The Q method as research and intervention tool in organizations: a systematic literature review

Within organizational studies, behavior is one of the key topics (Duriau et al., 2007). Organizational behavior (OB) refers to the systematic study and application of knowledge of employees’ individual behavior, the behavior of groups of employees, the overall behavior of employees and the interaction between different levels within organizations (Bauer and Erdogan, 2009; Heath and Sitkin, 2001). Within this domain, theories and empirical studies have developed into a highly relevant, valid and useful research area (Miner, 2003). The field of OB is currently thriving, as is illustrated by the increasing numbers of topics, methodologies and scholars (Oldham, 2024). Nevertheless, the research area’s complexity implies enduring methodological challenges.

Individuals’ perceptions and opinions emerge through mechanisms that can be studied along quantitative stages, such as statistical factor analyses and multi-level analyses and along qualitative stages, such as sense-making (Ramlo, 2016). Ideally, these approaches can be combined and integrated by iteratively testing each approach against the other (Spector and Meier, 2014). A mixed method provides a better understanding of complex problems and phenomena than a qualitative or quantitative approach alone (Molina-Azorin and Cameron, 2010). A method that takes conceptual and theoretical complexity into account and integrates both a qualitative and a quantitative approach can be found in Q methodology (Stainton Rogers, 1995). Q methodology studies the individual’s point of view, which can take many different shapes and forms (Amin, 2000; Karim, 2001). Although this approach seems highly valuable for OB, it is, compared to other techniques, not often applied in OB studies. Even the use of mixed methods in general is scarce in OB research. Currently, in OB, quantitative methods dominate (Molina-Azorin and Cameron, 2010).

The Q method is gaining popularity among scholars in general (Dieteren et al., 2023). Over the past years, several systematic literature reviews of the application of the Q method have been conducted, varying in time scope (from 1 year to all years), sample size (from 14 articles to 613 articles) and level of comprehensiveness of the analysis, some focused on methodological choices (Dieteren et al., 2023; Dziopa and Ahern, 2011), others focused on specific research fields, namely educational research (Lundberg et al., 2020) and healthcare research (Churruca et al., 2021). For OB, a review of the application of the Q method was still missing. This systematic literature review of the Q method within the field of OB contributes to the literature on OB by outlining opportunities of this method for OB in particular and by defining lessons learned, benefiting the design and prosecution of future OB studies that will apply this method.

OB research aims to describe and understand behaviors in organizations, the factors that determine these behaviors and the outcomes of these behaviors. OB studies focus on one or more of three different levels of analysis – the individual, the group or team and the organization – and may focus on one or more OB topics (Buchanan and Huczynski, 2019; Konopaske et al., 2017; Sinding et al., 2014). The variety and multiplicity of topics and levels are conducive to the complexity of the OB domain. To fully understand behavior in organizations, “how” questions as well as “to what extent” questions must be addressed. In other words, both qualitative and quantitative approaches are needed to study OB issues. While qualitative research generally aims at theory development (Grodal et al., 2020), quantitative analysis involves probability testing of explanations (King et al., 2021). Integrating both approaches provides empirical rigor and cutting-edge insights (Molina-Azorin et al., 2017). However, OB research is dominated by quantitative research methods (Molina Azorin and Cameron, 2010; Molina-Azorin et al., 2017). Some exceptions to the qualitative versus quantitative split in OB entangle mixed-methods efforts like the combination of literature review, interviews and surveys (e.g. Arnold et al., 2000; Kooij et al., 2014). These exceptions prove to be crucial in theory development and advancement. This type of ‘triangulation’ has been advocated over time but is rarely implemented (Yauch and Steudel, 2003). Within OB research, a more balanced and integrated application of quantitative and qualitative methods is needed for several reasons.

First, at the origin and advancement of new topics or concepts, different interpretations and the richness thereof are often narrowed down to scales that fail to capture the full scope of these concepts. The step toward measuring concepts and empirical work is often taken quickly in light of the need to be able to measure the concept, thereby risking undervaluing the full concept. The importance of taking the multitude of interpretations into account in quantitative studies with complex concepts, such as “cognitive ability” and “work motivation” has been suggested in previous research (e.g. Berry et al., 2011; Van den Broeck et al., 2021). However, the development of scales, scale construction, operationalizations and studies of dimensions of specific concepts that have a strong qualitative character, often adopt a quantitative design, which may be less appropriate (e.g. Morgeson and Humphrey, 2006). The issue of content validity is often limitedly addressed, not reported or completely ignored. The original item generation is rarely elaborated upon (e.g. in leadership research, Eva et al., 2019) and the focus mainly lies on incremental validity and predictive validity of scales. This results in neglecting the fine-grained nature of content or interpretations of OB concepts (e.g. Joshi et al., 2011).

