Are academics willing to forgo citations to publish in high-status journals? Examining preferences for 4* and 4-rated journal publication among UK business and management academics

Academics often judge themselves and are judged by others according to the status of the journals in which they publish. Little is known about whether individual scholars would choose to publish a paper in a high-status journal if it would garner similar or lower levels of scholarly impact than a paper published in a lower-status journal. Drawing upon status theory, we explore whether and how much business school academics are willing to ‘pay’, as captured by a hypothetical level of ‘forfeited’ citations, to publish in high-status 4* journals rather than leading specialized 4-rated journals. Using choice-set design and survey data from UK business and management scholars, we suggest and empirically demonstrate that the willingness to forgo citations to publish in 4* journals is strongest among academics who have already published in 4* and/or 4-rated journals. Contrary to our expectations, we find that an individual’s existing scholarly impact, as captured by prior citations, has no effect on this preference. We also show that academics working in high-ranked institutions would give up more citations for 4* journal publication than those working at lower-ranked institutions. We explore the implications of these findings for theories of academic status, journal rankings and research assessment systems. application of empirical methods to solve real-world issues of work, the causes and consequences of inequality of unintended consequences of research evaluation on the of knowledge.


Introduction
An increasingly common mechanism for assessing the value of research is publication in high-status journals (e.g. Garfield, 2005;Hudson and Laband, 2013;Osterloh and Frey, 2020). The status of a journal is determined by various metrics and lists, but also by its social evaluation by academics. A high-status journal can bestow benefits to the publications included within it, as well as to the authors of these publications. Accordingly, publications in high-status journals are increasingly important to individual academics in gaining academic rewards, such as scientific status, promo-tion, faculty pay and research resources (De Rond and Miller, 2005;Gomez-Mejia and Balkin, 1992;Heckman and Moktan, 2020). These publications may also be highly valued by academic institutions, because they represent a means to improve their reputation and their position in institutional rankings. In this context, the perceived value of high status in relation to a journal is rooted not just in its relative position within the academic community, but also within the wider rankings' ecosystem.
Using journal rankings to assess research value is particularly common in the field of business and management, where a small group of three-letteracronym journals are considered 'elite', 'top' or 2 R. Salandra, A. Salter and J. T. Walker 'A' journals. These journals, which we refer to as 4* journals, are often described as generalist management journals, as compared to specialist outlets. They have been labelled 'bastions', because they tend to be dominated by a small group of 24, mostly US-based, institutions (Walsh et al., 2017). These outlets are likely to receive high levels of scholarly attention, both in terms of reading time and number of citations. Yet, publications in 4* journals may not necessarily be of 'higher quality' (Singh, Haddad and Chow, 2007), and a large proportion of these journals' citations are earned by a small number of highly influential papers (Baum, 2012). Recent work has argued that highstatus journals are increasingly seen as 'the new bottom line for valuing academic research' (Aguinis et al., 2020). Indeed, publication in 4* journals may influence hiring, recruitment, promotion, positions in international and national rankings, and research funding allocations (e.g. Morris and Lancaster, 2006;Walker et al., 2018).
Differences in status between journals, the use of journal lists and the growing pressures on academics to publish in high-status journals to gain professional rewards have been widely commented on (e.g. De Rond and Miller, 2005;Pettigrew and Starkey, 2016), with critics opining that journal rankings should play no role in the evaluation of research, noting for example the affinity of ranking with the commodification of research (e.g. Connelly and Gallagher, 2010;Hogler and Gross, 2009). Yet, there are no studies, as far as we are aware, that explore whether and how the status of the journals in which their publications appear matters to individual academics. Therefore, we ask: What kind of academics are, or are not, subject to the lure of 4* journal status? How is a journal's status valued in comparison to the scholarly impact that a publication can exert when measured by its citation levels? Specifically, we set out to explore whether academics' willingness to forgo citations in exchange for 4* journal status correlates with any salient characteristics at the level of the individual researcher or their institution.
To investigate these issues, we designed a choiceset survey question that focused on individuals' preferences for publication in 4* journals in relation to a hypothetical level of citations in subsequent years. The idea is to determine at what point (i.e. at what level of 'forfeited' citation) academics would switch their publication choice away from these high-status, generalist outlets (4* journals) and towards other leading, but more specialized, outlets (4-rated journals). We use citations as a proxy for scholarly impact on the assumption that, despite their notable limitations, citations have a perceived value to individual academics as a means of demonstrating their influence on their peers. In other words, our research design seeks to capture whether and how much individuals are willing to 'pay' for a 4* journal publication, as measured by the hypothetical citations they would be prepared to forfeit in relation to publishing in a lower-status journal instead.
