identified in headings in the report.

◆ The mixed methods researcher uses the qualitative data to refi ne the results from the quantitative
data. This refi nement results in exploring a few typical cases, probing a key result in more detail, or
following up with outlier or extreme cases.

This design has the advantage of clearly identified quantitative and qualitative parts, an
advantage for readers as well as for those designing and conducting the study. Unlike the convergent
design, the researcher does not have to converge or integrate two different forms of data. This design
also captures the best of both quantitative and qualitative data to obtain quantitative results from a
population in the first phase, and then refi ne or elaborate these findings through an in-depth qualitative
exploration in the second phase. The difficulty in using this design, however, is that the researcher needs
to determine what aspect of the quantitative results to follow up on. This follow-up means deciding on
the participants to sample in the second qualitative phase as well as the questions to ask in this follow-
up phase that builds on the initial quantitative phase. Also, this design is labor intensive, and it requires
both expertise and time to collect both quantitative and qualitative data.

A two-phase project by Ivankova and Stick (2007) is a good example of an explanatory design.
Their research examined factors contributing to students’ persistence in a dis tributed (on-line) doctoral
program in educational leadership in higher education. They called their study a “sequential explanatory
study” (Ivankova & Stick, 2007, p. 93). They first gathered quantitative survey data from 278 current and
former students and then followed up with four qualitative case study respondents to explore the
survey responses in more detail. This project illustrates rigorous quantitative methods using good
sampling and sophisticated data analysis as well as persuasive qualitative case study procedures that
show the development of themes for each case and a cross-case comparison. They present a good fi
gure of their mixed methods procedures showing the quantitative and the qualitative phases, following
by summarizing both results. By studying four specif c cases as a follow-up, they were able to gain
greater insight into the important predictors of student persistence.

The Exploratory Sequential Design

Rather than first analyzing or collecting quantitative data as is done in the explanatory design,
the mixed methods researcher begins with qualitative data and then collects quantitative information.
The purpose of an exploratory sequential mixed methods design involves the procedure of first
gathering qualitative data to explore a phenome non, and then collecting quantitative data to explain
relationships found in the qualitative data. A popular application of this design is to explore a
phenomenon, identify themes, design an instrument, and subsequently test it. Researchers use this
design when existing instruments, variables, and measures may not be known or available for the
population under study. Again refer to Figure 16.2. In this design,

◆ The mixed methods researcher emphasizes the qualitative data (QUAL) more than the quantitative
data (quan). This emphasis may occur through presenting the overarching question as an open-ended
question or discussing the qualitative results in more detail than the quantitative results.

◆ The mixed methods researcher has a sequence to data collection that involves first collecting
qualitative data followed by quantitative data. Typically in these designs, the researcher presents the
study in two phases, with the first phase involving qualitative data collection (e.g., interviews,
observations) with a small number of individuals, followed by quantitative data collection (e.g., a survey)
with a large, randomly selected number of participants.

◆ The mixed methods researcher plans on the quantitative data to build on or explain the initial
qualitative findings. The intent of the researcher is for the quantitative data results to refine and extend
the qualitative findings by testing out an instrument or survey developed using the qualitative findings
or by testing a typology or classifi cation that developed from the qualitative findings. In both cases, the
initial qualitative exploration leads to detailed, generalizable results through the second quantitative
phase.

One advantage of this approach is that it allows the researcher to identify measures actually
grounded in the data obtained from study participants. The researcher can initially explore views by
listening to participants rather than approach a topic with a predetermined set of variables. However, it
has the disadvantage of requiring extensive data collection as well as the time required for this process
is long. The testing of an instrument adds considerably to the length of time this design requires to be
implemented. It also asks researchers to make decisions about the most appropriate qualitative data
(e.g., quotes, codes, themes) to use in the follow-up quantitative phase of the study.

In the exploratory sequential mixed methods design by Meijer, Verloop, and Beijaard (2001), the
authors studied language teachers’ practical knowledge about teaching reading comprehension to 16-
to 18-year-old students. They first conducted a qualitative study of teachers’ practical knowledge from
13 teachers about reading comprehension by collecting semistructured interviews and concept-mapping
assignments. They then used information from this qualitative phase to identify six categories of
teachers’ knowledge. They also used the teachers’ expressions from the qualitative data to form Likert-
type items and scales for the follow-up questionnaire. Thus, the second phase of their study con sisted
of developing and testing an instrument based on their qualitative data. Sixty-nine teachers completed
the questionnaire and the results were used to assess the teachers’ shared practical knowledge about
teaching reading comprehension and the variations that existed among the teachers. A signifi cant
element of this study is that the reader learns about the detailed process of designing and developing an
instrument based on initial qualitative data.