Second, empirical studies often test theoretical models assuming their interpretation, which fails to critically and fundamentally assess theoretical models and their applicability. The frame of reference driven by theory delineates the area within which quantitative studies take place and thus limits this scope at the same time. Hypotheses testing and the search for confirmation of these hypotheses often remain the main or even sole focus. Moreover, a widespread desire for advanced methodological designs and longitudinal data leads to neglecting a critical assessment of the selected theoretical framework. Researchers acknowledge the limitations of inferential conclusions within quantitative research (King et al., 2021). In qualitative research, the grounded theory approach gains popularity in developing new theories and fundamentally addressing existing theories. Overall, empirical work underlines the added value of an integrated approach of triangulation, especially in light of theory development (e.g. Judge et al., 2002; Parris and Welty Peachy, 2013). Both in quantitative and qualitative research there is an increased awareness of the risk of neglecting more critical and fundamental assessments of OB theories and their applicability in empirical work.

Third, there is a tendency for quantification which fails to incorporate qualitative elements, findings and conclusions. This tendency is clearly demonstrated by meta-analyses in OB: these often choose not to include qualitative studies (e.g. Kristof-Brown et al., 2005; Van Iddekinge et al., 2018). Furthermore, quantitative meta-analyses usually consist of coding as a method for simplifying and arranging complex data and the associated studies. This process is bound to the chosen perspective and the labels implicated by that perspective. For example, tenure may refer to job or functional tenure or organizational tenure but is usually subsumed into one overall category labeled “tenure.” In addition, the process in meta-analyses is bound to methodological rules implied by the quantitative method applied in the original studies. Taken together, these aspects of quantitative meta-analyses limit the potential for extensive theory building (e.g. Slemp et al., 2020) and risk making inferior generalizations without a qualitative element.

In order to avoid this jumping to interpretations, conclusions and generalizations, OB research would benefit from a combined approach. By applying mixed methods, multiple explanations can occur, causing richer substantiation, increased validity, better understanding of complex phenomena and stimulation of theory building and testing. In sum, an integration of qualitative and quantitative approaches in the field of OB seems imperative, for three reasons. First, OB measurements and operationalizations need reconsideration to be optimized and to better meet the complexity and issues of OB at hand. Second, theory development in OB needs to take different potential explanations into account. And third, since both qualitative and quantitative designs face inevitable limitations, the advancement of the OB research field requires an integration of both approaches. The Q method is an inherently mixed-methods approach that has excellent potential for a wide range of OB studies, especially considering the practical relevance of the OB research field.

The Q method refers to “a distinctive set of psychometric and operational principles that, conjoined with statistical applications of correlational and factor-analytic techniques, provides researchers with a systematic and rigorously quantitative procedure for examining the subjective components of human behavior” (McKeown and Thomas, 2013, p. ix). The Q method can be applied to study how individuals think and feel about a certain topic. The viewpoints of multiple individuals provide insight into the extent to which shared and different perceptions exist. This methodology studies the concourse on topics through subjective perspectives: how and why people think the way they do (Bartlett and DeWeese, 2015). The results of a Q study are based on correlations between the viewpoints of persons or even within one person (Van Exel and De Graaf, 2005; Woods, 2011; Silvius et al., 2017). For example, results may show a particular combination of topic-related themes that are considered important by a group of participants (Watts and Stenner, 2005). Q methodology is widely applied within numerous disciplines such as nursing, environmental sciences, public administration and education (Van der Steen et al., 2018; Jeffares and Skelcher, 2011; Twijnstra and De Graaf, 2013; Sharifi et al., 2014), but to a lesser extent in other domains.

The Q method involves several successive steps. Churruca et al. (2021) summarize these in seven stages: (1) identifying the topic, (2) sampling the concourse and developing the Q-set, (3) performing a pilot, (4) selecting participants, (5) Q-sorting, (6) performing quantitative analyses and (7) qualitatively interpreting factors. In the first stage, the concourse is demarcated (Brown, 1993). The concourse represents the breadth or range of all potentially related indicators, statements, pronouns and views with regard to the focal topic (Amin, 2000). In the second stage, the Q-set, which is the selection of inventoried statements or words, is constructed (Stainton Rogers, 1995). In the third stage, piloting is performed, for example through interviews or focus group sessions (e.g. Chinnis et al., 2001). The fourth stage identifies the group of participants who will perform the Q-sorting, which is called the P-set. The fifth stage represents the Q-sorting by the participants according to specific instructions provided by the researchers. This involves rating the Q-set items, for example, by the level of agreement, in a fixed or open number of categories (McKeown and Thomas, 2013; McKeown et al., 1999; Reber et al., 2000). During the sixth stage, quantitative analyses like factor analyses and multidimensional scaling are conducted (Watts and Stenner, 2005). Finally, during the seventh stage, interpretational analyses take place, comparing results yielded by the previous stage (Stainton Rogers, 1995; Watts and Stenner, 2005). The final result of the analyses is formed by the chosen clusters, categories, facets or dimensions and how the Q-set is situated in light of these.

