You are on page 1of 20

The current issue and full text archive of this journal is available at www.emeraldinsight.com/1352-2752.

htm

Mixed methods designs in marketing research


Robert L. Harrison
Western Michigan University, Kalamazoo, Michigan, USA, and

Mixed methods designs

Timothy M. Reilly
University of Nebraska-Lincoln, Lincoln, Nebraska, USA
Abstract
Purpose The purpose of this article is to determine the uses of mixed method research designs published in major marketing journals. Design/methodology/approach This study involved a content analysis of 2,166 articles published between 2003 and 2009 in nine prominent marketing journals. Findings A total of 34 mixed method studies implemented data-collection procedures sequentially (79 percent), eight implemented them concurrently (19 percent) and one combined both sequential and concurrent procedures (2 percent). On the whole, priority was skewed more toward quantitative strands, with 27 articles prioritizing quantitative data (63 percent), three articles prioritizing qualitative data (7 percent), and 13 articles prioritizing both equally (30 percent). Research limitations/implications It is clear that marketing scholars recognize the benet of mixing qualitative and quantitative research; however, as a discipline we are not demonstrating knowledge of the mixed method literature or procedures, as only one article recognized or mentioned knowledge of mixed method procedures or cited mixed method research. Practical implications This study provides guidance for researchers in identifying design types appropriate for various rationales or research objectives and models of different design types that have been published in marketing journals. In addition, implications for designing mixed methods studies in marketing include highlighting the need for scholars to specically address issues such as the timing and priority given to each data type (i.e. sequential or concurrent), and the integration (or mixing) of the both data types. Originality/value Until now, the role of mixed methods designs in marketing has not been the subject of formal examination. The delineation of the major forms in mixed method designs provides a framework for looking at such design types, which helps to provide more credibility to the eld of marketing by providing examples of research designs that are substantially different than single strand studies. Keywords Research methods, Marketing, Publications Paper type Research paper

In marketing, mixed methods research has received little coverage, despite the apparent movement in many of the social sciences toward such research designs. While national research foundations such as the National Institute of Health, the National Science Foundation, the National Research Council, and others have offered workshops promoting the use of mixed method research techniques (Creswell et al., 2003), discussions in marketing regarding this methodological approach have focused
The authors would like to thank Amanda Garrett, John Creswell, and Vicki Plano Clark in the Ofce of Qualitative and Mixed Method Research, University of Nebraska Lincoln, Ann Veeck, and James W. Gentry.

Qualitative Market Research: An International Journal Vol. 14 No. 1, 2011 pp. 7-26 q Emerald Group Publishing Limited 1352-2752 DOI 10.1108/13522751111099300

QMRIJ 14,1

primarily on the philosophical assumptions that guide qualitative, quantitative, and mixed method research approaches (Bahl and Milne, 2006). The post-positivist and interpretivist paradigms have been discussed in terms of their ontological, axiological and epistemological assumptions about the nature, values, and epistemological assumptions that guide research (Hudson and Ozanne, 1988). An important assumption guiding post-positivist research mandates an objective view of reality, in which research is aimed to measure or explain, creating knowledge that is generalizable across different people, time, and place (Hudson and Ozanne, 1988). The interpretivist ontology assumes the existence of multiple realities that are socially constructed and is focused on understanding behavior rather than predicting it. It is argued that mixed method research with its pragmatic approach does not align itself with a single system or philosophy (Creswell et al., 2003), and instead is most often driven by the research question, rather than being restrained by paradigmatic assumptions ( Johnson and Onwuegbuzie, 2004). Paradigmatically, mixed method research makes use of pragmatism as a system of philosophy. The logic of pragmatic inquiry includes the use of induction (or discovery of patterns), deduction (testing of theories and hypotheses), and abduction (uncovering and relying on the best of a set of explanations for understanding ones results) ( Johnson and Onwuegbuzie, 2004; Morgan, 2007). The mixing of methods in social research has been given many names including blended research, integrative, multi-method, multiple methods, triangulated studies, ethnographic residual analysis, and mixed research. In marketing, multi-method and mixed method research are the most commonly used labels. In the Handbook of Mixed Methods Research, distinctions have been made between these two terms: (1) multi-method which involve multiple types of qualitative inquiry (e.g. case study and ethnography) or multiple types of quantitative inquiry (surveys and experiments); and (2) mixed methods which involve the mixing of the two types of data (Morse, 2003). Mixed methods research has become the most popular term for mixing qualitative and quantitative data in a single study (Johnson et al., 2007) and is the label used in this study. The denition of mixed methods research used in this study is one based on an analysis of denitions used by leaders in the eld of mixed methods research:
Mixed methods research is the type of research in which a researcher or team of researchers combines elements of qualitative and quantitative research approaches (e.g. use of qualitative and quantitative viewpoints, data collection, analysis, inference techniques) for the broad purpose of breadth and depth of understanding and corroboration ( Johnson et al., 2007, p. 123).

This denition will be used in this study, as published marketing research will be evaluated, particularly addressing the use of two strands of data (i.e. qualitative and quantitative), the timing (i.e. sequential or concurrent), the priority given to each data type, and the integration (or mixing) of the data. The development of the mixed method research approach began in the 1980s, when scholars began expressing concerns about the mixing of quantitative and qualitative data without articulating defensible reasons for doing so (Greene et al., 1989). Subsequently, scholars began identifying a number of rationales for combining data-collection methods and research questions particular to different mixed method research designs. In a study,