The Embedded Design

A second form of mixed methods design is similar to both the parallel and the sequen tial
design, with some important differences. The purpose of the embedded design is to collect quantitative
and qualitative data simultaneously or sequentially, but to have one form of data play a supportive role
to the other form of data. The reason for collecting the second form of data is that it augments or
supports the primary form of data. The supportive data may be either qualitative or quantitative, but
most examples in the literature support adding qualitative data into a quantitative design. For example,
during a quantitative experiment, the researcher may collect qualitative data to examine how
participants in the treatment condition are experiencing the intervention. Also, the researcher may
collect qualitative data either before or after the experiment to help support the experimental study.
Collecting data before the experiment can help to design an intervention that is tailored to the
participants. Collecting data after the experiment can help to explain and follow up on the quantitative
outcome results. As another example, during a correlational study, the researcher may gather secondary
qualitative data to help understand the reasons for the correlational results. In some embedded designs,
the pro cedures are sequential, with the secondary form of data gathered before the experiment (or
the correlational study) begins (e.g., to help determine the best means for recruiting participants) or
after it concludes (e.g., to follow up and help explain the results).

How does the process of an embedded study work? The researcher collects both quantitative
and qualitative data during a single study (e.g., an experiment or a correlational study), the two datasets
are analyzed separately, and they address different research questions. For example, the quantitative
data will address whether the intervention had an impact on the outcomes, whereas the qualitative
data will assess how the participants experienced the intervention. As shown in Figure 16.2, in this
design:

◆ The mixed methods researcher gives priority to the major form of data collection (e.g., often QUAN)
and secondary status to the supportive form (e.g., often qual) of data collection. The secondary form is
used in the mixed methods study to support and provide additional information to the primary form.

◆ The mixed methods researcher collects both the quantitative and qualitative data simultaneously or
sequentially. Both forms of data are collected during the study at roughly the same time or in sequence.
It is important to understand and describe the purpose for which the secondary data is being collected.
◆ The mixed methods researcher uses the secondary form of data to augment or provide additional
sources of information not provided by the primary source of data. The augmentation is to gather
information that typically addresses a different question than asked for by the primary form of data. For
example, the collection of qualitative data during an experiment may be to understand the “process”
the participants are going through, whereas the quantitative data assesses the impact of the treatment
on the outcomes.

The strength of this design is that it combines the advantages of both quantitative and qualitative data.
Quantitative data are more effective at recording outcomes of the experiment than identifying through
qualitative data how individuals are experiencing the process. It also provides a type of mixed methods
design in which the researcher can collect qualitative data, but the overall design still emphasizes
quantitative approaches. In some fields new to qualitative research, this role of qualitative data helps to
legitimize the use of such forms of data. One challenge in using this design is to be clear about the intent
of the secondary database. In addition, the two databases may not be easily compared because the data
address different research questions. There is also the possibility that introducing qualitative data
collection during an experiment (or correlational study) will influence the outcomes. Strategies need to
be put into place to minimize this effect (e.g., collecting qualitative data at the end of the experiment,
having participants complete journals of their experience that are turned in after the experiment).
Further, like the convergent design, the simultaneous data collection of quantitative and qualitative data
may be labor intensive for a single researcher.

Harrison’s (2007) correlational study of an undergraduate mentoring program in teacher education

illustrates the embedded mixed methods design. Using quantitative longitudinal analysis, she followed
18 undergraduates in a leadership program over 2 years as they learned how to forge mentor–mentee
relationships in an undergraduate teacher education program. Harrison collected quantitative data on
an instrument, the Working Alliance Inventory (WAI), during six administrations over a 2-year period.
This quantitative information represented the major source of information during her study, and her
correlation model suggested that a number of factors (e.g., number of times the mentors–mentees met)
would infl uence the building of positive relationships. She also collected limited data in the form of
three qualitative focus group interviews with the students. She plotted the longitudinal trends in
relationship building using the WAI scores over time, and then used the secondary data, the focus group
information, to help her understand why some mentor–mentees forged closer relationships, plateaued,
or formed more distant relationships over time. Her study was a good example in education of an
embedded design with a major quantitative correlational component and a smaller, sup portive
qualitative focus group element.

The Transformative Design

At a more complex level than the four previous designs, we have the transformative mixed
methods design. The intent of the transformative mixed methods design is to use one of the four
designs (convergent, explanatory, exploratory, or embedded), but to encase the design within a
transformative framework or lens (Creswell & Plano Clark, 2011).
 This framework provides an orienting lens for the mixed methods design. It informs the overall
purpose of the study, the research questions, the data collection, and the outcome of the study. The
intent of the framework is to address a social issue for a marginalized or underrepresented population
and engage in research that brings about change. Thus, strength of this design is that it is value-based
and ideological ( Greene, 2007). The typical frameworks found in mixed methods are feminist, racial,
ethnic, dis ability, and gay or lesbian perspectives. A challenge in using this design is that we are still
learning about how to best integrate the framework into a mixed methods study.