To explore these questions, we build on status theory (Merton, 1968;Podolny, 1993;Sauder, Lynn and Podolny, 2012), which suggests that those who aspire to status will be willing to pay to gain and maintain an association with high-status social institutions. This aspiration for status may be driven by personal motivations and also by the organizational context in which people work. In particular, we suggest that individuals who have already had their work published in 4* journals will show a higher willingness to forgo citations to publish in 4* journals in preference to 4-rated ones, as a means to further reinforce their status and to sustain their status advantage over others. We also propose that there is a strong aspirational preference for 4* journal publication among those who have published in 4-rated journals, as they seek to obtain the status benefits associated with 'breaking into the 4* journal world'. In addition, we suggest that those with high scholarly impact, as captured by their level of prior citation, are more willing to sacrifice citations for 4* outputs as a means to maintain and reinforce their position in the academic elite. Finally, we propose that individuals who work at high-ranked research institutions will be willing to 'pay' more for 4* journal publication than faculty at low-ranked institutions, because they are subject to greater status expectations in their local environment, which are often directed towards the production of such outputs.
We test our hypotheses by making use of a multisource dataset, which builds upon information we collected from a large-scale survey of academics working in UK business and management schools. In this context, we leverage the widespread dissemination of the Association of Business Schools (ABS) journal list, later renamed the Academic Journal Guide (the AJG/ABS list). The list ranks journals on a five-point scale (4*, 4, 3, 2, 1), with the highest ranking being 'Journal of 3 Distinction' or '4*', which almost exactly overlaps with the club of 'A journals' as defined within the US scholarly community. This is followed by journals with internationally leading research, referred to as 4-rated journals. 1 We find support for most of our hypotheses, indicating that the allure of highstatus journals may lead a significant number of academics to sacrifice some of the scholarly impact of their research, as measured by the citations that their future work might hypothetically garner. We explore the implications of our findings for understanding the effects of status signals on the preferences of academics, and for the management of research in business schools.
Our study makes two important contributions. First, it enhances understanding of how the aspiration for status shapes academics' attitudes to ranking systems, leading individuals either to be indifferent to them or to choose to give up significant potential influence on their peers to obtain high-status affiliations. Second, the study documents which individuals are most likely to feel these pressures on the basis of where they work and their prior research achievements, highlighting the self-reinforcing nature of rankings, status and professional aspiration. In doing so, the study helps to enrich understanding of how scholarly norms are being redefined by such systems.

Individual and institutional factors determining preference for 4* journal publications
Social status refers to the 'extent to which an individual or group is respected or admired by others' and, within a community, there is often a social hierarchy leading to the 'ordering of individuals and groups according to the amount of respect by others' (Magee and Galinsky, 2008, p. 352). These status hierarchies are often self-enforcing, because people ascribe competence to those with high status, and can create a level of expectation confirmation about relative status positions. In addition, as Magee and Galinsky (2008) suggest, when people have expectations of their status, they may behave in a way that asserts that status, leading to a behavioural confirmation of their position in the status hierarchy. Such confirmation may be internalized by different actors, leading them to accept potential inequalities and also leading to these status hierarchies becoming entrenched and embedded in organizational practices and institutional norms (Magee and Galinsky, 2008).
However, as Sauder, Lynn and Podolny (2012) suggest, status and quality may be uncoupled where there is high uncertainty about outcomes or there are competing norms of performance assessment. Indeed, Sorenson (2014, p. 63) states that the presumed link between status and quality often relies on 'superstitious learning, surmising that status and quality must move in tandem'. For example, Simcoe and Waguespack (2011) demonstrated that when high-status names and institutional affiliations were removed from computer science proposals, academics were less likely to support them. In status hierarchies, systems of worth are created that help to institutionalize the values placed on some activities and behaviours relative to others, with the underlying criteria for these evaluations embedded in different categorizations (Lamont, 2012).
In the context of academic communities, the aspiration for status can be strong, motivated by the desire of individuals and groups to gain respect from their peers for their contribution to knowledge, and also to obtain the benefits associated with such status, such as higher pay, greater resources for research, promotion, higher institutional rankings or more favourable work conditions. Although it can be expected that the aspiration for higher academic status may be strong among all academics, individuals may have different preferences in terms of how to earn these rewards. For many, publishing remains the 'primary currency' of academic life, because although service and teaching might be rewarded locally, the main driver of academic 'recognition, prestige and mobility is that of publication' (De Rond and Miller, 2005, p. 322). Within the social context of the research community, faculty members are expected to generate a strong record of publications, and failing to do so may lead to feelings of dissimilarity to the prototype of the in-group. Authors of rejected manuscripts, for example, 'face psychological rejection in that their status as authentic members of their professional social identity may be endangered' (Day, 2011, p. 708).