investigating the reasons given for combining qualitative and quantitative research, Bryman (2006) identied 16 rationales for conducting mixed method research. Descriptions of these rationales are listed in Table I, in addition to an added column that highlights what is identied as appropriate mixed method design type for each rationale. These design types are described in detail in the Findings section. In a content analysis of research published between 1993 and 2002 in three prominent marketing journals, 173 articles were found to mix qualitative and quantitative data (Hanson and Grimmer, 2007). Implicitly, the marketing discipline encourages mixed methods research because of the emphasis on rigorous research (Woodruff, 2003) and, as Hunt (1994) pointed out, research using qualitative methods could usefully complement quantitative analyses. Nonetheless, though the potential of qualitative research is undisputed, mixed methods studies in marketing are relatively rare (Hanson and Grimmer, 2007). Again, the existing literature has taken the beginning steps towards understanding this methodological approach by rst discussing the philosophical assumptions of such research (Bahl and Milne, 2006) and identifying trends as far as the numbers of studies employing the approach (Hanson and Grimmer, 2007). This study takes the next step in this methodological discussion by updating the recent trends in the use of mixed methods research and identifying trends in terms of the types of mixed methods designs being employed in marketing research. Despite the availability of mixed methods-related books, chapters, and journal articles, the use of mixed methods research designs in marketing still is quite limited. The general absence of mixed method research designs may be due to a number of factors including the historical precedent of favoring quantitative research in marketing (Hunt, 1994), the general lack of attention to diverse methodological approaches in graduate education and training, and the difculty in learning both method types. Scant methodological coverage by marketing scholars suggests a need to answer the following question: What types of mixed method research designs have been published in marketing? Further, marketing scholars must be able to assess the appropriateness of the different design choices and anticipate challenges with each choice. We address this issue by proving an overview of mixed method design types and an examination of how these designs are successfully being applied and reported in mainstream marketing journals, providing a guide for future researchers conducting mixed method marketing research. The paper is structured as follows. First, rationales for conducting mixed methods research are discussed. This is followed by the presentation of a content analysis of the mixed method marketing research including outlines of the four major mixed method designs. The paper concludes with a discussion of the methodological implications and future research recommendations. Content analysis and data-collection procedures In this paper, mixed method articles were reviewed across nine journals from 2003 to 2009: Journal of Consumer Research, Journal of Marketing, Journal of Marketing Research, Journal of the Academy of Marketing Science, Journal of Retailing, Journal of Consumer Psychology, Marketing Science, International Journal of Research in Marketing, and European Journal of Marketing. The rationale for selecting these journals is that JM, JMR, JCR, and Marketing Science are listed as the A-level marketing journals according to the American Marketing Association, with JCP, JR,

Mixed methods designs

10

QMRIJ 14,1

Rationale Concurrent

Triangulation

Quantitative and qualitative combined to triangulate ndings to be mutually corroborated Offset Combining strands offsets their weaknesses to draw on the strengths of both Completeness Bringing together a more comprehensive account if both quantitative and qualitative research is employed Process Quantitative provides an account of structures in social life but qualitative provides sense of process Different research questions Quantitative and qualitative each answers different research questions Explanation One is used to help explain ndings generated by the other Unexpected results When one strand generates surprising results that can be understood by employing the other Instrument development Qualitative is employed to develop questionnaire and scale items Sampling One approach is used to facilitate the sampling of respondents or cases Credibility Employing both approaches enhances the integrity of ndings Context Qualitative providing contextual understanding coupled with either generalizable, externally valid ndings or broad relationships among variables uncovered through a survey Illustration Qualitative to illustrate quantitative ndings (putting meat on the bones of dry quantitative ndings) Utility Among articles with an applied focus, the combining the two approaches will be more useful to practitioners and others Conrm and discover This entails using qualitative data to generate hypotheses and using quantitative research to test them within a single project Diversity of view Combining researchers and participants perspectives through quantitative and qualitative research, respectively, and uncovering relationships between variables through quantitative research while also revealing meanings among research participants through qualitative research Explanatory

Note: aAdded classication Source: Adapted from Bryman (2006)

Table I. Rationale for mixed methods research and design types Description Design typea Concurrent Exploratory, explanatory, or concurrent Exploratory or explanatory Concurrent Explanatory Explanatory or embedded Exploratory Exploratory or explanatory Exploratory, explanatory, or concurrent Exploratory or explanatory Exploratory, explanatory, concurrent, or embedded Exploratory Concurrent or embedded

JAMS, IJMR and EJM also listed as top-level journals according to some surveys (Theoharakis and Hirst, 2002). These journals were chosen to provide data from eminent journals and it is assumed that they represent trends in best practice research. The rationale for selecting articles after 2003 coincides with the publication of the Handbook of Mixed Methods Research (Tashakkori and Teddlie, 2003). Mixed method scholars often refer to studies published before and after this time period as pre- and post-handbook research, respectively. Further, the intent of the paper is to provide a specic content analysis of mixed method research, analyzing more recent trends in marketing research following Hanson and Grimmers (2007) lead. The purpose of the Hanson and Grimmer article was to assess the mix of qualitative and quantitative research being published (Hanson and Grimmer, 2007, p. 60). While they also identied trends in the sheer number of mixed method studies published in three major marketing journals before 2002, this study examines research conducted after 2002 and includes a focused discussion of the different mixed methods research design types being used. Because the focus of their study was not mixed methods research specically, their denition of mixed methods may have allowed for the inclusion of studies that discuss the use of multiple data sources, without the mixing of the quantitative and qualitative results. It is clear from the number of articles found by Hanson and Grimmer that mixed methods research is being conducted by marketing scholars. What is unclear are mixed method design trends in the marketing discipline, that is, what types of mixed methods designs are being used in marketing studies. Other disciplines have conducted reviews in this manner to gauge trends with regard to mixed method design types within the eld (Hanson et al., 2005; Johnson et al., 2007; Plano Clark et al., 2008; Powell et al., 2008; Truscott et al., 2010). Content analysis is an observational technique which allows for a systematic evaluation of recorded communications (Kolbe and Burnett, 1991). In the current investigation, the recorded communications of interest are journal articles within the marketing journals previously identied. For the purpose of coding, articles were required to contain qualitative research centered on primary data collection in non-numerical form (words, images, symbols, etc.) and quantitative research centered on data collection in numerical form. To identify mixed method research studies, we used a coding scheme developed by the founding editor of the Journal of Mixed Methods Research, who uses these criteria when judging article t with his journal (Table II). Consistent with this scheme and others (Bryman, 2006), studies that foreground the fact that they are using both qualitative and quantitative research, but either report just

Mixed methods designs

11

X Quantitative rigor Use of theory Good sampling procedure Good sample size Identication of instruments Validity and reliability of scores Detailed procedures of data collection Level of sophistication of data analysis

X Qualitative rigor Emerging design type of design mentioned (e.g. case study, ethnography) Good sampling procedure Good sample size Questions for data collection Procedures for data collection Coding/thematic development Advanced level of use themes (e.g. taxonomy, chronology)