Although these steps seem straightforward, different applications and operationalizations of Q methodology exist and are used (Dieteren et al., 2023; Donner, 2001; Karim, 2001; Stainton Rogers, 1995). Applications of the Q method may skip one or more steps, like piloting, or may go back and forward between different steps, for example, between Stages 1 and 2 or between Stages 6 and 7. Nevertheless, overall, the Q method steps can be divided into sorting activities and analyzing activities (Dziopa and Ahern, 2011). During both kinds of activities, interpretations are key and develop into a narrative (Stainton Rogers, 1995; Stenner and Stainton Rogers, 2004). The methodological foundation of the Q method comprises an alternation of interpretational and statistical steps that can be consecutive and alternated (cf. Stainton Rogers, 1995) and are strongly intertwined.

In order to address the objectives of this paper, a systematic review was conducted on a systematically collected sample of scholarly sources reporting on studies primarily addressing organizational behavior topics and employing Q methodology. The review study was designed following the steps and principles for systematic reviews in the field of management and organization studies as described by Denyer and Tranfield (2009). Inclusion criteria were set as follows. Most importantly, sources to be included report on a study that primarily addresses organizational behavior topics and that applies the Q method. Moreover, sources to be included in the review sample should be peer-reviewed journal articles, be written in English and report on an empirical study. We conducted many searches in various databases before deciding on the final research strategy, a transparent strategy with few and clear steps and resulting in as many sources meeting the inclusion criteria as possible. Records of sources that were found in the preliminary search stages but did not flow from this final systematic search strategy were saved to be added to the final sample.

We searched the Web of Science core collection on sources published until 2020 that included one or more of the following search terms in title, abstract or keywords: q method*, q-method*, q sort* and q-sort*. We refined the list of 69,612 search results by selecting articles only (leading to 66,848 records) and by selecting the following WoS categories: management, business, social psychology, applied psychology and social sciences interdisciplinary. The remaining list of 406 records was further refined by only including the WoS citation topics meso categories management (6.3) and social psychology (6.73). This resulted in 53 and 58 records respectively, together with 109 WoS records. We screened the abstracts of these 109 records on the inclusion criteria. Some studies did not apply Q method but used Tobin’s Q. Some studies did not address an OB research question but mainly focused on other research areas such as marketing, supply chain management, higher education or environmental studies. In case of doubt, we retrieved and scanned the full article and discussed it within our group of four authors. Among the 109 WoS citations, 20 publications were considered to meet the inclusion criteria. The review sample was further enlarged by adding sources identified in other ways, such as during preliminary searches or by screening references lists of articles in the sample. Most of the additional sources, however, were found through Google scholar, using the search terms q method and q sort in combination with organization and management. The final review sample included 47 sources. The search strategy is visualized in Figure 1.

From the articles in the sample, we extracted (1) country of study, (2) topic, (3) aim of study and research question, (4) reasons for using Q, (5) determination of concourse, (6) Q-sampling strategy, (7) P-sampling strategy, (8) Q-sorting strategy, (9) Q-related strengths (if mentioned) and (10) Q-related limitations (if mentioned). The papers were categorized according to the OB topic.

As is common when reporting systematic literature reviews, we first present a description of the sample based on general study features (Tranfield et al., 2003). The oldest article in our sample dates back to 1969. As Figure 2 shows, there is a clear upward trend over time in a number of publications that meet the selection criteria: over half of the studies were published during the last decade included in the sample, from 2010 until 2020. Relatively many studies in our sample are based on data sets collected in the USA, followed by Europe (especially Belgium, the Netherlands and UK).

In Table 1, we have listed methodological characteristics of the studies included in our sample. A relatively large part, almost 40%, of the studies used existing Q sets, or existing Q set items. The remainder of the studies have developed Q sets from a concourse, based on either a (non-)systematic literature review, focus groups or interviews, or other input. One study applied photographs instead of verbalized perspectives. Both the number of items in the Q set and the number of respondents in the P set vary widely across the sample. The smallest Q set holds seven items and the largest Q set holds 222 items. The smallest P set counts seven respondents and the largest P set counts 1,299 respondents. In many cases, the P set counts around 30 or 40 respondents. In several studies, the P set is not a number of respondents but a range of descriptions that are used as perspectives. In our sample, sortings are predominantly conducted in forced (normal) distributions, often with a number of nine categories. Ratings are in accordance with levels of, for example, distinctiveness, agreement, representativeness, similarity, and other criteria. The majority of the studies describe paper and pencil routines. Only six studies mention an electronic mode of collecting the data. Other articles do not specify the way sortings were administered.