Business and management academics have to navigate an increasingly competitive research environment and are, accordingly, changing the way they approach publication (Nedeva, Boden and Nugroho, 2012), using various strategies to facilitate the production of top-tier publications (Seibert et al., 2017). In business and management studies, there is also a 'fragmented adhocracy' with a low consensus on research goals, and little clarity about quality standards (Vogel, Hattke and Peterson, 2017). In this context, there are strong incentives for performance managers to turn to lists to help them with evaluations, and to 'infer the quality of publications from the quality of the journals in which they are published and link contingent rewards to it, such as funding, promotion and pay' (Vogel, Hattke andPeterson, 2017, p. 1707). Indeed, Drivas and Kremmydas (2020) found that business and management academics had internalized this approach, with scholars more likely to cite papers from journals that had been promoted on the AJG/ABS list, and less likely to cite those from a journal that had been downgraded. Along these lines, Walker et al. (2019) found that business and management academics were less hostile to journal lists, if they themselves benefited from a re-grading of outlets where they had previously published. Despite this, Bryce, Dowling and Lucey (2020) found a significant journal-quality perception gap between AJG 2018 rankings of journals and the views of the UK business and management academics who are arguably most familiar with them. In theory, objective measures of research impact, such as citation levels, might provide an antidote to such 'list fetishism'. However, citations to individual papers remain an imperfect measure of scholarly influence, because they may take time to emerge (Wang, Veugelers and Stephane, 2017), may be a product of visibility rather than quality (Merton, 1973) and may arise from the strategic citation efforts of other academics (Baum, 2012).
Drawing on status theory (Merton, 1968;Podolny, 1993;Sauder, Lynn and Podolny, 2012), we suggest that some business and management academics will be willing to accept fewer citations of their work in order to gain the benefits of being associated with higher-status journals. We start by focusing on the case of those who have previously published in 4* journals, expecting the willingness to 'pay' for publication in 4* rather than 4-rated journals -in terms of hypothetical citations -to be strongest among this group. First, individuals who have had their work published in 4* outlets will seek opportunities to further reinforce their status, helping to sustain and enhance their status advantage over others (Sauder, Lynn and Podolny, 2012). These individuals may already have reaped the benefits associated with previous 4* outputs in terms of higher pay or better working conditions and will, therefore, have critical information about the advantages stemming from these outputs, which others may be less aware of. Second, because publishing in 4* journals is difficult and only a select few can achieve it consistently (Baum, 2012), 4* journal publication will provide a source of visibility or a signal of status to others, in a way that publication in 4-rated outlets may not. Thus, additional publications in 4-rated journals, even with significant impact, will be less valued than the potential to signal reaffirmation of an affiliation to a high-status outlet. Thus: H1a: The willingness to pay, in terms of citations, for 4* journal publications in preference to 4rated ones will be greater among those who have previously published in 4* journals than those who have not.
The willingness to pay in terms of citations for publishing research in 4* journals is also liable to be strong among those individuals who have published in 4-rated journals, yet aspire to even higher academic rewards. First, individuals who have demonstrated their ability to publish in leading specialized journals clearly have academic competency in terms of high-impact outputs. However, such competency may not be fully recognized by those working outside the particular core field domain concerned. Publication in a 4* journal may be regarded by these academics as an opportunity to claim a 'golden ticket' and demonstrate that their status within their specific domain is not dependent on exclusive features of that context. Such publications may be seen as the 'icing on the cake' for their scholarly performance (Gomez-Mejia and Balkin, 1992). This is because 'these second-tier publications are less likely to be rewarded in isolation, in the absence of a demonstrated scholarly record in top-tier, or premier, outlets' (Gomez-Mejia and Balkin, 1992, p. 924). Second, citations of a 4-rated journal publication may be perceived by these individuals as having less value than publication in a 4* journal, because these latter are rarer and have potentially higher labour-market value. Thus, scholars with 4-rated journal outputs may be willing to pay a higher price in terms of forfeited citations than scholars with portfolio outputs from 3-rated journals or lower. Furthermore, scholars without a 4-rated output in their portfolio may consider a 4* journal paper to be beyond their current aspiration level. For these individuals, looking from a distance at the 4-rated journal club, the potential status benefits of a 4* journal publication compared to a 4rated one will be seen as a secondary concern, and they will, therefore, show a lower willingness to pay in terms of citations for such an outcome. Thus: H1b: The willingness to pay, in terms of citations, for 4* journal publications in preference to 4rated ones will be greater among those who have previously published in 4-rated journals than those who have not.
We expect that the willingness to pay in terms of citations for 4* journal publications will also be great among those possessing the highest scholarly impact, as measured by numbers of citations to their prior work. These scholars are likely to regard 4* journal outputs as important drivers of status reinforcement, demonstrating their ability to participate in the elite level of academic competition. These individuals, inclined to judge their relative status in the academic community against their leading international peers, might also see publication in 4* journals as a means for recognition by other members of this in-group, helping to reaffirm and cement their affiliation to the status hierarchy. For these individuals, a publication in a generalist journal can add more value than an output in a leading specialized journal, where their status may already be visible. In addition, an output in a 4* journal may help garner more citations of their past work, because it is a notable status-raising achievement, akin to winning a prize (Azoulay, Stuart and Wang, 2014).