Table II. Mixed method scoring scheme

QMRIJ 14,1

12

the quantitative or qualitative data, were not included in the study. Content analyses or studies that quantify qualitative data were excluded from this study. Multiple approaches were used to identify articles of the study. The rst approach involved keyword searches for phrases such as quantitative and qualitative, multi-method, mixed method, or triangulation appeared in the title, keywords, or abstract. Relatively, few articles were generated by this search. The second approach involved manually searching each article in each journal. This process was accelerated through the use of online databases such as Science Direct, which offers search tools that include bookmarks that outline each article, and Emerald and Springerlink that provide HTML fulltext options that made for more rapid scrolling through methods and results sections. All 2,596 articles published in the selected journals from 2003 to 2009 were examined and 99 articles were identied that mentioned the use of both qualitative and quantitative methods. These articles were then limited to include studies that reported both qualitative and quantitative results, rather than merely mentioning that such data were collected. From this search, 43 articles met the denition of mixed methods research and were included in the purposive sample. These articles were then sorted to identify examples of research employing different mixed method design types (Table III). Table IV lists the design features of each selected article. Included as a design feature is the priority given to the different data strands. To highlight priority and timing, Morse (1991, 2003) developed a procedural notation system that uses plus ( ) symbols and arrows ( ! ), and capital and lowercase letters to represent different mixed method procedures. A plus sign indicates that both data strands were collected concurrently and an arrow indicates that the data were collected sequentially. Capital letters are used to indicate higher priority for a particular method, with lowercase indicating lower priority. For example, QUAL ! quan indicates a sequential study where qualitative data was collected prior to the quantitative data, and that top priority is given to the qualitative data. QUAL QUAN indicates that both data strands were collected at the same time and were given equal priority. This notation system is featured in Table IV. Findings The timing, or sequence, in mixed method designs refers to the order in which researchers use the two data strands. Timing procedures vary by design type, with concurrent designs employing concurrent data-collection procedures, explanatory and exploratory designs employing sequential data collection, and embedded designs employing either sequential or concurrent data-collection procedures. In this study, 34 mixed method studies implement data-collection procedures sequentially (79 percent), and eight implemented them concurrently (19 percent), and one combined both sequential and concurrent procedures (2 percent). On the whole, priority was skewed more toward quantitative strands, with 27 articles prioritizing quantitative data (63 percent), three articles prioritizing qualitative data (7 percent), and 13 articles prioritizing both equally (30 percent). This is not surprising as it has been suggested that most articles employing mixed methods in marketing literature have a positivist orientation (Bahl and Milne, 2006). Exploratory designs Four mixed methods design types appeared in the selected marketing journals during the designated time period. In the rst type, i.e. exploratory designs, researchers rst

Study Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential Sequential exploratory Sequential exploratory Sequential exploratory Instrumental development Taxonomy Typology exploratory exploratory exploratory exploratory exploratory exploratory exploratory explanatory exploratory exploratory exploratory exploratory exploratory exploratory exploratory exploratory exploratory exploratory exploratory Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Follow-up explanations Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development Instrument development

Design

Variant

Priority qual ! QUAN qual ! QUAN qual ! QUAN QUAL ! quan qual ! QUAN qual ! QUAN qual ! QUAN QUAN ! qual QUAL ! QUAN qual ! QUAN qual ! QUAN QUAL ! QUAN qual ! QUAN qual ! QUAN qual ! QUAN qual ! QUAN QUAL ! QUAN QUAL ! QUAN QUAL ! QUAN qual ! QUAN QUAL ! QUAN QUAL ! QUAN (continued )

Walsh and Beatty (2007) Wolnbarger and Gilly (2003) Bruner and Kumar (2007) Commuri and Gentry (2005) Grayson and Martinec (2004) Hand et al. (2009) Russell et al. (2004) Thomson (2006) Ulaga and Eggert (2006) Arnold and Reynolds (2003) Venable et al. (2005) Du et al. (2008) Grifths-Hemans and Grover (2006) Brennan et al. (2003) Luna et al. (2008) Avlonitis and Indounas (2007) Argouslidis and McLean (2004) Mekonnen et al. (2008) Marcella et al. (2002) and Marcella and Davies (2004) Botti et al. (2009) Sogn-Grundvag and Ostli (2009) Kirmani and Campbell (2004)

Mixed methods designs

13

Table III. Design features of mixed methods studies published in marketing

14

QMRIJ 14,1

Study Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential explanatory Sequential embedded Sequential embedded Sequential embedded Embedded Embedded Embedded Concurrent convergence Concurrent convergence Concurrent convergence Concurrent convergence Sequential exploratory with concurrent follow-up Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Follow-up explanations Embedded experimental model Embedded experimental model Embedded experimental model Embedded methodology Embedded experimental Embedded experimental Convergence model Convergence model Convergence model Typology Hybrid

Steenkamp and Geyskens (2006) Dellande et al. (2004) Hewett et al. (2006) Coulter et al. (2005) Jayachandran et al. (2004) Horsky et al. (2004) Grace and OCass (2003) McMullan and Gilmore (2008) West and Prendergast (2009) Paul et al. (2009) Dahl and Moreau (2007) Scott and Vargas (2007) Harris and Blair (2006) Daly et al. (2009) Andrade and Ho (2009) Lau-Gesk and Meyers-Levy (2009) Coulter et al. (2003) Mathwick et al. 2008 Coelho and Easingwood (2008) Voorhees et al. (2006) Simoes et al. (2005)

Note: n 34

Table III. Design Variant Priority QUAN ! qual QUAN ! qual QUAN ! qual QUAN ! qual QUAN ! qual QUAN ! qual QUAN ! qual QUAL ! quan QUAN ! QUAL qual ! QUAN QUAN (qual) QUAN (QUAL) qual ! QUAN QUAL QUAN QUAN (qual) QUAN (qual) QUAL quan QUAL QUAN QUAL QUAN qual QUAN (Qual ! Quan) ! (Qual Quan)

Design type Variants Concurrent

Timing

Weighting Usually equal Unequal

Mixing

Notation

Mixed methods designs

Convergence Concurrent: quantitative and qualitative at the same time Concurrent or Embedded Embedded experimental sequential Embedded correlation Embedded methodology Explanatory Follow-up Sequential: explanations quantitative followed by a qualitative Exploratory Instrument Sequential: development qualitative followed by Taxonomy a quantitative

Merging the data QUAN QUAL during the interpretation or analysis Embed one type of QUAN (qual) or QUAL (quan) data within a larger designusing the other type of data QUAN ! qual

15

Usually Connect the data quantitative between the two phases Usually qualitative Connect the data between the two phases

QUAL ! quan

Note: Words in italics denote added variant Source: Adapted from Creswell and Plano Clark (2007)