With regard to methodological choices in applying the Q method, in the review sample, in general, studies did not seem to build on previous OB studies: how the method was applied does not show a general development and refinement over time. Methodological choices, if described at all, are mostly justified referring to key publications on Q from the past. Exceptions in this respect are studies using pre-established Q samples; these studies are built further on previous studies using the same or a similar Q sample. However, these developments resulted in the advancement of specific OB measures and not in the advancement of the application of the Q method in OB.

Table 1 also indicates for each study in the sample the methodological context within which the Q method was applied. Twenty studies have integrated the Q method as part of a theory-testing study design. In many of these studies, the Q method was harnessed to measure certain variables that were part of the hypotheses to be tested. A handful of studies utilized the Q method for the sole purpose of either instrument development (Peterson et al., 1999; Ryan and Schmit, 1996; Zand and Sorensen, 1975) or instrument testing (Ceschi et al., 2019). The remaining 23 studies, which is slightly less than half of the studies in the sample, employed the Q method in studies that were designed for theoretical or conceptual development.

The sample of studies that we analyzed involves studies at all three OB levels – at the individual level, the group or team level and the organizational level. To enable a coherent description of the topics and aims present in the studies included in the sample, we have grouped the studies together into several topic categories (see Figure 3). At the individual level, we have grouped together studies on Human resource management (HRM)-related topics. At the group/team level, we have grouped together studies on leadership, studies on group decision-making and studies on collaboration. At the organizational level, finally, we have grouped together studies on culture and studies on organizational change and development.

The studies within the topic category of HRM applied the Q method in diverse ways and with regard to a diversity of HRM-themes. To start with, two sources used the Q method in order to assess person-environment congruence in relation to employee selection (Barrett, 1995; McCulloch and Turban, 2007). More specifically, Barrett (1995) described and demonstrated the Performance Priority Survey, a Q-sort procedure in which supervisors and applicants rate the relative importance of work behaviors. McCulloch and Turban (2007) tested several hypotheses regarding the impact of Person-Organization (PO) fit, which was measured through the Q sort instrument CultureFit.

Also, Denelsky and McKee (1969) used the Q method for assessing congruence, however, not congruence between person and organizational environment, but between predicted and actual performance of employees working overseas. Predicted performance was based on judges’ Q sortings of preemployment psychological assessment reports and actual performance was based on judges’ Q sortings of descriptions of effectiveness in employees’ fitness reports.

Several studies within this topic category reported on Q studies that portray perspectives on HRM-related topics among professionals or scholars. Chinnis et al. (2001) described viewpoints on employee needs based on a Q method study among 41 nurses and other non-physician medical staff at an emergency department in the USA. Lingard et al. (2015) outlined three perception clusters regarding occupational health and safety based on a Q study among 60 construction professionals in Australia. An interesting detail of this study is that the Q set consisted of photographs. Bachkirova et al. (2015) revealed one dominant shared viewpoint on coaching based on a Q study among 41 professional coaches located on various continents. Baek and Kim (2017) identified perspectives on the state and future of critical human resource development (CHRD) based on a Q study among 13 CHRD scholars in the USA and UK Paul et al. (2017) studied perceptions regarding the relative importance of variables influencing research productivity among 200 agricultural scientists in India.

Several other studies within this group applied the Q method in relation to the construction or validation of a conceptual model or the validation of a construct. Büssing et al. (1999) conducted a free Q sort study in combination with other methods among 46 German nurses in order to validate a work satisfaction model. Craik et al. (2002) integrated a Q instrument in a study of the construct validity of managerial performance ratings. These authors applied the California Q set (see Block, 1978), consisting of 100 phrases appropriate for composing a psychological portrait of a person. Derous et al. (2003) tested and adapted the Social Process Model of selection based on multidimensional scaling of Q sorts by 30 experts on selection. Dries et al. (2008) combined interviews, a Q study among 30 subject matter experts, and multidimensional scaling for the construction of a model of career success. Peeters et al. (2019) tested and found support for a conceptual framework on employability capital based on their Q study among 29 academic experts.

One final study within this group used the Q method in combination with other methods for a scale development purpose. Ceschi et al. (2019) conducted an interview study to develop items regarding non-technical skills in aviation and tested the logical principles, intelligibility and clarity of the items in a Q study.

Roughly half of the studies within the topic category of leadership applied the Q method in a straightforward way, whereas the other half applied the Q method as part of a research design that also includes other methods. Several articles reported on studies that portray perspectives on leadership roles, mostly among managers themselves, in various countries. Woods (2011) studied perspectives on the role of the school business manager in leadership in English schools. Mitiku et al. (2017) studied perspectives on leadership roles among Ethiopian civil servant managers. Sales et al. (2020) studied perspectives on leadership roles and empowerment among women leaders from diverse ethnicities. Also, Militello and Benham (2010) brought a full paper based on a Q study, however, the subject of their study was the outcomes of a leadership development program in six American cities. These authors emphasized the threefold value of their Q study: the Q method as a data collection tool for scholarly purposes, the Q method as a tool for participants in research and the Q method as a tool for evaluators. One study within this topic category applied the Q method as a template matching technique, using the Organizational Culture Profile (see Organizational culture articles), in order to compare the value system of a spiritual leader with other value systems (Huang and Shih, 2011).

Studies in this topic category that employed the Q method as one among several data collection techniques substantiated their use of the Q method from various and rather diverging viewpoints. Some studies simply applied the Q method as one study among a range of studies with equal status within a multi-method design (Crossan et al., 2017; Dries and Pepermans, 2012; Van Dun et al., 2017). Dries and Pepermans (2012) applied multi-dimensional scaling of sorts in structured piles in order to develop a model of leadership potential. Crossan et al. (2017) conducted a Q study to examine how business leaders define character, as part of their study aimed at bringing the construct of character into management research. Van Dun et al. (2017) used the Q method to study the values of lean managers. Based on the results of their full study, they produced propositions regarding effective lean leadership. In two other studies, the place of the Q study within the research design could be described as supporting research goals that are primarily addressed by a different method (Fu et al., 2010; Liden et al., 2016). Fu et al. (2010) applied the Q method as a means for validating a measure of CEO’s personal values in China, which was used in their survey study. Liden et al. (2016) applied the Q method for measuring CEO values in a Chinese context, arguing that this should counter the social desirability problem inherent in using value scales among Chinese people.

The studies within the topic category of group decision-making applied the Political Group Dynamics Q sort (PGDQS) (Peterson, 1997) or the Group Dynamics Q sort (GDQ) (Ferguson et al., 2019; Peterson et al., 1998, 1999, 2003; Wong et al., 2011). These instruments allow “for a systematic approach to studying elite level groups that are normally not open to quantitative study” (Peterson et al., 1998, p. 279), for example, by using publicly available data sources. Peterson (1997) used the PGDQS to study the effects of leader directiveness in group decision-making. In a subsequent study, in collaboration with others, he adapted the PGDQS into the GDQ (Peterson et al., 1998). This instrument forms a common data language to describe processes in and across groups. It involves a set of 100 items within eight process indicator scales, e.g. Leader weakness-strength and Optimism-Pessimism. The scales represent broad group process dynamics. From the set of 100 items, other scales can be developed to test specific constructs or theories.

The GDQ instrument is described in detail in a methodological article that includes empirical data from a demonstration case (Peterson et al., 1999). The remainder of the studies that applied the GDQ are empirical papers with the theme of top management teams as a central topic. In the majority of these, hypotheses were tested and GDQ was applied in order to assess one or more of the variables in the hypotheses (Ferguson et al., 2019; Peterson et al., 2003; Wong et al., 2011). Hypotheses in the studies regarded top management dynamics in relation to CEO personality (Peterson et al., 2003), in relation to corporate social performance (Wong et al., 2011), and in relation to boundary spanning and financial performance (Ferguson et al., 2019). Data about the group dynamics in the top management teams in the selected companies were retrieved from archival sources such as memoirs or books (Peterson et al., 2003) or from the business press and industry journals (Ferguson et al., 2019; Wong et al., 2011). One study applied template matching by creating a template by Q sorting group dynamics described by groupthink and vigilant decision-making theory (Peterson et al., 1998). This application of the Q method allows for a comparison between theoretical ideal types and actual group dynamics in empirical cases through Pearson’s correlations. In this way, in Peterson et al. (1998), the value of specific models as explanations of failure and success were explored.

The studies within the topic category of collaboration portrayed perspectives on a collaboration-related theme based on a Q study among practitioners. Sharifi et al. (2014) described perspectives on knowledge transfer management in higher education, based on Q sorts by 31 senior practitioners of eight UK universities. Suprapto et al. (2015) outlined perspectives on the essence of collaborative relationships in engineering and construction projects, based on Q sorts by 30 project practitioners of nineteen Dutch firms. Moalusi and Coetzee (2018) reported ways of perceiving trust in business partnerships, based on Q sorts by 25 managers of emerging and mainstream African companies. Lai and Smith (2019) investigated the topic of effective coaching alliances by combining critical incident interviews with two levels of Q sorting among coaches, coachees and organizational stakeholders.

With two exceptions, all studies within the topic category of organizational culture applied the Organizational Culture Profile (OCP), or an adapted version of this Q instrument. The OCP was developed and validated by O’Reilly et al. (1991) and is aimed at assessing the person-organization fit. The set of value statements contains 54 items to be sorted into nine categories. Organizational culture profiles can be obtained by key informants in a particular organization through Q sortings. These organizational profiles are then applied for template matching with the value profiles of (groups of) employees. The original OCP was unstructured, however, Sheridan (1992) discerned seven dimensions in 27 of the original 54 items. Also, Howard (1998) adapted the original OCP, by selecting 23 items for seven dimensions and adding 25 items for two extra dimensions.