Conversely, it can be expected that faculty with more modest levels of scholarly impact in the academic community will assign less value to the status advantages of 4* journals compared to those of good-quality specialized journals. For these less influential scholars, 4* journals might represent too distant a target, and they may consider these journals well-fortified bastions that they lack the skills and networks to penetrate, leaving them unwilling to forgo citations in order to achieve this (Walsh et al., 2017). For these scholars, it might also be the case that targeting 4* journals means not publishing at all, losing not just citations, but volume of publications too. Thus: H1c: The higher an individual's scholarly impact, as reflected by their previous citation count, the greater their willingness to pay, in terms of citations, for 4* journal publications in preference to 4-rated ones.
The preference of an academic in terms of journal status may also relate to their work context. There is considerable variation in business and management schools. In the UK, for example, some institutions have a strong orientation towards research, whereas others are more directly concerned with servicing local industry and/or educational functions. These institutional differences are liable to matter when it comes to shaping individuals' attitudes towards 4* journals, because academics will need to conform to the expectations of their institutions and align their behaviours with the goals of their employers (Pelger and Grottke, 2015). The use of outputs in specific journals in a wide range of international ranking exercises, such as the business school rankings from the Financial Times (FT) and the QS World University Rankings, and the strong link between the relative position of these institutions in such rankings and their teaching income, also mean that institutions are liable to tailor their recruitment, promotion and performance systems to the production of research that can be published in the top journals (Van Fleet, McWilliams and Siegel, 2000).
Because highly ranked research institutions are likely to expect their faculty to engage in elite research status competitions, we anticipate that academics working in such institutions will be more willing to pay in terms of citations for outputs in 4* journals. In these institutions, 4* journal publication is liable to confer strong personal career advancement benefits and, in turn, the institutions will seek to recruit and promote those who demonstrate the ability to enter the bastions of the 4* journal club. Publishing in 4* journals also helps high-ranked institutions maintain their relative status position in various external rankings, which plays an important role in student recruitment. These organizations also tend to hire from other elite institutions, wherein early-career scholars are also advised to target 4* journals (Van Fleet, McWilliams and Siegel, 2000).
In contrast, at low-ranked research institutions, the perceived pressure on individual faculty to publish in 4* journals as opposed to 4-rated ones is likely to be modest. These organizations are unlikely to organize their hiring, promotion and reward systems around 4* journal publication, focusing instead on more realistic and immediate outputs. In these contexts, a 4-rated journal publication may itself be considered a significant achievement, and lead to rewards for faculty members. Moreover, because there is a paucity of 4* journal papers in these institutions in general, it can be expected that there are only weak expectations on faculty to publish in such outlets. In addition, rarely do these institutions rely on their relative position in international rankings to drive student recruitment, and they are, therefore, less likely to place pressure on their faculty to publish in the journals embedded in these rankings. Thus: H2: The willingness to pay, in terms of citations, for 4* journal publications in preference to 4rated ones will be greater among those at highranked research institutions than those at lowranked ones.

Research setting, data and measures
UK business schools and the AJG/ABS list. In the UK, the national assessment of institutional research quality -initially, the Research Assessment Exercise (RAE) and subsequently, the Research Excellence Framework (REF) -is based on peer review by panels of subject matter experts (see Bessant et al., 2003;Geary, Marriott and Rowlinson, 2004;Pidd and Broadbent, 2015). Although these experts are instructed to give no weight to the status of journals in their assessment, research managers at UK business and management institutions developed lists of journals to try to preempt the judgements of these review panels. These journal lists, started as institution-specific exercises, were consolidated by the ABS in 2007. The ABS list was later renamed the Academic Journal Guide, and hence is usually referred to as the AJG/ABS list. The AJG/ABS list ranks journals on a scale, with the highest ranking for 'Journal of Distinction' (4* journals), followed by 4rated journals with internationally leading re-search, which tend to be more specialist in nature and more likely to be based outside the USA.
The institutionalization of the AJG/ABS list provides an ideal setting in which to explore whether academics prefer to publish papers in journals of a given rank. First, the list has a clear status hierarchy, which is not determined as simply as in other citation-based journal metrics. Second, almost all business schools in the UK use the AJG/ABS list and, therefore, business and management academics are highly aware of the list and its rankings. Third, the diversity of sub-disciplines within UK business and management schools, and the ubiquity of the list, together provide a useful basis for comparing how academics from a broad range of subject backgrounds view journal status, while operating within similar institutional settings.

Data sources
This study is based on a survey of academics working in business and management schools in the UK, which we ran in 2015, combined with other data sources. It is a part of a larger research programme, including Salter et al. 2017, Walker et al. 2019, that explores attitudes toward and use of the ABS/AJG list. We compiled our survey list by accessing the websites of all the business and management schools in the UK, identified as the institutions that submitted to the 2008 RAE, plus University College London. We collected email address, rank and gender for all scholars. Because our focus was on research-active and permanent staff, we excluded faculty members who were not assistant, associate or full professors. These procedures allowed us to identify a population of 8,002 research-active individuals affiliated to 90 UK business and management schools. The subsequent survey received 1,945 responses, representing a 24% response rate.