Table IV. Major mixed methods design types

collect qualitative data, analyze the qualitative data, and then build on the qualitative data for the quantitative follow-up. Building can involve identifying the types of questions that might be asked, determining the items/variables/scales for instrument design, and generating a typology or classication. Exploratory designs were the most common type of design used (47 percent, n 20). Exploratory designs are useful for exploring relationships when study variables are unknown; developing new instruments, based on initial qualitative analysis; generalizing qualitative ndings; and rening or testing a developing theory. Exploratory designs are most often conducted when qualitative data are only an initial exploration to identify variables, constructs, taxonomies, or instruments for quantitative studies (Creswell et al., 2003). Two common variants of this design type are the instrument design model and the taxonomy development model. In instrument development designs, qualitative ndings are used to develop scale items for a quantitative survey instrument. In taxonomy development designs, qualitative results are used to develop a taxonomy (or classications system), or to develop an emergent theory, and the secondary, quantitative phase tests or develops the emergent theory, or tests or studies the quantitative results in more detail (Creswell et al., 2003). To explain, a research question employing this design-type might read, To what extent do the qualitative ndings generalize to a specied population? Though it was found that marketing scholars rarely discussed rationales for conducting mixed method research, Grayson and Martinec (2004, p. 299) described explicitly their reasons for using both types of data:
Rather than assuming that our conceptualization is appropriate for understanding and meaning perceptions of authenticity at two research sites, we used open-ended interview questions to hear, in consumers own words, how they experienced the sites.

QMRIJ 14,1

16

In one example from marketing literature, Commuri and Gentry (2005) used an exploratory design to investigate household resource allocation in households where wives earn more money than their husbands. The authors interviewed 20 couples a total of 64 times and developed a grounded theory. They then tested implications of the emergent theory in an online survey among 126 married participants. The results from the qualitative and quantitative data were presented separately and mixed in the Discussion section. Arnold and Reynolds (2003) also used a sequential exploratory design, but to develop an instrument. They described the scale development process in three phases. The qualitative inquiry phase included 98 depth interviews, coded thematically by a three-member coding team. Phenomenological meaning statements were used to develop domain denitions and survey items, and a table of statements was included in an appendix. The scale purication phase involved detailed item analysis, exploratory factor analysis, conrmatory factor analysis, an assessment of scale reliability, unidimensionality, and convergent and discriminant validity. The scale validation phase activities involved the replication of the conrmatory factor structure on an independent sample, assessment of nomological validity by correlating the emergent constructs with existing constructs, and performing a cluster analysis. The ndings from the two data strands were presented separately. About 14 other studies used a similar sequential exploratory design to develop an instrument (Table II). Kirmani and Campbell (2004) used an exploratory design to investigate how consumers respond to inuence attempts by salespeople and service personnel. Three qualitative data sets included consumer diaries from 36 participants, 34 semi-structured interviews, and nine depth interviews. The qualitative data were used to develop a response strategy typology. A total of 96 participants then participated in an experimental study to test relationships within the developed typology. The results from the qualitative study and the experimental study were presented separately and were mixed in the Discussion section. Voorhees et al. (2006) also used sequential exploratory designs to test typological relationships. The exploratory design was found to be the dominant design type used in this study. Of the articles that were excluded from this study, most of the articles did not fully develop or analyze the qualitative strand of the study. The benet of mixing methodologies is utilizing the strengths of each approach. Many of the studies merely discussed conducting interviews to develop scales. The strength in qualitative research is the depth of knowledge gained from analyzing the experience of participants. Interviewing respondents only indicates a surface level of understanding a phenomenon. The multi-layered analytic approach used in qualitative methodologies allow for a much deeper understanding of a phenomenon and in turn should lead to the development of more specic and focused questions to be asked in the quantitative follow-up phase. Explanatory designs In explanatory designs, researchers rst collect and analyze quantitative data, then build on those ndings in a qualitative follow up, which seeks to provide a better understanding of the quantitative results. Building can involve either using the quantitative data to select cases or to identify questions that need further explorations in the qualitative phase. A total of 11 articles were found that employed explanatory designs (25 percent). In explanatory designs, quantitative results are collected and followed by qualitative data

and unequal priority is given to the quantitative data. As the label implies, the explanatory design is useful in explaining relationships or study ndings. Explanatory designs are most often conducted when qualitative data are needed to help explain or build on initial quantitative data. Two variants of explanatory designs include follow-up explanations and participant selection models (Creswell et al., 2003). In follow-up explanation models, specic qualitative results are used to explain or expand on quantitative results. For example, statistical differences among groups, individuals who scored at extreme levels, or unexpected results are explored qualitatively. A research question employing this design-type might read, How do qualitative data explain the quantitative results? In both exploratory and explanatory designs the data sets, or strands, are usually connected, or mixed, during the interpretation stage and in the Discussion section. Jayachandran et al. (2004) used an explanatory study of consumer response capability. After the development of a conceptual model, (31) depth interviews were conducted to triangulate the study ndings from the survey data and to generate further understanding of the underlying phenomenon ( Jayachandran et al., 2004, p. 225). The authors then discussed the rationale for mixing the data:
Since the objective of this (qualitative) phase of the research study was to gain further understanding of the results phase, we integrate the discussion of the ndings from the depth interviews with that of the survey results ( Jayachandran et al., 2004, p. 255).

Mixed methods designs

17

Four other studies used sequential explanatory designs, all of which employed the follow-up variant. Overall, the articles that collected and analyzed quantitative data rst presented a much more thorough analysis and presentation of the qualitative data in the second phase. This may be explained by the nature and rationale for the design type. For the most part, the qualitative phase was included to address specic issues that resulted from the quantitative phase. Because the issues were focused, the qualitative study was more specic in its development and analysis. An overall trend in the marketing literature is the movement away from survey data in recent years, as scholars are questioning the reliability of survey data. The explanatory design is an alternative that may potentially address these concerns. The strength in survey research is to uncover generalizable trends in specic population and qualitative follow-up can be used to address gaps that results from unique ndings. Thus, explanatory design has the potential to illuminate the strengths of survey data. Embedded designs In embedded designs, researchers collect both quantitative and qualitative data either sequentially or concurrently with one form of data playing a supporting role, or both forms of data playing a supporting role in a larger design. In the former design, the qualitative data may play a supporting role within an experiment or correlational study. In the latter, the qualitative and quantitative data play a supportive role in a case study, ethnography, narrative, or other qualitative research design. This latter variant is identied as embedded methodology in Table I. A key question with embedded designs is whether the secondary data type is playing a supplemental role, i.e. would the results of the secondary data type be meaningful if it were not embedded within the other data (Creswell and Plano Clark, 2007).