A diagram of the procedures is found in Figure 16.2. For purposes of discussion, this figure
shows the use of a transformative lens within an explanatory sequential design (i.e., quantitative data
collection followed by qualitative data collection). In this design:

◆ The mixed methods researcher uses either a convergent, explanatory, exploratory, or embedded
design. The basic designs provide the cornerstone for the transformative design, but the transformative
design goes beyond simply the use of the basic design.

◆ The mixed methods researcher uses an overall orienting lens in the study as a transformative
framework. This framework may be a feminist perspective, a racial or ethnic perspective, or some other
perspective. It is this framework that shapes many aspects of the mixed methods design, such as the
framing of the title, the questions, the methods, and the conclusions. The framework basically addresses
an issue for an underrepresented group and presents research intended to bring about change for that
group.

◆ The mixed methods researcher calls for change that will address the social issue faced by the group
under study. A strong key to a good transformative mixed methods study is whether the research calls
for reform or changes at the end of the study. This call may be an explicit request for change or steps
that will be required to bring about change.

In a sequential explanatory mixed methods study by Buck, Cook, Quigley, East wood, and Lucas
(2009), the authors used a feminist lens to study 89 African American girls’ personal orientations toward
science learning. The study begins with a theoretical framework about the writings of feminist
researchers in science education and what it means to be a girl in science as well as an African American
girl in science. In this sequential design, a quantitative first phase study of the girls’ attitudes toward
science was collected on an inventory to provide descriptive results. This was followed by a qualitative
second phase consisting of focus group interviews and themes related to girls’ definition of science, the
importance with science, and their experiences and success in school science. As a final step, the
authors linked the qualitative themes to categories. In the conclusion of the article, the authors call for
reforming the instructional strategies in science for girls to be positively connected to science, and they
comment how their study has “illustrated the use of a feminist lens in a study in which specific myths
have been dispelled and recommendations for change made.” (p. 408)

Multiphase Design
 Like the transformative design, the multiphase design is a complex design that builds on the
basic convergent, explanatory, exploratory, and embedded designs. Multiphase mixed methods designs
occur when researchers or a team of researchers examine a problem or topic through a series of phases
or separate studies. The groups of phases or studies are considered to be a mixed methods design and
the intent of the design is to address a set of incremental research questions that all advance one
programmatic research objective (Creswell & Plano Clark, 2011). The phases or studies may employ a
combination of concurrent or sequential designs and this form of design is popular in large-scale health
research and in evaluation research. The strength of this design lies in the use of multiple projects to
best understand an overall program objective. Challenges include forming a research team that can
work comfortably together given diverse method orientations, making sure that the phases or studies
link together, and having all of the studies provide insight into an overall project objective. As shown in
Figure 16.2, the major elements of this design are:

◆ The mixed methods researchers use either a convergent, explanatory, exploratory, or embedded
design in multiple phases or projects in the study. The multiphase design builds on the basic mixed
methods designs and adds to these designs multiple phases or projects conducted over time. Any one
phase may have a combination of concurrent and sequential mixed methods designs. In addition, this
form of research is most amenable to large-scale funded investigations.

◆ The mixed methods researchers need to clearly identify projects or phases that help address a larger
program objective. These researchers also need experience in large scale research. Teams might be
composed of individuals with quantitative, qualitative, and mixed methods research skills.

◆ The mixed methods researchers need to interrelate the different phases or projects so that they tie
together to address a common research objective. Typically, one phase or project leads to another and,
in this sense, the phases or projects build on (or inform) each other throughout the study.

In a multiphase mixed methods study, Nastasi et al. (2007) provide a program development
research study of culture-specific definitions of mental health constructs (e.g., stressors, competencies)
for adolescents in SriLanka. This study was part of the Sri Lanka Mental Health Promotion Project. In the
beginning of the article we are introduced to twelve different phases in the research project that form
formative research, instrument development, program development, and evaluation research. At these
different phases, the researchers engaged in combinations of quantitative and qualitative research,
some presented as concurrent and some as sequential. A table in their article showed how they
combined the qualitative and quantitative data in each phase (e.g., in the fi nal evaluation phase they
combined qualitative and quantitative results based on an experimental design and post-intervention
interviews). In this study, they used the basic mixed methods designs in different phases, conducted
research toward the program objective of establishing culturally relevant understanding of mental
health, and engaged in this large scale investigation over several years involving at least six primary
researchers.
When Maria starts with a survey and follows it up with an interview in her mixed methods study, what
type of design did she use? What would be the possible priority and sequence for her design? What
would be the reason for using this type of design?