We downloaded the publication portfolio of the surveyed individuals using Scopus. Finally, we collected information at the level of the institutions, such as their ranking in the UK's 2014 REF assessment. After cleaning and matching the data, we were left with a sample of 1,154 usable questionnaires.

Measures
Dependent variable. The dependent variable is Preference for 4* journals. Because the issue of whether academics would prefer to publish in a 4* journal rather than a 4-rated one had not been addressed in the previous literature, we developed a new survey question. Our question design was inspired by studies of managerial choice in innovation and research policy studies Henkel, 2012, 2013;Salter et al., 2017;Schillebeeckx et al., 2016), as well as research on risk preferences in behavioural economics (Cohen and Einav, 2007;Santos-Pinto et al., 2015). In these studies, individuals are offered a set of hypothetical staged choices between two options, which have different degrees of risk. The choice-set fixes the value of one option and then incrementally increases the reward associated with the second option. The idea is to determine at what stage individuals shift their preference from the first option to the second. Building upon this logic, we designed a hypothetical choice-set question to assess the preference for a publication in a 4* journal versus one in a 4-rated journal. Because citations are commonly used, easy to measure and well understood by academics, we use them in the hypothetical set as a proxy for the reward associated with each option. To present this as neutrally as possible, we specified that the question related to individual preference and hence that there was no right or wrong answer. The exact wording of the question is shown in Figure 1.
We started with the simple case in which the 4* journal publication was compared to a 4-rated one with an equal number of citations. In this case, each paper would receive 25 citations in the 10 years following publication, indicative of a paper of modest, but not high, academic interest. Any respondent that chose a 4-rated journal in preference to a 4* one in this scenario would 'depart' the question, having suggested that they would prefer a 4-rated journal publication regardless. For the remainder, indicating an initial preference for 4* journals, we progressively increased the offer in terms of citations associated with a 4-rated journal publication, trying to induce a shift in preference to the 4-rated journal and capturing the point at which this shift might occur. The underlying logic being that the higher the number of citations required to favour a 4-rated journal publication, the higher the respondent's preference for a 4* journal one.
Although the choice-set is hypothetical in the sense that it is, ex ante, impossible to know the level of citations a paper will receive, it does enable the respondent to express their preferences for differently rated journals in a form of 'currency' that is understood and valued in the community. The choice-set was intended to unearth preferences in relation to journal status, removing the possibility that a respondent would be able to 'have their cake and eat it too' by simultaneously opting for a high-status journal and a high citation level. 8 R. Salandra, A. Salter and J. T. Walker Key independent variables. At the level of the individual, we first examined whether an individual had any publications in 4* journals in their publication history, a measure that captures their existing ability to gain entry to these 'bastion' journals (Publication in 4* journals -a dummy variable taking the value 1 if the individual had published in 4* journals, and 0 otherwise). Second, we captured whether an individual had published in 4rated journals (Publication in 4-rated journals). Finally, we considered an individual's scholarly impact. To measure this, we counted the total number of citations an individual had received across their publication portfolio, as indicated in their Scopus record (Citations).
At institution level, we considered the institution's REF Grade Point Average (GPA). This is a measure of the average quality of research, computed from the REF Summary for business and management. The GPA was captured in the variable Research rank of the institution.
Control variables. Several control variables were included in the model to better understand the sample and to account for possible training effects.
Gender (Male). To control for any differences in the preferences for journal status between female and male academics, we created a dummy variable that took the value 1 for male, and 0 for female.
Non-academic work experience. Business and management academics have often had hybrid careers, working both inside and outside academe (Clarysse, Tartari and Salter, 2011;Lin and Bozeman, 2006). Individuals who have spent a long time outside academe may be less socialized into pressure to publish. To account for this potential influence on journal publication preference, we controlled for the number of years an individual had indicated having worked outside academe in our survey.
Years since obtained PhD. To control for potential influences on individual preferences arising from their length of stay in academia (e.g. the newer generation of business and management academics are more conditioned to journal rankings, because of recent changes in the labour market and the widespread role of such rankings in hiring and promotion), we incorporated a variable to capture the number of years elapsed since a respondent had gained their PhD. This was based on a survey question.
Obtained PhD in the USA. The USA has a dominance in the editorial boards of most 4* journals and a tenure structure that is tightly aligned to publication in 4* journals. To capture these influences, we created a binary variable that took the value 1 if individuals had received their PhD in the USA, and 0 otherwise. We derived this information from a survey question.
Rank. To account for differences in pressure to publish in 4* journals experienced within different academic ranks, we considered an individual's position within their institution. Based on a survey question, we created three dichotomous variables accordingly: Lecturer, Associate Professor and Professor (baseline category, removed).
Size of the institution. We included the headcount of each institution to control for any potential influence on individuals' preferences arising from differing levels of resources available in their host institutions. These figures were based on our web search of staff at the business and management schools included in the survey.