QMRIJ 14,1

18

Six studies were found employing embedded designs (14 percent). Embedded designs are most often conducted when there are different questions requiring different data. The two variants of embedded designs include experimental models, wherein qualitative data are embedded in experimental, and correlational models, in which qualitative data are used to explain how mechanisms work in a correlation model (Creswell et al., 2003). For example, a research question employing this design-type might read, How do the qualitative ndings enhance the interpretation of the experiments, or correlational outcomes? Scott and Vargas (2007) employed an embedded design in their study of consumer response to pictures as a form of writing. The qualitative phase was embedded in a larger experimental design to understand better what aspects of each picture respondents were using to deduce the intended message. Following an experimental study, 20 interviews were conducted to understand how mechanisms worked in the experiment. Specically, the interviews examined what aspects of images were used by respondents to deduce the intended message. This was followed by an experimental study of the emergent aspects and other relationships. Dahl and Moreau (2007) also employed an embedded design in their study to understand motivations and enjoyment in creative activities. They employed an experimental model, wherein 12 semi-structured interviews were used to shape the intervention. Using central themes identied in the interviews, the authors tested experimentally the roles of constraints in the creative experience. They then identied and tested empirically an important moderator of the creative experience, namely poor skill level. The results from the qualitative and quantitative data were presented separately in the Theoretical contributions section of the article. The Daly et al. (2009) study offered the closest example of an embedded methodology design in their examination of the value of improvisation training for service employees. While the qualitative case study method was the overarching design used in this study, both qualitative and quantitative data were collected simultaneously, then analyzed and mixed in the Results section. Concurrent designs In concurrent designs, researchers collect both qualitative and quantitative data simultaneously, analyze both data strands separately, and then mix the databases by merging the data. Five articles were found employing concurrent designs (12 percent). Concurrent designs are most often conducted to bring together the strengths of both quantitative and qualitative research to compare results or to validate, conrm, or corroborate quantitative results with qualitative ndings. In convergence design models, qualitative and quantitative data are analyzed separately and the different results are converged during the interpretation (Creswell et al., 2003). A research question employing this design-type might read, To what extent do the qualitative results conrm the quantitative results? Mathwick et al. (2008) collected both qualitative and quantitative data in their concurrent design study examining the relational norms that determine social capital in virtual communities. The authors collected online survey data from 1,011 visitors to a virtual community website. They also collected observational data using netnography, that is, they analyzed discussion threads to develop insight into community interactions (Mathwick et al., 2008). A measurement model was estimated using the quantitative data

with individual item reliability and convergent and discriminate validities being assessed (Mathwick et al., 2008). Themes that emerged from the qualitative data provided additional support for their initial characterization of virtual activity. The themes were discussed until the authors achieved interpretive convergence. The results for each data strand were presented separately and mixed in the Discussion section. Voorhees et al. (2006) used a concurrent design to investigate non-complainers, or customers who experience service failures but do not voice complaints. They used a critical incident (CIT) survey that included both qualitative and quantitative sections to gather data. The rst goal of this study was to explore the reasons why customers do not complain following dissatisfactory service experiences. In this study, qualitative data were gathered from consumers in which they described a recent service experience with which they were dissatised and then scale items were used to gather quantitative data surrounding their response to said incident. The results of the qualitative section of their CIT survey reveal six general categories and 16 subcategories of non-complainer motivations. The second goal was to compare non-complainers with complainer groups across several key outcome variables. The results of the quantitative section of the CIT survey indicate that the levels of repurchase intentions and negative outcomes for non-complainers are in among the other investigated groups. Collectively, the ndings both supported and extended current knowledge of the motivations of non-complainers and the consequences of their actions (Voorhees et al., 2006). The results were presented separately and mixed in the Discussion section. Because the data in concurrent design are collected simultaneously and equal weight is given to both strands, the nature of this design lends itself to rigorous collection and analysis in both data strands. While all of the concurrent design studies asked and answered different research questions with each data strand, a few did not provide equal weighting and may have illuminated more with a balanced design. Hybrid design One study was found to incorporate aspects of multiple design types (2 percent). This study, not included in the original classication scheme, is labeled a hybrid design. A hybrid design may be used when multiple rationales require the use of multiple designs. This study developed a measure of the corporate identity management construct drawing on literature, in-depth interviews, survey data, and a follow-up phase (Simoes et al., 2005). The rst two phases of the study are indicative of an exploratory design as they interviewed managers to understand better their experience in order to develop an instrument. The instrument was developed using exploratory factor analysis, with discriminant and convergent validity being assessed. The authors then incorporated a two-part follow-up concurrent design phase to further validate, interpret and develop the ndings (Simoes et al., 2005, p. 160). This validation phase included interviews with ten managers and a follow-up questionnaire sent to 110 (27 returned) managers who had been involved with the initial survey to assess the representativeness of the scale. The data were presented separately and mixed in the Discussion section. Discussion Mixed methods research allows researchers to be more exible, integrative, and holistic in their investigative techniques, as they strive to address a range of complex research questions that arise (Powell et al., 2008, p. 306). However, mixed methods

Mixed methods designs

19

QMRIJ 14,1

20

researchers have suggested that there is a need for distinguishing between studies that utilize the two types of data without serious integration and studies that mix the data sets effectively (Tashakkori and Creswell, 2007). The benets of such integration include, but are not limited to, the possibility of redirecting one or both studies given earlier results, adding questions to a revised questionnaire so that the quantitative data address the issue more systematically, or seeing new issues from other responses to surveys that can be supplemented qualitatively. A study that includes both data types without integration is merely a collection of methods. Strong mixed methods studies address the decision of how to integrate the data as well as timing and priority (Creswell and Plano Clark, 2007). Thus, there is a need for guidance for conducting mixed method research and for assessing the rigor of data collection and analysis of both data types. Until now, no information has been provided as to the extent to which marketing research is using mixed methods designs. That is, prior to this article, the role of mixed methods design types in marketing has not been the subject of formal examination. The delineation of the major forms in mixed method designs provides a framework for looking at such design types, which helps to provide more credibility to the eld of marketing by providing examples of research designs that are substantially different than single strand studies. The framework also helps advance a common language for researchers using mixed methods techniques and provides guidance and direction for marketing researchers to design mixed methods studies. Bryman (2006) found when interviewing mixed methods scholars that they had a difcult time identifying exemplary mixed methods research and an absence of best practice templates from which to draw upon when it comes to combining ndings. This study addresses this issue for marketing scholars by identifying current trends and offering examples of research employing different design types. While it is clear that marketing scholars recognize the benet of mixing qualitative and quantitative research, as a discipline we are not demonstrating knowledge of the mixed method literature or procedures. This is demonstrated by the fact that only one article recognized or mentioned knowledge of mixed method procedures or cited mixed method research (Argouslidis and McLean, 2004). Further, having a mixed methods design plan will improve an authors rhetoric as to why specic designs were used to address research problems and force them to consider what can be learned by using both qualitative and quantitative data. It is also evident that researchers may not be maximizing the extent to which they are using mixed method research. For example, several authors mentioned conducting interviews to collect information prior to a quantitative study, but data from the interviews were often under-reported and the method by which the data were analyzed was often not used or dened. On the other hand, there were a few mixed method articles with weak quantitative strands (e.g. simple descriptive statistics). This is not to say that these are weak studies, they merely were not labeled mixed methods in this study. A majority of the studies placed emphasis on the quantitative component (63 percent, n 27). Quantitative research is more apt for answering questions about relationships between specic variables, and questions of who, where, how many and how much. Qualitative research is more apt for answering why and how questions. Understanding experiences and meanings attached to experiences of individuals and groups is the hallmark of good qualitative research. Mixed method research provides strengths that offset weaknesses in both qualitative and quantitative research.