Several studies within this topic category reported on hypotheses testing studies (Chatman and Jehn, 1994; Cooper-Thomas et al., 2004; O’Reilly et al., 1991; Sheridan, 1992; Vandenberghe, 1999). Person-organization fit was studied in relation to, for example, job satisfaction, organizational commitment (O’Reilly et al., 1991; Vandenberghe, 1999) and job performance (Lee and Yu, 2004; Sheridan, 1992). Cooper-Thomas et al. (2004) tested, among other hypotheses, the impact of socially-oriented socialization tactics on changes in “newcomer perceived” and actual, value-based person-organization fit.

Another recurring theme among studies in this topic category is the testing of the OCP instrument in different national cultures and different industries. Chatman and Jehn (1994) found that stable organizational culture profiles existed and varied more across industries than within them. In a study by Vandenberghe (1999), the OCP was tested in a Belgian sample, replicating the results of the study conducted by O’Reilly et al. (1991). Lee and Yu (2004) showed the robustness of the OCP by applying it in a study of Singaporean companies. Howard (1998) applied the OCP to test a theory, the competing values perspective on organizational culture, in firms across different industries.

Two sources on organizational culture did not apply the OCP. In a study conducted by Nemcsicsné Zsóka (2007), a Q set on environmental awareness was developed in order to examine the role of environmental values in corporate pro-environmental behavior in a Hungarian company in the paper industry. Silvius et al. (2017) used the Q method in order to investigate perspectives on sustainability values among twelve project managers from various industries in the Netherlands.

The four studies in the topic category on organizational change and development applied the Q method in various sorts of ways. Zand and Sorensen (1975) used the Q method as a step towards the composition of a questionnaire studying forces affecting change. Here, Q sorting by seven behavioral scientists was conducted in order to classify phases of change content units derived from interviews and select items for the questionnaire. Ryan and Schmit (1996) applied the Q method in their Organizational Fit Instrument (OFI). This instrument shows some similarity to the OCP (see Organizational culture articles) since it also assesses the fit between a person and the work environment. However, the OFI focuses on organizational-climate characteristics instead of organizational culture and values. In Klaus et al. (2010), the Q method formed the main research method. This study used a Q set derived from a focus group and interviews in order to investigate user perspectives on enterprise system implementations. In a study by Pfeiffer et al. (2019) on participation in the implementation of Industry 4.0, the Q method formed part of a multi-method design, together with interviews and a survey. The Q sortings were subjected to a Principal Component Analysis, reducing the shared patterns to a one-component model.

A variety of limitations and strengths of the Q method are described in the sample, often as part of discussion sections of the articles. We list the most prominently brought forward, general limitations and strengths of Q for OB that emanate from the analysis of the sample. A first limitation of the Q method regularly mentioned by authors is the limited generalizability of results (e.g. Sales et al., 2020; Vandenberghe, 1999). Results may not be transferable to broad populations or to other organizational contexts or nations. A second limitation indicated by the authors is a potential bias in responses, for example, due to social desirability. This limitation may apply to face-to-face Q data collection, during which respondents face a researcher in front of them, more than to online Q data collection. Another way in which respondents’ answers may be biased is that respondents “may simply respond more positively to elements with which they are familiar” (Crossan et al., 2017, p. 1010). A third limitation, in the words of Peterson et al. (1999), is that the Q method neither has “as much descriptive richness as a case study, nor permits the clean causal claims that experiments do” (p. 107).

The major strengths of Q for OB pointed out by the studies in the sample revolve around connecting, in theoretical and practical ways. Q connects theory and practice, or, put differently, Q connects the academic domain with the professional domain. More specifically, first, Q as a method is extraordinary in that it allows for engaging stakeholders in theory-building processes. By studying perspectives of organizational stakeholder groups, it shapes “theory in ways that are in concert with practice” (Crossan et al., 2017, p. 1010). Second, by collecting and clarifying varying definitions of a certain concept that are common within different disciplines or groups, the application of Q is considered to be a means to connecting different disciplines or groups (e.g. Lee and Yu, 2004; Peenstra and Silvius, 2018). Third, a Q study can serve academic and professional aims at the same time. In addition to the collection of data for research purposes, conducting a Q study can form an organizational “tool for mobilizing discussions, engaging in reflection, and planning new activities” (Militello and Benham, 2010, p. 630).

We will discuss three patterns that surfaced from investigating the OB studies included in the sample in our study. First, we found that many studies in our sample only partly report the research process, resonating with results from other systematic literature reviews of the application of the Q method (Churruca et al., 2021; Dieteren et al., 2023). Which Q set was used and how many participants performed sorting was often reported, however, how the items in the Q set were developed from the concourse and how these were sorted into ratings was often not reported. This restricts the reliability and replicability of these studies. For example, sorting cards with a focus on “importance” versus sorting cards with a focus on “similarity” bares different conclusions. Moreover, if the concourse remains unclear, the extent to which generalizing is allowed remains vague. The concourse is important for a good comprehension of the topic domain to which the conclusions of a study apply. In addition, based on what is included in the concourse, and what is not included in the concourse, it is possible to identify unexplored areas and opportunities for future research. In line with these observations, when it comes to applying Q methodology, we deem it paramount that studies carefully delineate the concourse and describe the Q method steps applied in detail (see also Churruca et al., 2021; Dieteren et al., 2023). Therefore, we recommend that OB scholars who apply the Q method in their research take this to heart. We refer to Churruca et al. (2021) and Dieteren et al. (2023) for checklists of Q study details that may be considered relevant reporting.

Second, surprisingly, the range of OB topics that are represented in the sample of our review is rather restricted. Many OB topics are hardly represented or even entirely absent in the sample. For example, topics such as teamwork, meetings, work engagement and work stress are lacking in the sample, whereas they are suitable for the Q method and research on these topics could benefit from a Q approach. Therefore, we recommend that OB scholars more often consider the application of Q methodology in their research, especially for research into attitudes, emotions and behaviors that are currently not yet studied with the application of Q methodology. This may be facilitated by recent publications that are starting to bring more clarity and guidance on methodological choices in applying the Q method (Churruca et al., 2021; Dieteren et al., 2023), although, we admit, there is still a way to go. Diversity in guiding instructions and sometimes lack of evidence for advised methodological choices in Q study research may complicate researchers’ endeavors and demand further evolutions in the application of the Q method (Dieteren et al., 2023). Nevertheless, we agree with Watts and Stenner that Q methodology has “almost boundless potential” (2005, p. 50). In our view, OB is a research field that could particularly benefit from this method.

Third, remarkably, in more than half of the studies in our sample, the Q method was applied for other purposes than for revealing subjective positions. The systematic literature reviews of the application of Q within healthcare and environmental research did not reveal such patterns. Apparently, this is quite typical for the application of the Q method within OB. This pattern seems to resonate with the, in the introduction of this paper indicated, tendency within OB research that quantitative methods dominate. We may describe this type of use of the Q method as the application of Q technique without applying Q methodology (Dziopa and Ahern, 2011). For example, pre-established Q samples that measure theorized phenomena might be considered “essentially quantitative analysis […], which ignores the main premise of subjectivity of Q methodology” (Dziopa and Ahern, 2011, p. 42). Only less than half of the OB studies within our sample applied the Q method for the main goal of revealing subjective positions. This is surprising since research on complex organizational behavioral matters might gain much from deeper insights into perspectives of various stakeholders involved in organizational matters. Therefore, although we do not generally object to variations in the way the Q method is harnessed, we consider the Q method promising for OB research, especially with regard to the study of subjectivity.

This study contributes to the field of OB in the following ways. First, the results of our study may help OB researchers to make sound decisions on applying the Q method or not, and, if yes, for which type of research questions specifically. Thereby, this study may be a first step toward a more coherent application of the Q method within the OB field and building a common body of knowledge. This will enhance the range of options OB researchers may choose from for conducting relevant research and can enrich the development and refinement of OB concepts and theories, thereby better meeting the challenges of understanding complex OB concepts.

Second, new users of the Q method within the OB field may be referred to this study as a starting point. It provides relevant considerations and literature references to sources that can guide scholars through methodological choices involved in the application of the Q method and that provide scholars with standards for adequate reporting of Q studies. Following such guidelines is imperative to counter the current confusion about conditions for sound Q study designs as well as the current trend of incomplete and inadequate reporting of Q studies, and to move forward to more rigorous Q research in OB.

Third, we discussed developments regarding the application of the Q method within the field of OB in light of developments regarding the application of the Q method in general. OB is an interdisciplinary research field and is intertwined with many adjacent fields. Therefore, evidently, advancements in the OB field may be inspired, boosted and enriched by learning from methodological developments in other research fields and vice versa. This study provides insights into how the application of the Q method within OB over the past decades compares to how the method was applied and evolved more in general, thereby adding to a balanced and nuanced perspective on the application of the Q method in OB.