Area of expertise. Different specialties may exhibit different publication and citation norms (Liu, Olivola and Kovács, 2017). Thus, we included area-level dummies, based on the primary field of expertise indicated by our survey respondents, given a choice of 22 discipline areas derived from the field classifications used in the 2015 AJG/ABS list.

Results
Descriptive statistics of the main variables are given in Table 1. The largest proportions of respondents were male (57%) and were Associate Professors (36%). Only 12% of the surveyed individuals had published in a 4* journal during their career, while around 37% had published in a 4-rated journal. This suggests that for much of our sample, both 4* and 4rated journals represent alluring but hard-to-reach targets. 2 Pairwise correlations between all variables are also included in Table 1. As expected, some of the 2 A small minority of survey participants had published in 4* but not 4-rated journals (n = 11). In robustness checks, we removed these individuals from the sample and re-ran the analysis. The findings were not qualitatively different from those presented (in Table 3).    Table 2 reports the means of variables across the different options chosen, that is, the different levels of citations to a paper in a 4-rated journal that an individual would require before forgoing publication in a 4* journal. Splitting the sample between the three trade-off categories introduced in the survey (i.e. 25, 50 and 100+ citations) highlights some observable differences within a number of individual characteristics. First, a substantial share of UK business and management academics do not differentiate between 4* journals and 4-rated journals. Thus, given the same level of citations in each, about 44% of our survey respondents (504 out of 1,154, falling in the first column of Table 2) opted for publication in a 4-rated journal rather than a 4* one.
There are also substantial differences in the outcomes in relation to gender, non-academic work experience and PhD training location, which will be discussed in more detail later in this section. 3 Notably, almost 10% of respondents that indicated a strong preference for 4* journals (as reported in the last column of Table 2) have a PhD from a US institution, compared to 2.6% of those who preferred 4-rated journals (as reported in the first column of Table 2).
Although the proportion of the sample that has published in 4* journals is small (12%), these individuals require a considerable number of citations to compensate for a 4-rated journal not being a 4* one (19.0% of these individuals fall into the last column of Table 2, compared to 6.7% in the first column).
In order to appropriately model the sequential nature of the decision-making process in the choice-set, we used a sequential logit model, as detailed by Buis (2013). Table 3 reports the results of this, which takes '4-rated journal paper with 25 citations preferred to a 4* journal paper with 25 citations' as the reference category. To ease interpretation, odds ratios (ORs) are calculated and reported throughout. Coefficients greater than 1 indicate a preference for a 4* journal, in other words an increased likelihood of switching from a 4* journal to a 4-rated journal publication only at 50 or 100 citations, compared to 'accepting' the latter with just 25 citations.
In Table 3, Model 1 represents the baseline model, including only the control variables; Models 2-5 introduce the key independent variables, while Model 6 includes all the variables. All models include area-of-expertise dummies to control for heterogeneity across business and management sub-disciplines. Notes: z-statistics in parentheses.
Baseline category = 4-rated journal paper with 25 citations preferred to a 4* journal paper with 25 citations (n = 1,154).

R. Salandra, A. Salter and J. T. Walker
We focus our discussion on the results for Model 6, which is the full model. The first column refers to the option of a '4-rated journal paper with 50 citations preferred to a 4* journal paper with 25 citations'; the second column option is a '4rated journal paper with 100 citations preferred to a 4* journal paper with 25 citations'. First, we observe that most coefficients are only significant in relation to the second column. This suggests that while the academics requiring 100 citations before they will choose a 4-rated journal publication are significantly different from those accepting a 4-rated journal paper at the first option given in the choice-set (our baseline category), the respondents opting for the middle option (requiring 50 citations) are not significantly different from the baseline group in relation to the dimensions explored. For example, gender appears to play a role in informing preferences for journal status, but only for those academics that chose the highest number of citations before they would 'renounce' a 4* publication. Thus, male academics are significantly more likely than female ones to require 100 citations (as opposed to 25 citations, our baseline) before preferring a 4-rated journal (OR = 1.39), but gender is not statistically significant when we consider academics that require 50 citations before preferring a 4-rated journal (OR = 1.21).
Some of the results for the other control variables are interesting too. The length of time an individual had been in academia (Years since obtained PhD) and their orientation to practice, as proxied by their Non-academic work experience, both correlate with a lower offer of citations for a paper in a 4* journal (OR = 0.96 and 0.97, respectively). By contrast, individuals in our sample who earned their PhD from US universities have a strong preference for 4* journals (i.e. they are more likely to require a high offer in terms of citations before they will prefer a 4-rated journal) in comparison to colleagues who trained elsewhere (OR = 2.33). With regard to Rank, although the odds ratios for Associate Professor and Lecturer are both below 1, suggesting that middle and lower-ranked academics are less likely than Professors to require 100 citations before they will prefer a 4-rated journal publication, the effects are not statistically significant.
The field of expertise does not appear to influence preference for 4* journal publication. Other than for the Finance sub-discipline, whose scholars are more likely than those in other fields to require 100+ citations to prefer a 4-rated journal (OR = 2.06), area-level dummies, not reported in the table, are insignificant. 4 We find that individuals who had published in 4* journals have a stronger preference for these outputs, compared to those who had not published in 4* journals (OR = 1.55), providing support for H1a. However, it is worth remembering that this variable is highly skewed, given that only a relatively small proportion of academics in our sample had published in 4* journals at all. Individuals who had published in 4-rated journals also have a strong preference for publishing in 4* journals (OR = 2.45). As such, H1b is also supported. Interestingly, we find that this odds ratio is higher for those who have already published in 4-rated journals than for those who have already published in 4* journals (OR = 2.45 vs OR = 1.55). By contrast, scholarly impact, as captured by prior citations, does not appear to have a substantial effect on individual preferences (OR = 1.00). Hence, H1c is not supported.
Finally, faculty employed in institutions where research quality is deemed to be high (as measured by the GPA) have a strong preference for 4* journals (OR = 2.73). As such, we find support for H2.

Discussion
To help understand how status considerations shape academic preferences towards publishing in differently ranked journals, we created a hypothetical choice-set to prompt academics to express such preferences. We found that for many academics in our sample, the appeal of 4* journal status is fairly modest; almost half of our survey respondents (44%) preferred publication in a 4-rated journal to that in a 4* journal, even given the same hypothetical 'offer' of 25 citations from either one. This suggests that a large proportion of academics, in this case UK-based business and management scholars, may choose to distance themselves from the 'race for status' deriving from 4* journals, and are content to focus on advancing their research in 4-rated ones instead. For example, many individuals may perceive that their research is distant from the types of research that are published in 4* journals. This apparent indifference to 4* journals could also arise from an individual's affinity with a specific field; for example, they may prefer to publish in journals that are specialized accordingly, as opposed to being generalist. More pragmatically, some academics might also believe that publishing in a 4* journal takes much longer than publishing in a lower-ranked journal, resulting in lower publication counts over time.
We found that the preference for high-status journals was high among those academics with prior 4* and/or 4-rated journal papers. Some individuals in our sample are so socialized to the benefits of 4* journal publication that they would forgo significant scholarly impact (as measured by citations) to achieve it. Indeed, scholars who have published in 4-rated journals appear to be the keenest to obtain such 4* journal distinction, because they are already 'knocking on the door'. This suggests that those with, and those with aspirations to obtain, high-status affiliations may view such outputs as the 'primary currency' of academic life (De Rond and Miller, 2005, p. 322), indicating that status hierarchies via ranking systems have become entrenched and embedded in the preferences of a significant share of business school academics. However, we do not find any conclusive evidence regarding the relationship between existing scholarly impact, as captured by citation levels, and preference for journal status. This could be because those with high scholarly impact already have high academic status, and therefore do not perceive the need to sacrifice citations in exchange for 4* journal publication.
Exploring institutional differences, we found that faculty based at high-ranked institutions were more likely to prefer publication in 4* journals than in 4-rated ones. Interestingly, this indicates that the willingness to sacrifice scholarly influence is greatest in those environments most focused on the production of 4* outputs, which appears to be in direct contradiction to the declared goals of these institutions, of being recognized for their scholarly impact. It may be the case that ranking systems have led these institutions to shift away from ostensive demonstrations of scholarly influence, and towards performative ones.

Theoretical implications
This study contributes to an understanding of how status aspiration shapes academics' attitudes towards ranking systems. Our findings indicate that while some scholars will seek opportunities to reinforce their status on the basis of the journals in which they publish (e.g. Willmott, 2011), others place less value on publications in high-rank journals than on the potential to generate scholarly impact on the academic community, as captured by citations. Academics in the former group are, in effect, indicating that they would rather publish with modest academic impact in a 4* journal than have a major impact on their field while publishing in less prestigious outlets. This would appear to be a case in which status-quality links lead to what Sorenson (2014) described as a form of 'superstitious learning', based on the idea that status and quality should move together. Of course, these individual preferences may be fed by wider pathologies that trickle down to academics through reward and ranking systems that privilege outputs over influence. Such views appear to be concentrated within the upper reaches of the academic hierarchy, with those with high aspirations for status being the ones that seek publication in 4* journals most avidly, highlighting the self-reinforcing cycle of rankings and professional aspirations.
The study also helps to advance understanding of how scholarly norms are redefined by rankings and aspiration for status. The findings suggest that Merton's (1973) original formulation of academic status needs to be broadened beyond the individual and their pursuit of scholarly influence via publication, and towards a more mixed model in which individuals' personal aspirations for status are bundled into the value accorded to their research appearing in specific outlets. In effect, academic status may no longer be judged by an individual's influence among their peers, but also by their presence in particular high-status outlets. This shift in outlook appears to be partly driven by the need for academic institutions to perform well in various rankings, thereby placing their staff under increasing pressure to align their expectations and behaviour with the needs of their employers (Pelger and Grottke, 2015).