This review nds that a mixed methods design typology can be used to describe the types of mixed method research currently being used by marketing scholars. The ndings highlight patterns in mixed method research designs conducted in marketing. Unlike other disciplines that have been found to use predominantly concurrent designs (Hanson et al., 2005), or an evenly distributed combination of the concurrent and sequential designs (Plano Clark et al., 2008), we found that an overwhelming majority of the marketing studies identied were sequential designs (79 percent, n 34). One reason for these results could be that a second data type came at the behest of reviewers. For example, an author of one of the sequential designs said that reviewers required that a quantitative study be added. Another explanation for this trend could be the inuence of commercial market research with its focus on preliminary focus groups and textbooks that suggest that the time for qualitative research is to develop hypothesis to be quantied by empirical research. In identifying mixed methods studies, a limitation was identied in the collecting and compiling mixed method studies wherein only the qualitative or quantitative results were published in different journals. By only focusing on studies wherein results from both data types are presented, we may have missed studies that employed mixed methods designs, but presented ndings in different journal outlets. For example, Marcella and Davies (2004) and Marcella et al. (2002) employed an exploratory sequential design, publishing ndings in different journals. While these authors specically discussed this issue of multiple publications, other authors using similar strategies without exclamation could not be identied in the search. The subjective nature of the article selection processes is recognized as a limitation; however, this study presents a current snapshot of how mixed methods designs are being used in marketing research. It is also acknowledged here that the nature of the review process may result in the removal of qualitative or quantitative content, due to space limitations and resulting manuscript streamlining. As such, it is suggested that while a larger instance of mixed methods research may be undertaken, such research is not making its way into the knowledge repositories of the discipline. In a recent discussion on mixed methods research (Freshwater, 2007), warned against uncritically accepting mixed methods as an emerging dominant discourse. She criticizes mixed methods research because of its desire for certainty, suggesting that the gaps in mixed methods text is that there should be no gaps (Freshwater, 2007). This is based on the apparent goal of grasping the whole through mixed methods inquiry, with little critical reection in the assumption that the whole is always partial in itself. That is, Freshwater (2007, p. 141) suggests that there is little recognition of the impossibility of deciding between two or more competing interpretations. She suggests that mixed methods research brings different perspectives to bear, but does not allow for competing interpretations to coexist. Further, she suggests that it is important to cross-validate results; however, the rationale for doing so is dependent on the motivations of the research and the desire for such accuracy. Echoing Johnson and Onwuegbuzies (2004, p. 23) sentiments, it is accepted that qualitative, quantitative, and mixed method are all superior under different circumstances and researchers must determine how to use different approaches. If mixed methods is the desired approach, this study provides guidance for researchers in identifying design types appropriate for various rationales or research objectives and models of different design types that have been published in marketing journals.

Mixed methods designs

21

QMRIJ 14,1

In addition, implications for the designing mixed methods studies in marketing include highlighting the need for scholars to specically address issues such as the timing and priority given to each data type (i.e. sequential or concurrent), and the integration (or mixing) of the both data types. Managerial implications For marketing managers, the inclusion of both deductive and inductive logic allows for a more holistic view of a given marketing problem rather than a strictly positivistic or interpretivist slant to the data. From a practical standpoint, many seemingly empirical articles already utilize qualitative methods while conducting scale construction, pre-tests, manipulation checks, and other research tasks where simply asking research subjects their opinions is more diagnostic to researchers than running multiple tests. These results are often left out of the research or under-reported. Conversely, the inclusion of quantitative work to qualitative research often increases its generalizability to a larger population while maintaining the understanding and insight which has already been generated. In this vain, it has been suggested that mixed methods research allows for more trustworthy results than any single research paradigm, and that the methods often offset each other in complimentary ways (Bahl and Milne, 2006).
References Andrade, E.B. and Ho, T.-H. (2009), Gaming emotions in social interactions, Journal of Consumer Research, Vol. 36 No. 4, pp. 539-52. Argouslidis, P.C. and McLean, F. (2004), Service elimination decision-making, European Journal of Marketing, Vol. 38 Nos 11/12, pp. 1355-81. Arnold, M.J. and Reynolds, K.E. (2003), Hedonic shopping motivations, Journal of Retailing, Vol. 79 No. 2, pp. 77-95. Avlonitis, G.J. and Indounas, K.A. (2007), An empirical examination of the pricing policies and their antecedents in the services sector, European Journal of Marketing, Vol. 41 Nos 7/8, pp. 740-64. Bahl, S. and Milne, G.R. (2006), Mixed methods in interpretive research: an application to the study of self concept, in Belk, R.W. (Ed.), Handbook of Qualitative Research Methods in Marketing, Edward Elgar, Northampton, MA, pp. 198-218. Botti, S., Orfali, K. and Iyengar, S.S. (2009), Tragic choices: autonomy and emotional responses to medical decisions, Journal of Consumer Research, Vol. 36 No. 3, pp. 337-52. Brennan, D.R., Turnbull, P.W. and Wilson, D.T. (2003), Dyadic adaptation in business-to-business markets, European Journal of Marketing, Vol. 37 Nos 11/12, pp. 1636-65. Bruner, G.C. and Kumar, A. (2007), Gadget lovers, Journal of the Academy of Marketing Science, Vol. 35 No. 3, pp. 329-39. Bryman, A. (2006), Integrating quantitative and qualitative research: how is it done?, Qualitative Research, Vol. 6 No. 1, pp. 97-113. Coelho, F. and Easingwood, C. (2008), An exploratory study into the drivers of channel change, European Journal of Marketing, Vol. 42 Nos 9/10, pp. 1005-22. Commuri, S. and Gentry, J.W. (2005), Resource allocation in households with women as chief wage earners, Journal of Consumer Research, Vol. 32 No. 2, pp. 185-95.