As every study comes with limitations, so does this systematic literature review. First, the main challenge we encountered during the execution of this study was how to systematically collect the review sample. From the onset, we explored a wide variety of search designs, covering various search terms and various databases. Paradoxically, the more systematic the search design, the less useful the list of records generated: we retrieved thousands of records that appeared to include only a very, very limited number of studies on the Q method. Furthermore, we noted that OB studies that were already within our sight due to previous search activities, were, surprisingly, often missing in these search outcomes. These search challenges and the way we dealt with them, show similarities with reports of other recent systematic literature reviews of Q (Dieteren et al., 2023; Lundberg et al., 2020). Similar to Dieteren et al. (2023), we chose to focus on one database (since adding other databases would substantially add to the number of records without adding much to the number of records that met the inclusion criteria), and, similar to Lundberg et al. (2020), our final sample was built for a substantial part from studies that were identified via alternative search methods. Due to this search strategy, we may have missed some studies that do meet the inclusion criteria set for our review sample. Nevertheless, considering our extensive efforts to capture all OB studies until 2020 that use the Q method, we think our sample is still sufficiently solid.

Second, during the sample collection and analysis, we did not apply a quality appraisal criterion, as we intended to gain an overview of Q studies in OB as completely as possible. All studies in the sample were considered on an equal basis within the analysis, neglecting potential differences with regard to paper quality. If we had applied quality criteria, the results of this study could have been different. A third limitation of this study is that we may have delved deeper into the opportunities of the Q method for specific OB themes and with regard to dominant OB questions still to be answered. We call for future studies to further explore and address this research avenue.

Q studies conducted in organizations can yield important managerial insights (Molaei et al., 2019). In a number of studies in our sample, Q was applied not only for academic purposes but also for organizational goals. Already several decades ago, Zand and Sorensen (1975) pinpointed the potential of Q in facilitating interactive processes for change. For example, Q can play a role in the participation of employees in organizational change processes, whereby Q sortings initiate iterative processes of sensemaking, dialogue and the creation of shared meanings (cf. Lai and Smith, 2019). In addition to internal stakeholders, organizations can apply the Q method in order to involve external stakeholders, such as clients, consumers or suppliers, in participatory processes (cf. Peenstra and Silvius, 2018).

Among the studies in our sample, we can identify two areas in which Q served as a management intervention tool: organizational change and development and HRM management and development. With regard to organizational change and development, this study includes examples of how Q was successfully applied as a tool to identify divergent interests and perspectives of internal and external stakeholders (e.g. Ferguson et al., 2019; Howard, 1998). As regards HRM management and development, this study includes examples of how Q was successfully applied in personnel selection processes (e.g. Barrett, 1995; McCulloch and Turban, 2007), employee training (e.g. Dries et al., 2008; Mitiku et al., 2017) and as a diagnostic tool for improving employee satisfaction (e.g. Chinnis et al., 2001).

We believe that the potential of the Q method as a managerial diagnostic and intervention tool spans far beyond the specific examples that were found in this study’s sample. In the light of the notion that large percentages of organizational change efforts fail to meet their goals due to choices that lack stakeholders’ support, we recommend that managers of change consider making use of the Q method as an instrument to collect stakeholder perspectives, facilitate dialogue and create shared language and understanding. We do acknowledge that this involves challenges. One such challenge is realizing a connection and collaboration between managers and OB researchers. This requires aligning management needs with opportunities for research and identifying shared goals for managers and researchers. This may be facilitated, for example, by platforms for discussion between OB insights producers and OB insights users (cf. Edgeley et al., 2020). Besides finding and meeting with one another and establishing a partnership, researchers and managers need to get to a shared understanding of the goals and scope of a Q study project before this can materialize. Researchers may face organizational barriers that should be dealt with. For the organization, it is important that the process and results of a shared Q project are relevant from a management point of view. If productive partnerships are created, Q can play a role in fostering participative, successful change processes in groups, organizations and larger systems. This notion deserves to become much more widespread among organizations and managers of change specifically.

This study is based on extensive efforts to capture all OB studies until 2020 that use the Q-method. We have shown that very few OB studies using Q were conducted between 1960 and 1990, and that in more recent decades, especially since 2010, the number of OB studies using Q is accumulating somewhat faster. Most OB studies that applied Q have focused on group-level topics, although, in our sample, also the individual level and the organizational level were addressed. Typically, and different from what reviews of the Q method in other research fields show, many OB studies harnessed the Q method for other purposes than for the systematic study of subjectivity along the principles of Q methodology. Nonetheless, the results of this review also illustrate the unique strengths of Q in making connections in multiple respects – connections between quantitative and qualitative data analysis techniques, connections between academics and practitioners and connections between divergent groups of stakeholders. The method allows for engaging organizational stakeholders in theory building processes. Moreover, studies using Q can meet academic and practical organizational aims at the same time. Despite its still limited application within OB so far, Q as a “connecting” method has the potential to further advance the research domain of OB in important ways as well as to foster organizational processes of participative, inclusive and successful change that lead to enhanced organizational performance.