Practical implications
Our research has important practical implications. Researchers and university managers should be mindful that the pursuit of status may drive changes in the way that academics approach publication (Nedeva, Boden and Nugroho, 2012), and might explain the use of various strategies (e.g. coauthoring) to facilitate the production of top-tier publication (Seibert et al., 2017). It is important to acknowledge that there is still considerable diversity among business and management academics in their willingness to 'pay' in terms of scholarly influence for publishing in 4* outlets, with many academics appearing immune to the lure of journal status in preference to academic influence. This diversity is a bulwark against the tendency to focus on the performative aspects of journal rankings and lists in the face of the ostensive challenge of generating academic influence. We would encourage research managers not to lose sight of the value of scholarly influence (as well as social and economic impact) as an important goal in itself, rather than simply attempting to service the demands of ranking systems.
In terms of the design of research systems and scientific performance indicators, it may be the case that 'hybrid' systems that combine revealed preferences (e.g. based on citations) and stated preferences (e.g. expert-based) are better aligned with the range of preferences that exist within the academic community. Although citation data have become more accessible and powerful, our results confirm that, for many, journal-level metrics still reign supreme over paper-level and authorlevel metrics in the appreciation of the 'value' of research. These preferences are sustained even when the high skew in the distribution of citations among papers in 4* journals has been repeatedly demonstrated (Baum, 2012). Until paper-level citations are accorded greater value within the academic community and by institutions, it is likely that many academics will continue to prioritize the status of the outlet over the impact of the work reported therein. Aguinis et al. (2020, p. 141) sum matters up thus: 'a 4* journal paper may often be celebrated as a victory with relatively little conversation about the study's content, the quality of its methodology and data, and the implications of its findings for theory and practice'. The continued use of journal rankings, such as the AJG/ABS list, may further entrench the prominence of 4* journals at the expense of more diverse perspectives (Hussain, Liu and Miller, 2020) and more substantive discussion of the research itself. As such, our findings reinforce the call for the use of 'responsible metrics', such as the Leiden Manifesto (Hicks et al., 2015), to combat reification of journal status.

Limitations, future research and conclusions
Our study has a number of limitations that give rise to a range of future research questions. First, our study is based on a survey of academics at UK business and management schools. As such, our results may be hard to generalize to other academic communities. While lists and journal metrics are used extensively outside the UK, institutions elsewhere are also subject to a range of domestic pressures and regulations. Indeed, our finding relating to the academics who trained in the USA provides some indication that such contextual differences may be substantial. Future research should seek to explore the preferences for journal status among scholars working in different national contexts and in other academic fields, as well as within the various business and management sub-disciplines. These should include the finance sub-field, which has exhibited some distinction from other fields in business and management in our study.
Second, our survey focused on the trade-off between 4* journal and 4-rated journal publications. However, this trade-off may not be representative of the publication options faced by the wider academic community. While our study explores the premium placed on a small group of elite 4* journals, with the internationally highly regarded 4rated journals serving as a benchmark, future research could consider other trade-offs that may be more prevalent within UK institutions; for example, that between 4-rated and 3-rated journals. 5 Relatedly, our choice-set question focuses on the case of a single publication. Yet, it is often the case that publications are seen in terms of a portfolio of quality versus quantity -for example, academic probation criteria may refer to publication of three 4-rated journal papers versus one 4* jour-nal paper. By taking a portfolio view, future research could give greater attention to individual preferences for different publication profiles. In addition, we use citations as the 'currency' of our choice-set question, which centres around a hypothetical trade-off. Citations are liable to be well understood by academics because they are routinely used to provide a measure of the scholarly impact of their work, and in recent years there has been a generalized increase in the use of citation tools (e.g. Scopus, Web of Science, Google Scholar) by institutions and ranking systems. However, citations have notable limitations and are only one of many sources of academic recognition for individual academics (e.g. awards, honorary positions, speed of promotion). Follow-on work could explore how stated preferences may change according to the 'currency' under consideration, and how academics may then differently communicate and contemplate journal status and citations (e.g. in appointments, promotions, research grants).
Third, future work should investigate more closely the mechanisms at play and the relative weight of individual and institutional factors, which we were only able to partly capture. For example, future studies could investigate in more detail the pressures for high-status publications that arise both informally (e.g. relative to expectations) and formally (e.g. from probation or promotion practices).
To conclude, despite these limitations, we hope that our findings may stimulate further research on the factors informing academics' preferences for status and scholarly impact, and on how assessment and incentive systems should be designed to align with such varied preferences.
Rossella Salandra is a Senior Lecturer at the School of Management at the University of Bath. Her work focuses on the dissemination of scientific information and on the working practices and preferences of academics, and has been published in the British Journal of Management, Research Evaluation and Research Policy.
Ammon Salter is a Professor of Innovation and Technology Management at the School of Management at the University of Bath. His recent research focuses on open innovation, university-industry collaboration, R&D decision-making and research assessment in business and management.