22

Coulter, R.A., Price, L.L. and Feick, L. (2003), Rethinking the origins of involvement and brand commitment: insights from postsocialist Central Europe, Journal of Consumer Research, Vol. 30 No. 2, pp. 151-69. Coulter, R.A., Price, L.L., Feick, L. and Micu, C. (2005), The evolution of consumer knowledge and sources of information: Hungary in transition, Journal of the Academy of Marketing Science, Vol. 33 No. 4, pp. 604-19. Creswell, J.W. and Plano Clark, V.L. (2007), Designing and Conducting Mixed Methods Research, Sage, Thousand Oaks, CA. Creswell, J.W., Plano Clark, V.L., Gutmann, M.L. and Hanson, W.E. (2003), Advanced mixed methods research designs, in Tashakkori, A. and Teddlie, C.B. (Eds), Handbook of Mixed Methods in Social and Behavioral Research, Sage, Thousand Oaks, CA, pp. 209-40. Dahl, D.W. and Moreau, C.P. (2007), Thinking inside the box: why consumers enjoy constrained creative experiences, Journal of Marketing Research, Vol. 44 No. 3, pp. 357-69. Daly, A., Grove, S.J., Dorsch, M.J. and Fisk, R.P. (2009), The impact of improvisation training on service employees in a European airline: a case study, European Journal of Marketing, Vol. 3/4, pp. 459-72. Dellande, S., Gilly, M.C. and Graham, J.L. (2004), Gaining compliance and losing weight: the role of the service provider in health care services, Journal of Marketing, Vol. 68 No. 3, pp. 78-91. Du, S., Sen, S. and Bhattacharya, C.B. (2008), Exploring the social and business returns of a corporate oral health initiative aimed at disadvantaged Hispanic families, Journal of Consumer Research, Vol. 35 No. 3, pp. 483-94. Freshwater, D. (2007), Reading mixed methods research: contexts for criticism, Journal of Mixed Methods Research, Vol. 1 No. 2, pp. 134-46. Grace, D. and OCass, A. (2003), Child care services, European Journal of Marketing, Vol. 37 Nos 1/2, pp. 107-32. Grayson, K. and Martinec, R. (2004), Consumer perceptions of iconicity and indexicality and their inuence on assessments of authentic market offerings, Journal of Consumer Research, Vol. 31 No. 2, pp. 296-312. Greene, J.C., Caracelli, V.J. and Graham, W.F. (1989), Toward a conceptual framework for mixed-method evaluation designs, Educational Evaluation and Policy Analysis, Vol. 11 No. 3, pp. 255-74. Grifths-Hemans, J. and Grover, R. (2006), Setting the stage for creative new products: investigating the idea fruition process, Journal of the Academy of Marketing Science, Vol. 34 No. 1, pp. 27-39. Hand, C., Riley, F.D.O., Harris, P., Singh, J. and Rettie, R. (2009), Online grocery shopping: the inuence of situational factors, European Journal of Marketing, Vol. 43 Nos 9/10, pp. 1205-19. Hanson, D. and Grimmer, M. (2007), The mix of qualitative and quantitative research in major marketing journals, 1993-2002, European Journal of Marketing, Vol. 41 Nos 1/2, pp. 58-70. Hanson, W.B., Creswell, J.W., Plano Clark, V.L., Petska, K.S. and Creswell, D. (2005), Mixed methods research designs in counseling psychology, Journal of Counseling Psychology, Vol. 52 No. 2, pp. 224-35. Harris, J. and Blair, E.A. (2006), Consumer preference for product bundles: the role of reduced search costs, Journal of the Academy of Marketing Science, Vol. 34 No. 4, pp. 506-13.

Mixed methods designs

23

QMRIJ 14,1

Hewett, K., Money, R.B. and Sharma, S. (2006), National culture and industrial buyer-seller relationships in the United States and Latin America, Journal of the Academy of Marketing Science, Vol. 34 No. 3, pp. 386-402. Horsky, D., Nelson, P. and Posavac, S.S. (2004), Stating preference for the ethereal but choosing the concrete: how the tangibility of attributes affects attribute weighting in value elicitation and choice, Journal of Consumer Psychology, Vol. 14 Nos 1/2, pp. 132-40.

24

Hudson, L.A. and Ozanne, J.L. (1988), Alternative ways of seeking knowledge in consumer research, Journal of Consumer Research, Vol. 14 No. 4, pp. 508-21. Hunt, S.D. (1994), On rethinking marketing: our discipline, our practice, our methods, European Journal of Marketing, Vol. 28 No. 3, pp. 13-25. Jayachandran, S., Hewett, K. and Kaufman, P. (2004), Customer response capability in a sense-and-respond era: the role of customer knowledge process, Journal of the Academy of Marketing Science, Vol. 32 No. 3, pp. 219-33. Johnson, R.B. and Onwuegbuzie, A.J. (2004), Mixed methods research: a research paradigm whose time has come, Educational Researcher, Vol. 33 No. 7, pp. 14-26. Johnson, R.B., Onwuegbuzie, A.J. and Turner, L.A. (2007), Toward a denition of mixed methods research, Journal of Mixed Methods Research, Vol. 1 No. 2, pp. 112-33. Kirmani, A. and Campbell, M.C. (2004), Goal seeker and persuasion sentry: how consumer targets respond to interpersonal marketing persuasion, Journal of Consumer Research, Vol. 31 No. 3, pp. 573-82. Kolbe, R.H. and Burnett, M.S. (1991), Content-analysis research: an examination of applications with directives for improving research reliability and objectivity, The Journal of Consumer Research, Vol. 18 No. 2, pp. 243-50. Lau-Gesk, L. and Meyers-Levy, J. (2009), Emotional persuasion: when the valence versus the resource demands of emotions inuence consumers attitudes, Journal of Consumer Research, Vol. 36 No. 4, pp. 585-99. Luna, D., Ringberg, T. and Peracchio, L.A. (2008), One individual, two identities: frame switching among biculturals, Journal of Consumer Research, Vol. 35 No. 2, pp. 279-93. McMullan, R. and Gilmore, A. (2008), Customer loyalty: an empirical study, European Journal of Marketing, Vol. 42 Nos 9/10, pp. 1084-94. Marcella, R. and Davies, S. (2004), The use of customer language in international marketing communication in the Scottish food and drink industry, European Journal of Marketing, Vol. 38 Nos 11/12, pp. 1382-95. Marcella, R., Davies, S. and Williams, D. (2002), The language strategies of Scottish exporters in the food and drink sector, Journal of Small Business and Enterprise Development, Vol. 9 No. 4, pp. 358-69. Mathwick, C., Wiertz, C. and de Ruyter, K. (2008), Social capital production in a virtual p3 community, Journal of Consumer Research, Vol. 34 No. 6, pp. 832-49. Mekonnen, A., Harris, F. and Laing, A. (2008), Linking products to a cause or afnity group, European Journal of Marketing, Vol. 42 Nos 1/2, pp. 135-53. Morgan, D.L. (2007), Paradigms lost and pragmatism regained: methodological implications of combining qualitative and quantitative methods, Journal of Mixed Methods Research, Vol. 1 No. 1, pp. 48-76. Morse, J.M. (1991), Approaches to qualitative-quantitative methodological triangulation, Nursing Research, Vol. 40 No. 2, pp. 120-3.

Morse, J.M. (2003), Principles of mixed methods and multimethod research design, in Tashakkori, A. and Teddlie, C. (Eds), Handbook of Mixed Methods in Social and Behavioral Research, Sage, Thousand Oaks, CA, pp. 189-208. Paul, M., Hennig-Thurau, T., Gremler, D.D., Gwinner, K.P. and Wiertz, C. (2009), Toward a theory of repeat purchase drivers for consumer services, Journal of the Academy of Marketing Science, Vol. 37 No. 2, pp. 215-37. Plano Clark, V.L., Huddleston-Casas, C.A., Churchill, S.L., Green, D.O.N. and Garrett, A.L. (2008), Mixed methods approaches in family science research, Journal of Family Issues, Vol. 29 No. 11, pp. 1543-66. Powell, H., Mihalas, S., Onwuegbuzie, A.J., Suldo, S. and Daley, C.E. (2008), Mixed methods research in school psychology: a mixed methods investigation of trends in the literature, Psychology in the Schools, Vol. 45 No. 4, pp. 291-309. Russell, C.A., Norman, A.T. and Heckler, S.E. (2004), The consumption of television programming: development and validation of the connectedness scale, Journal of Consumer Research, Vol. 31 No. 1, pp. 150-61. Scott, L.M. and Vargas, P. (2007), Writing with pictures: toward a unifying theory of consumer response to images, Journal of Consumer Research, Vol. 34 No. 3, pp. 341-56. Simoes, C., Dibb, S. and Fisk, R.P. (2005), Managing corporate identity: an internal perspective, Journal of the Academy of Marketing Science, Vol. 33 No. 2, pp. 153-68. Sogn-Grundvag, G. and Ostli, J. (2009), Consumer evaluation of unbranded and unlabelled food products: the case of bacalhau, European Journal of Marketing, Vol. 43 Nos 1/2, pp. 213-28. Steenkamp, J.-B.E.M. and Geyskens, I. (2006), How country characteristics affect the perceived value of web sites, Journal of Marketing, Vol. 70 No. 3, pp. 136-50. Tashakkori, A. and Creswell, J.W. (2007), Editorial: the new era of mixed methods, Journal of Mixed Methods Research, Vol. 1 No. 1, pp. 3-7. Tashakkori, A. and Teddlie, C. (2003), Handbook of Mixed Methods in Social & Behavioral Research, Sage, Thousand Oaks, CA. Theoharakis, V. and Hirst, A. (2002), Perceptual differences of marketing journals: a worldwide perspective, Marketing Letters, Vol. 13 No. 4, pp. 389-402. Thomson, M. (2006), Human brands: investigating antecedents to consumers strong attachments to celebrities, Journal of Marketing, Vol. 70 No. 3, pp. 104-19. Truscott, D.M., Swars, S., Smith, S., Thornton-Reid, F., Zhao, Y., Dooley, C., Williams, B., Hart, L. and Matthews, M. (2010), A cross-disciplinary examination of the prevalence of mixed methods in educational research: 1995-2005, International Journal of Social Research Methodology, Vol. 13 No. 4, pp. 317-28. Ulaga, W. and Eggert, A. (2006), Value-based differentiation in business relationships: gaining and sustaining key supplier status, Journal of Marketing, Vol. 70 No. 1, pp. 119-36. Venable, B.T., Rose, G.M., Bush, V.D. and Gilbert, F.W. (2005), The role of brand personality in charitable giving: an assessment and validation, Journal of the Academy of Marketing Science, Vol. 33 No. 3, pp. 295-312. Voorhees, C.M., Brady, M.K. and Horowitz, D.M. (2006), A voice from the silent masses: an exploratory and comparative analysis of noncomplainers, Journal of the Academy of Marketing Science, Vol. 34 No. 4, pp. 514-27. Walsh, G. and Beatty, S. (2007), Customer-based corporate reputation of a service rm: scale development and validation, Journal of the Academy of Marketing Science, Vol. 35 No. 1, pp. 127-43.

Mixed methods designs

25

QMRIJ 14,1

West, D. and Prendergast, G.P. (2009), Advertising and promotions budgeting and the role of risk, European Journal of Marketing, Vol. 43 Nos 11/12, pp. 1457-76. Wolnbarger, M. and Gilly, M.C. (2003), eTailQ: dimensionalizing, measuring and predicting e-Tail quality, Journal of Retailing, Vol. 79 No. 3, pp. 183-98. Woodruff, R.B. (2003), Serving the marketing discipline through journal reviews, Journal of the Academy of Marketing Science, Vol. 31 No. 3, pp. 327-30. About the authors Robert L. Harrison has a PhD from the University of Nebraska-Lincoln and is an Assistant Professor at Western Michigan University. He is a member of the AMA and ACR. His current research interests include family and experiential consumer behaviour. Robert L. Harrison is the corresponding author and can be contacted at robert.harrison@wmich.edu Timothy M. Reilly is a Doctoral Student at the University of Nebraska-Lincoln. His research interests include research methodologies and consumer responses to marketing communication crises.

26

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints