A cohort study of how faculty in LIS schools perceive and engage with open-access publishing

Abstract

This article presents results from a survey of faculty in North American Library and Information Studies (LIS) schools about their attitudes towards and experience with open-access publishing. As a follow-up to a similar survey conducted in 2013, the article also outlines the differences in beliefs about and engagement with open access that have occurred between 2013 and 2018. Although faculty in LIS schools are proponents of free access to research, journal publication choices remain informed by traditional considerations such as prestige and impact factor. Engagement with open access has increased significantly, while perceptions of open access have remained relatively stable between 2013 and 2018. Nonetheless, those faculty who have published in an open-access journal or are more knowledgeable about open access tend to be more convinced about the quality of open-access publications and less apprehensive about open-access publishing than those who have no publishing experience with open-access journals or who are less knowledgeable about various open-access modalities. Willingness to comply with gold open-access mandates has increased significantly since 2013.

Keywords

Faculty authors journals library and information science faculty open access open-access publishing scholarly communication

1. Introduction

The dawn of scholarly publishing is typically traced to the first Secretary of the Royal Society of London, Henry Oldenburg, who oversaw the publication of Philosophical Transactions beginning in 1665, although it was not until 1752 that the Royal Society took over the journal as a society journal. The scholarly publishing system remained largely confined to learned societies for roughly the following two centuries, until commercial publishers began to recognise and exploit the profit potential of academic literature. This more contemporary commercialisation of academic journal publishing has been traced to the 1947 launch of Biochimica et Biophysica Acta [1]. This transformation in academic publishing based on the profit motive proceeded apace during the following decades, resulting in a contemporary multi-billion dollar industry dominated by a handful of international publishing companies. Industry consolidation, working in tandem with the unique supply side and captured demand side of the market, gave rise to what is commonly referred to as a ‘serials crisis’, which is shorthand for a double-pronged dilemma faced by academic libraries beginning in the 1990s: skyrocketing journal prices coupled with static or declining library budgets. Partly in response to this, a sustained movement began emerging in the late 1990s that advocates for and develops open-access models for academic research. The undoubtedly most well-known statement about the importance of open access emerged from a conference convened by the Open Society Institute in Budapest in December 2001 to interrogate issues around open access to scholarly research. This conference, which laid the foundation for the subsequent Budapest Open Access Initiative (BOAI), was one of the defining moments of the then nascent open-access movement.

Informed by the underlying premise that scholarly research should be freely accessible online, the BOAI suggests two complementary strategies to achieve and sustain such access. The first suggestion is self-archiving. The second strategy relies on expanding the number of open-access journals, both de novo journals and those that are transitioned to open access [2]. Given the underlying emphasis on free access that informs it, the BOAI suggests that open-access journals should avoid price barriers to access by eliminating subscription or user fees. Almost immediately researchers began investigating electronic access to scholarly publications from a wide range of disciplinary and stakeholder perspectives. Thus, there now exists a robust corpus of literature about open access to scholarly literature, a topic germane to one of the leading foundational tenets of Library and Information Science/Studies (LIS) research and practice. While LIS faculty members and academic librarians have participated in and conducted some of this work, surprisingly little research that focuses exclusively on faculty in LIS schools has been undertaken. This represents a significant gap in the extant literature.

In order to respond to this gap, in 2013 the present author administered and reported on a comprehensive online survey that explored North American LIS faculty awareness of, assessment of and experience with open-access scholarly publishing [3,4]. This article discusses a follow-up project in 2018 that investigated these same elements of LIS faculty engagement with open-access publishing and whether any differences have occurred since 2013. The findings presented below are based on both descriptive and inferential analyses of these survey results. Inferential analyses were conducted for the 2018 data to explore whether attitudes towards and experience with open-access publishing differ across respondents based on faculty rank, perceptions and assessments of open-access publications, beliefs about how tenure and promotion committees would assess open-access publications, experience publishing in open-access journals, knowledgeability about open access and likelihood of publishing an article in an open-access journal within a year. Additional inferential analyses compared the datasets from 2013 and 2018 to ascertain whether there have been any statistically significant changes in attitudes towards and engagement with open access during these 5 years.

2. Literature review

2.1. Engagement with scholarly (open-access) publishing

While determining that a number of the original barriers to the adoption of open-access publishing have declined over the 10-year period, 2003–2013, Björk [5] has noted that the academic reward system as well as issues around marketing and critical mass continue to hinder broader engagement with both gold and green open access.¹ The green model remains very much dependent on the permissions that subscription-based journal publishers grant to authors, although research has revealed that green open access is limited not only by publisher permissions. Even among researchers writing about open access, engagement with this mode of scholarly publishing is often muted. In a relatively early analysis of articles indexed in Library and Information Studies Abstracts in 2008, Mercer [6] determined that academic librarians were not making as much of their published work available through green open-access methods as would have been permitted based on publisher policies. In their bibliometric study of the period 2003–2011, Grandbois and Beheshti [7] calculated that the rate of self-archiving (37%) by LIS authors writing articles about open access was much lower than that would have been permitted by publishers (98%). In a survey of faculty in the universities and colleges of the University System of Georgia about their open-access publishing practices and use of the system’s institutional repository, Campbell-Meier [8] ascertained that almost 73% of the faculty respondents had never deposited research outputs into the system repository. Of those respondents who had deposited a research output into the repository, only 14% had done so voluntarily. Such findings reinforce the importance of repository deposit mandates, which, according to Gargouri et al. [9], can triple the rate of engagement with green open access.

As Chang [10] has pointed out, the still relative youth of many open-access journals when compared with established traditional journals means that the former have often not yet accrued the prestige and the reputation for quality that typically help inform journal publication choices by academics who must navigate the scholarly evaluation system. Indeed, in their survey of faculty who had received funding from the Berkeley Research Impact Initiative to publish an article in a gold open-access journal, Teplitzky and Phillips [11] determined that prestige and quality of the journal were the top criteria that faculty authors considered when deciding where to publish their research. As the triennial Ithaka S+R US Faculty Survey of academics working at 4-year colleges and universities in the United States has found, the three highest-ranked considerations for deciding in which journals to publish were fit between scope of journal and research, wide circulation and a readership composed of scholars in the research field and impact factor or academic reputation of the journal [12]. Solomon and Björk [13] obtained similar results from their survey of 1038 authors who had recently published an article in an open-access journal that levied an article processing charge. The top three factors that their survey participants took into account when deciding where to submit a manuscript were fit of the research to the scope of the journal, the scientific quality of the journal, which, in some cases, was considered a reflection of the journal’s impact factor, and the speed of peer review and publication. Ranked behind these concerns was the open-access status of a journal, which 60% of the respondents considered to be important to some degree. While ostensibly this is a good percentage of survey participants who attributed importance to open-access availability, one might expect a larger number since the survey was administered to authors who had actually published in an open-access journal [13]. The triennial Ithaka S+R US Faculty Surveys have also consistently found that free online availability of their research was ranked lowest by faculty authors when contemplating appropriate publication venues.

2.2. Perceived advantages and disadvantages of publishing in open-access journals

In addition to analysing citation counts for 28,435 articles appearing from 2006 through 2011 in journals published by the American Chemical Society, Nelson and Eggett [14] surveyed open-access authors. They found that the two leading reasons why the majority of chemistry authors included in their sample made their articles open access were to comply with formal open-access mandates and to improve the likelihood of receiving more citations for their articles. Moreover, they determined that only 7% of these authors agreed that the article they made open access was better than most of their other articles. Nelson and Eggett [14] suggest that this finding goes some way in defusing the charge of self-selection bias articulated by some researchers critical of the claim about citation advantages for open-access journals [15].

Teplitzky and Phillips [11] found that an overwhelming majority (82%) of the open-access authors they surveyed believed that open-access availability gave their article greater overall impact than articles they had published in subscription-based journals. These researchers further determined that the top three perceived benefits of publishing in an open-access journal were greater number of views, more downloads, and increased citations. Similarly, research conducted among authors who had published in any of the Taylor & Francis’s journals in 2012 revealed that the three top-rated potential advantages of open-access publications compared with subscription-based journals were wider circulation (81%), higher visibility (65%) and faster publication (55%) [16,17]. Despite these positive qualities attributed to open access, this mode of publishing and dissemination continues to meet with some ambivalence and scepticism within the academy.

Based on an analysis of previous studies, Xia [18] ascertained that reasons for not publishing in open-access journals include unfamiliarity with appropriate venues, concerns about low prestige, lack of rigorous peer review, low impact factors and corresponding poor citation rates. Among survey respondents who had published in a Taylor & Francis journal in 2012, 35% agreed to some extent that open-access journals are of lower quality than subscription journals and 31% believed that open-access journals have comparatively lower production standards [16,17]. A little more than half of the faculty respondents in the University System of Georgia appeared sceptical of the scholarly rigour of open-access journals, believing that fewer articles would be rejected by such journals. Indeed, only 26% of these survey participants assessed open-access journals as being of high quality [8].

Dalton [19] observed that career-related factors strongly influenced library faculty and practitioners’ decisions about appropriate journals in which to publish their work. Similarly, although Moore [20] found evidence of emerging forms of digital scholarship, the constraints of the merit and tenure and promotion processes were perceived by faculty at the University of Toronto as reinforcing the traditional publishing system. In a survey of full-time faculty and administrators at the University of Scranton, Aulisio [21] determined that less than half of the respondents believed that publishing in an open-access journal would help a person’s tenure and/or promotion file and almost one-fifth of the respondents (18.5%) voiced concern that an open-access publication would be detrimental to tenure and promotion prospects. That belief increased to 28% among untenured faculty respondents. Interestingly, almost 59% of the respondents indicated apprehension about negative colleague perceptions of open-access publishing, while slightly less than a quarter of survey participants (23%) themselves articulated reservations about the quality of open-access journals [21].

2.3. Open-access article processing charges

Another potential hindrance to greater engagement with open-access publishing is the article processing charges that figure prominently in gold and hybrid models. In her work, Creaser [22] determined that researchers from the arts, humanities and social sciences were the least likely to know how to pay for these fees, while academics from the biological and medical fields were most likely to know how to meet these payment demands, with many indicating that these could be paid using grant funding. Indeed, all Warlick’s and Vaughan’s [23] biomedical faculty respondents who had paid an article processing charge to publish in an open-access journal did so using part of their grant funding. No doubt for this reason, most respondents did not believe that article processing charges pose a disincentive to publish in open-access journals [23].

Mischo and Schlembach [24] found concern among engineering faculty about the gold open-access model, including things such as the costs to authors, the overall economics of such a system and the potential for conflicts of interest between scholarly rigour and the lure of profit from article processing charges. Indeed, as other researchers have discussed, the publishers of open-access journals financed through article processing charges may have a disincentive to reject articles since they will incur the costs of submission and peer-review processes but not receive any revenue if an article is rejected [25].

In an analysis of article processing charges paid for by 23 universities in the United Kingdom for the period between 2007 and 2014, Pinfield et al. [26] determined that article processing charges were considerably higher for hybrid open-access journals than they were for gold open-access journals. Siler et al. [25] found similar differences in Global Health research, in which article processing charges for hybrid journals were, on average, substantially higher ($3275) than the charges levied by gold open-access journals ($2139).

3. Methods

The data for this study were collected using self-administered web surveys distributed in 2013 and 2018. The 2018 survey instrument was a slightly revised and shortened version of the one used for the 2013 study. For this reason, not all the findings presented in what follows have a comparison 2013 value. The web surveys were created and administered using Qualtrics online survey software.

The request for participation in the study was sent to faculty in all LIS programmes in North America (excluding Puerto Rico) that are accredited by the American Library Association (ALA). The names of all ALA-accredited schools were collected from the ALA’s searchable database. The names and email addresses of all faculty members to be asked to participate in the survey were obtained from the websites of each school. For schools housed in departments or faculties that include cognate disciplines such as communication/media studies or education, only those faculty members who were clearly identified on the school’s website as being associated with the LIS programme were included in the sampling frame. Since most adjunct faculty are not required to publish as part of their position, these faculty members were not included. The sampling frame contained 807 faculty members: 260 assistant professors, 321 associate professors and 226 full professors.

The research protocol was reviewed and approved by the author’s institutional review board. Potential participants were sent an email explaining the purpose and goals of the study and inviting completion of the survey. Those who accepted the invitation were asked to follow the provided URL to access and complete the survey. Participants were informed that no response data would be linked back to their identity during either the analysis or write-up and reporting stages of the project. The survey was open for 3 weeks. A reminder email was sent at the mid-point of this period to those faculty members who had not yet completed the survey. As an additional confidentiality safeguard, this reminder process was automated using Qualtrics survey software.

The survey instrument was pre-trialled and subsequently modified slightly before being distributed. The instrument contained 43 questions, although several of these were composed of multiple sub-questions. Most of the questions employed agree/disagree, Likert-type scale or ranking response categories. Some questions offered participants the opportunity to provide additional details. Respondents were able to skip any question to which they did not wish to respond. A total of 172 surveys were completed, which yields a response rate of just over 21%. The distribution of respondents according to faculty rank is not significantly different from the distribution of faculty ranks found in the broader population (χ²(2, N = 169) = 1.136, p = 0.567). Associate professors are slightly over-represented (44% completed the survey and this rank comprises 40% of the population), and assistant and full professors are very slightly under-represented (30% and 26%, respectively, completed the survey and these ranks comprise, respectively, 32% and 28% of the population).

Since the questions contained in the survey elicited responses on nominal and ordinal scales of measurement, the statistical analyses were conducted using chi-square tests for independence or goodness of fit. All tests were run using SPSS 25 and were considered significant at an alpha level less than 0.05. In order to avoid violating the expected cell count requirement of chi-square tests, several questions were recoded to combine semantically similar response categories. For example, the typical Likert-type scale of ‘strongly agree’, ‘agree’, ‘neither agree nor disagree’, ‘disagree’ and ‘strongly disagree’ was collapsed into ‘agree/strongly agree’, ‘neither agree nor disagree’ and ‘disagree/strongly disagree’. Similarly, the original response categories of ‘very important’, ‘important’, ‘not very important’ and ‘not at all important’ were transformed into ‘very important/important’ and ‘not very important/not at all important’.

4. Findings

4.1. Awareness of open access

Faculty in LIS schools are relatively knowledgeable about the different types of open access (Table 1). Although knowledgeability about open-access journals has increased by 8%, levels of knowledgeability about other modalities of open access have remained stable between 2013 and 2018.² Faculty in LIS schools who indicate that they are knowledgeable or very knowledgeable about open-access journals are more likely to have published in an open-access journal than their colleagues who are not very knowledgeable about this modality of open access (χ²(1, N = 157) = 11.85, p = 0.001, Cramér’s V = 0.28). Survey respondents were also asked for how long they have known about the different types of open access. Sixty-nine percent and 74% of the respondents, respectively, have been aware of institutional repositories and open-access journals for longer than 5 years. However, the proportion of respondents knowledgeable about disciplinary repositories is lower, with only a slight majority (52%) being knowledgeable for longer than 5 years. A similar proportion of respondents (51%) have been knowledgeable about hybrid journals for more than 5 years.

Table 1.

Levels of knowledge about types of open access.

	Very knowledgeable	Knowledgeable	Aware, but do not know much	Not aware	No response	Total
Institutional repositories of open-access content	61	68	37	2	4	172
%Within question total	35.5	39.5	21.5	1.2	2.3	100
Disciplinary repositories of open-access content	36	61	49	21	5	172
%Within question total	20.9	35.5	28.5	12.2	2.9	100
Fully open-access journals (all journal articles freely available without university or individual subscription)	57	89	22	0	4	172
%Within question total	33.1	51.7	12.8	0.0	2.3	100
Hybrid journals (authors can pay a fee to make their article open access in an otherwise subscription-based journal)	45	74	39	11	3	172
%Within question total	26.2	43.0	22.7	6.4	1.7	100
Blogs/wikis	57	78	30	3	4	172
%Within question total	33.1	45.3	17.4	1.7	2.3	100

4.2. Engagement with academic publishing

A substantial number of respondents in LIS schools (65%, up from 62% in 2013) agree that all scholarly articles should be free for everyone to access online. Sixty-six percent of the participants think that the dissemination of research is a common good that should not be monetised in any way (62% in 2013) and 78% agree to some extent that LIS scholars should be at the forefront of efforts to expand open access to research (72% in 2013). Despite such beliefs about access to academic research, and similar to findings from several of the studies reviewed earlier, only a small majority (54%) of the respondents consider open access to be an important criterion when deciding where to publish their work (Table 2). Instead, faculty in LIS schools are motivated by more traditional factors when deciding in which journals to publish their work, including the following in the order of importance: journal publisher’s reputation (97% in 2013), journal prestige, speed of publication (85% in 2013), weight of the publication venue in tenure and promotion decisions (81% in 2013) and impact factor (80% in 2013). When deciding where to submit their work for publication, those respondents who have published in an open-access journal are more likely than those who have not to consider important the ability to deposit the pre-publication version of their article into an electronic repository (χ²(3, N = 158) = 9.99, p = 0.019, Cramér’s V = 0.25) or that the journal is open access (χ²(3, N = 156) = 23.36, p < 0.001, Cramér’s V = 0.39).

Table 2.

Factors considered when selecting publication venues.

	Very important	Important	Not very important	Not at all important	No response	Total
Journal’s or book publisher’s reputation	106	58	5	0	3	172
%Within question total	61.6	33.7	2.9	0.0	1.7	100
Journal’s impact factor	51	73	33	12	3	172
%Within question total	29.7	42.4	19.2	7.0	1.7	100
Prestige of journal	76	74	19	0	3	172
%Within question total	44.2	43.0	11.0	0.0	1.7	100
Publication venue’s weight in tenure and promotion considerations	68	67	26	8	3	172
%Within question total	39.5	39.0	15.1	4.7	1.7	100
My ability to retain copyright in my work	29	65	62	13	3	172
%Within question total	16.9	37.8	36.0	7.6	1.7	100
My ability to deposit the pre-publication version of my work into an electronic repository	52	54	45	18	3	172
%Within question total	30.2	31.4	26.2	10.5	1.7	100
My ability to deposit the published version of my work into an electronic repository	31	77	45	16	3	172
%Within question total	18.0	44.8	26.2	9.3	1.7	100
Publication venue is open access	21	71	63	12	5	172
%Within question total	12.2	41.3	36.6	7.0	2.9	100
Speed of publication	43	94	26	4	5	172
%Within question total	25.0	54.7	15.1	2.3	2.9	100

While 81% of the respondents had submitted a manuscript or had an article published in a subscription-based journal in the 12 months prior to the survey (80% in 2013), slightly less than half had done the same in an open-access journal (49%, up from 37% in 2013). However, considered from a longer-term perspective 70% of the respondents have published at least once in an open-access journal, which is significantly higher than in 2013 when just over half (53%) had done so (χ²(1, N = 416) = 14.24, p < 0.001). Among those respondents who have published in an open-access journal, the following reasons were articulated as motivations for doing so: free access for readers (54%), increased visibility for the author’s work (49%), concerns about high prices for subscription-based journals (32%), speed of publication (31%), the ability to retain copyright (22%) and antipathy towards traditional publishers (17%).³ Respondents were also offered the opportunity to write in any additional motivations and the most repeated reason by far, and not surprisingly, was appropriateness of the venue to the subject matter of the research. In several of these cases, the open-access status of the journal was a welcome coincidence but not a driving criterion. A few respondents indicated that the open-access journal in which they published was a top one in their field of research.

A majority (63%, up from 57% in 2013) of the respondents claim to be likely or very likely to publish at least one article in an open-access journal within a year of the survey, while 22% are not very likely to do so and 1% will not (29% and 5%, respectively, in 2013). This difference between 2013 and 2018 is statistically significant (χ²(4, N = 439) = 10.15, p = 0.038). Those respondents who are more knowledgeable about open-access journals are more likely than expected to plan on publishing at least one article in an open-access journal within a year following the survey, while those less knowledgeable are more likely than expected not to do so (χ²(2, N = 166) = 18.88, p < 0.001, Cramér’s V = 0.34). Experience with open-access publishing is also strongly associated with the proclaimed likelihood of faculty in LIS schools publishing at least one article in an open-access journal within the next year (χ²(2, N = 158) = 33.79, p < 0.001, Cramér’s V = 0.46). Whereas 76% of those respondents who have already published an open-access article are likely or very likely to publish in an open-access journal within a year (81% in 2013), this proportion falls to 29% among those who have not already published in an open-access journal (27% in 2013).

4.3. Assessments and perceptions of open-access publishing

With the exception of two, most of the assessments and perceptions of open-access journals have not changed significantly since 2013 (Table 3). First, the proportion of respondents who attribute faster publication speeds to open-access journals compared with subscription-based journals has decreased significantly from 61% in 2013 to 43%, while almost half are uncertain (30% in 2013). In 2013, respondents were more likely than expected to agree to some extent and less likely than expected to neither agree nor disagree that open-access journals are published faster than traditional journals. 2018 respondents are less likely than expected to agree and more likely than expected to be uncertain (χ²(2, N = 435) = 16.68, p < 0.001). This finding raises the question of whether subscription-based journals have sped up publication times or have publishing cycles in open-access journals become slower. Second, the number of respondents who would welcome more open-access journals in their field of research has increased significantly from 67% in 2013 to 79% in 2018 (χ²(2, N = 436) = 11.07, p = 0.004). In 2013, respondents were less likely than expected to agree and more likely than expected to neither agree nor disagree about welcoming more open-access journals in their field of research. By 2018 the situation had reversed. This finding suggests that there remains a lot of opportunity to expand open-access publishing in LIS.

Table 3.

Perceived advantages and disadvantages of open-access publishing.

	Strongly agree	Agree	Neither agree nor disagree	Disagree	Strongly disagree	No response	Total
Open access offers wider circulation of research than publication in subscription-based journals	69	62	27	9	1	4	172
%Within question total	40.1	36.0	15.7	5.2	0.6	2.3	100
Open-access journals have a larger readership by researchers than subscription-based journals	31	34	78	22	4	3	172
%Within question total	18.0	19.8	45.3	12.8	2.3	1.7	100
Open-access journals have faster publication timeframes than subscription-based journals	27	47	84	10	0	4	172
%Within question total	15.7	27.3	48.8	5.8	0	2.3	100
I would welcome more open-access journals in my field of research	68	67	22	9	2	4	172
%Within question total	39.5	39.0	12.8	5.2	1.2	2.3	100
Articles in open-access journals are cited less frequently than articles in subscription-based journals	10	22	79	42	15	4	172
%Within question total	5.8	12.8	45.9	24.4	8.7	2.3	100
Open-access journals have less scholarly rigour than subscription-based journals	9	19	67	51	23	3	172
%Within question total	5.2	11.0	39.0	29.7	13.4	1.7	100
Open-access journals have lower production standards (e.g. copyediting, typesetting) than subscription-based journals	9	32	64	42	21	4	172
%Within question total	5.2	18.6	37.2	24.4	12.2	2.3	100
Publishing my work in open-access journals may adversely affect my tenure and/or promotion prospects	12	33	46	50	27	4	172
%Within question total	7.0	19.2	26.7	29.1	15.7	2.3	100
Publishing my work in open-access journals may adversely affect my chances of winning research grants	6	15	61	61	26	3	172
%Within question total	3.5	8.7	35.5	35.5	15.1	1.7	100
Publishing my work in open-access journals may limit the impact of my published work	8	20	51	56	34	3	172
%Within question total	4.7	11.6	29.7	32.6	19.8	1.7	100
There are no fundamental benefits to open-access publication	2	7	15	74	72	2	172
%Within question total	1.2	4.1	8.7	43.0	41.9	1.2	100

Those respondents who have published in an open-access journal are more likely than those who have not to agree to some extent that open access offers wider circulation of research than subscription-based journals (χ²(2, N = 157) = 15.81, p < 0.001, Cramér’s V = 0.32) and that open-access journals have a larger readership by researchers (χ²(2, N = 158) = 14.46, p = 0.001, Cramér’s V = 0.30). Although not statistically significant, there is a substantial amount of uncertainty among both those who have published open access (48%) and those who have not (40%) about whether open-access journals have a larger readership.

Conversely, survey participants who have published in an open-access journal are more likely than those who have not to disagree to some extent that open-access articles are cited less frequently than articles in subscription-based journals (χ²(2, N = 157) = 7.94, p = 0.019, Cramér’s V = 0.23). In this case, the substantial uncertainty about citation disadvantages for open-access articles among those who have published in an open-access journal (41%) and those who have not (65%) is statistically significant. Those respondents who have published in an open-access venue are also more likely than those who have not to disagree that publishing their work in open-access journals may limit the impact of their research (χ²(2, N = 158) = 19.55, p < 0.001, Cramér’s V = 0.35). Open-access authors are similarly more likely than non-open-access authors to evaluate a publication in a peer-reviewed open-access journal as being of comparable or better quality than an article in a traditional subscription-based journal (χ²(3, N = 159) = 8.80, p = 0.032, Cramér’s V = 0.24).

Those respondents who claim to be knowledgeable about institutional repositories (χ²(2, N = 166) = 8.41, p = 0.015, Cramér’s V = 0.23) or disciplinary repositories of open-access content (χ²(2, N = 165) = 7.06, p = 0.029, Cramér’s V = 0.21) are more likely than their less knowledgeable colleagues to agree to some extent that open access offers wider circulation of research than subscription-based journals. Knowledgeability about institutional repositories (χ²(2, N = 166) = 9.63, p = 0.008, Cramér’s V = 0.24), disciplinary repositories (χ²(2, N = 165) = 17.19, p < 0.001, Cramér’s V = 0.32), and open-access journals (χ²(2, N = 166) = 7.71, p = 0.021, Cramér’s V = 0.22) is also positively associated with the belief that open-access journals have a larger readership by researchers than subscription-based journals. Here too, however, uncertainty about this perceived benefit of open access is high and it contributes significantly to the relationships with knowledgeability about institutional repositories (41% of the knowledgeable respondents and 62% of those not knowledgeable) and disciplinary repositories (36% of the knowledgeable respondents and 59% of those not knowledgeable).

On the other hand, survey participants who claim to be more knowledgeable about open-access journals (χ²(2, N = 166) = 10.00, p = 0.007, Cramér’s V = 0.25) are more likely than those who are less knowledgeable to disagree to some extent that open-access articles are cited less frequently than articles in subscription-based journals. Here too the high levels of uncertainty about citation disadvantages for open-access articles contribute significantly to the relationship (44% among those knowledgeable and 73% among those less knowledgeable about open-access journals). Survey participants more knowledgeable about open-access journals are more likely than those less knowledgeable to disagree that publishing their work in such journals might limit the impact of their published work (χ²(2, N = 167) = 12.65, p = 0.002, Cramér’s V = 0.28). Analysis of the adjusted residuals also indicates that the relatively high levels of uncertainty among respondents about this possible impact of publishing in open-access journals contribute significantly to this relationship (26% among those knowledgeable and 55% among those less knowledgeable about open-access journals).

4.4. Open access and faculty career considerations

Although half of the respondents consider a publication in a peer-reviewed open-access journal to be of quality comparable with a publication in a traditional journal, fewer believe that members of tenure and promotion committees are as convinced about the quality of open-access journals (Table 4). Beliefs about peer assessments remain important because a good part of an academic’s success as a researcher depends critically on how that research is assessed by disciplinary peers. There is a statistically significant difference between 2013 and 2018 respondent perceptions about the quality of open-access publications (χ²(3, N = 438) = 14.31, p = 0.003). Respondents in 2013 are more likely than expected to evaluate open-access publications as being of lesser quality than a publication in a traditional peer-reviewed journal and less likely than expected to be unsure, while the reverse is true for 2018 survey participants. Those respondents who are more knowledgeable about open-access journals are less likely than expected to evaluate a publication in an open-access journal unfavourably compared with an article in a traditional subscription-based journal while those less knowledgeable are more likely than expected to do so (χ²(3, N = 167) = 10.85, p = 0.013, Cramér’s V = 0.26). Open-access authors are more likely than non-open-access authors to believe that members of a tenure and promotion committee at their institution would judge articles in open-access journals as being comparable in quality with articles published in traditional journals (χ²(3, N = 157) = 7.77, p = 0.05, Cramér’s V = 0.22).

Table 4.

Respondent assessments of the quality of publications in a peer-reviewed open-access journal versus a peer-reviewed subscription-based journal, and respondent beliefs about how members of a tenure and/or promotion committee would evaluate peer-reviewed open-access publications versus articles in peer-reviewed subscription-based journals.

	Respondents’ own assessments	Respondents’ beliefs about colleague assessments
Open access very unfavourably	2	10
%Within question total	1.2	5.8
Open access of somewhat lesser quality	21	48
%Within question total	12.2	27.9
Open access of comparable quality	86	66
%Within question total	50.0	38.4
Open access of somewhat better quality	2	2
%Within question total	1.2	1.2
Open access very favourably	21	6
%Within question total	12.2	3.5
Unsure	38	36
%Within question total	22.1	20.9
No response	2	4
%Within question total	1.2	2.3

Beliefs about the career impacts from publishing in open-access journals are related to several other variables. Those respondents who have published in an open-access publication are more likely than those who have not to disagree that publishing their work in open-access journals may adversely affect their tenure and/or promotion prospects (χ²(2, N = 157) = 7.97, p = 0.019, Cramér’s V = 0.23). Survey participants more knowledgeable about open-access journals are more likely than those less knowledgeable to disagree that publishing their work in such journals might adversely affect their tenure and/or promotion prospects (χ²(2, N = 166) = 7.38, p = 0.025, Cramér’s V = 0.21) or their chances of winning research grants (χ²(2, N = 167) = 8.12, p = 0.017, Cramér’s V = 0.22). Analysis of the adjusted residuals reveals that the relatively high levels of uncertainty among respondents about these possible impacts of publishing in open-access journals contribute significantly to both these relationships. Twenty-four percent of the respondents knowledgeable about such publishing venues and 50% of those not knowledgeable are uncertain about possible negative impact on tenure and/or promotion chances, while 33% of the knowledgeable respondents and 59% of those not knowledgeable are unsure about the impact of open-access publishing on their chances of securing research grants. Similarly, those respondents more favourable in their assessments of open-access publications are more likely than those who would judge an open-access publication unfavourably to disagree that publishing their work in an open-access journal may adversely impact their tenure and/or promotion applications (χ²(6, N = 168) = 27.05, p < 0.001, Cramér’s V = 0.28) or chances of winning research grants (χ²(6, N = 169) = 38.43, p < 0.001, Cramér’s V = 0.34). Put another way, respondents who have experience with or are knowledgeable about open-access publishing or who believe open-access publications are of quality comparable with publications in traditional journals tend to be less concerned that publishing their work in open-access journals may negatively affect their career or research funding prospects.

4.5. Financing open-access publishing

Thirty-eight percent of the respondents (down from almost half in 2013) believe that neither they nor their institutions should have to pay any fees to publish their research in an open-access journal, while 22% (similar proportion in 2013) are unsure about what would be an appropriate amount for such fees. Respondents in 2018 are less likely than expected to think that there should not be any fees to publish their research and are more likely than expected to accept article processing charges between $500 and $1500, although, while contributing to the difference found, the willingness to accept fees over $500 is limited to a small minority (5% in 2013 and 17% in 2018). The likelihood of accepting these levels of fees is reversed for 2013 respondents (χ²(5, N = 432) = 19.8, p = 0.001).

Compared with their colleagues in other disciplines, as outlined in the literature review, faculty in LIS schools have less experience with paying an article processing charge to publish in an open-access journal, although the number of respondents who have paid such fees has increased significantly from 8% in 2013 to 18% in 2018 (χ²(1, N = 265) = 16.85, p < 0.001). The article processing charges 2018 authors paid range from $150 to $3000. When asked about hybrid open-access models, in which authors pay a fee to have their article made freely accessible to readers in an otherwise subscription-based journal, only 17% of the respondents would definitively remit such fees to a publisher (up from 5% in 2013). Twenty-nine percent might possibly pay such fees (down from 33% in 2013), while 26% and 19%, respectively, would probably or definitely not (these proportions were 37% and 22% in 2013). Respondents in 2018 are more likely than expected to pay to publish in hybrid journals and less likely than expected to probably refuse to pay such fees, while the reverse is the case for 2013 respondents (χ²(4, N = 431) = 38.88, p < 0.001).

4.6. Electronic repositories (green open access)

Recent deposition of scholarly output into electronic repositories (i.e. green open access) remains limited to a minority of faculty in LIS schools, with only 47% and 33%, respectively, having deposited an article or other research output, such as working papers or technical reports, over the 12 months prior to the survey (35% and 24%, respectively, in 2013). However, the proportion of deposition is higher when considered over a longer time frame; 70% of the respondents indicate that they had deposited at least one research output into an electronic repository at one time or another, which is significantly higher than the 50% of the survey participants who had done so in 2013 (χ²(1, N = 432) = 15.68, p < 0.001). As might intuitively be expected, respondents who have published their work in an open-access journal (83%, up from 63% in 2013) are significantly more likely than their colleagues who have not published open access (17%, down from 36% in 2013) to have deposited any of their research outputs into an electronic repository (χ²(1, N = 159) = 13.45, p < 0.001, Cramér’s V = 0.29).

Among those who had posted their work to an electronic repository, the following proportions of respondents attributed some degree of importance to these reasons: the potential for electronic repositories to broaden the dissemination of research more generally (94%, up minimally from 93% in 2013), expands exposure of both previously published (92%, up from 88% in 2013) and not previously published (80%, up from 73% in 2013) work, increases an academic institution’s leverage with commercial publishers (52%, up from 44% in 2013), improves the prospects for tenure and/or promotion (45%, up from 41% in 2013) and mandated by institution (52%, up from 24% in 2013, which is a statistically significant difference (χ²(3, N = 166) = 18.77, p < 0.001)).

Sixty-two percent of the respondents indicate that their institution had developed an open-access publishing initiative within the past 5 years, which is a statistically significant difference from 2013 when 45% of the respondents worked at institutions that had done so (χ²(1, N = 424) = 8.59, p = 0.003). There is a positive, albeit weak, relationship between institutional open-access publishing initiatives and likelihood of having published in an open-access journal (χ²(1, N = 160) = 5.88, p = 0.015, Cramér’s V = 0.19). Those respondents working at universities that have developed open-access initiatives are also more likely to have deposited at least one research output into an electronic repository (χ²(1, N = 167) = 16.01, p < 0.001, Cramér’s V = 0.31). Similarly, those survey participants working at institutions with open-access initiatives are more likely than those at universities without such initiatives to be knowledgeable about institutional repositories (χ²(1, N = 166) = 5.47, p = 0.019, Cramér’s V = 0.18).

Unlike in 2013 when faculty in LIS schools exhibited a much greater readiness to comply with green rather than gold open-access mandates, willing compliance is fairly similar for both models of open access (Tables 5 and 6). There is a statistically significant difference between 2013 and 2018 levels of willingness to comply with mandates to publish in open-access journals, with respondents in 2018 more likely than expected to willingly accept such a requirement and 2013 respondents more likely than expected to unwillingly accept or outright refuse a gold open-access directive (χ²(3, N = 435) = 57.84, p < 0.001).

Table 5.

Willingness to comply with research funding mandates to publish research results through open access (gold model).

Gold open access	Count	%Within question
I would willingly accept such terms	133	77.3
I would accept such terms, but unwillingly	19	11.0
I would not accept such terms and would look elsewhere for funding	5	2.9
Unsure	10	5.8
No response	5	2.9
Total	172	100

Table 6.

Willingness to comply with employer or funder mandates to deposit copies of publications in electronic repositories (green model).

Green Open Access	Count	%Within question
I would do so willingly	142	82.6
I would do so, but unwillingly	12	7.0
I would not be prepared to do so	3	1.7
Unsure	5	2.9
No response	10	5.8
Total	172	100

5. Discussion

Unsurprisingly given the foundational tenets of the LIS discipline and profession, almost two-thirds of faculty in LIS schools believe that research is a common good and are very supportive of free online access to scholarly research. Nonetheless, when faculty in LIS schools make publishing decisions the open-access status of a journal trails substantially other more traditional considerations such as journal publisher’s reputation, journal prestige, speed of publication, weight of the publication venue in tenure and promotion decisions and impact factor. These findings comport with those found by previous researchers across a range of disciplines and underscore the fact that career considerations tend to outweigh mode of delivery as a determining factor for academics when selecting journals in which to publish.

One of the biggest overall comparative findings is that awareness and assessments of open access have remained relatively stable over the 5 years between 2013 and 2018, although engagement with open access has increased significantly. Somewhat unexpectedly, and in contrast to the 2013 survey results, comparatively fewer relationships were detected in the 2018 data when analysing whether attitudes towards and engagement with open access differ across respondents based on faculty rank, perceptions and assessments of open-access publications, beliefs about how tenure and promotion committees would evaluate open-access publications, experience publishing in open-access journals, knowledgeability about open access, and likelihood of publishing an article in an open-access journal within a year of the survey. The two main variables found to be related to others are experience with open-access publishing and knowledgeability about open access. As might intuitively be expected, those respondents who have published in an open-access journal are more likely to be more knowledgeable about such publishing venues than those faculty in LIS schools without such publishing experience.

Among those respondents who have published in an open-access journal, the two most agreed upon motivations for doing so are free access for readers and increased visibility for research. These two considerations are clearly related as the former contributes to the latter and they indicate that faculty in LIS schools take seriously the foundational importance attributed to access while remaining cognisant of the research imperatives in the academy. A prominent additional reason a number of respondents articulated for publishing in an open-access journal is fit with scope of the journal. The importance attributed to these considerations, coupled with the large number of respondents who would welcome more open-access journals in their field of research, reinforces the proposition made earlier that open-access publishing in LIS remains underdeveloped. Similar to some of the motivations to publish in open-access journals, contributing to the broader dissemination of research and increasing exposure of work both previously published and unpublished are rated most highly by survey participants as reasons for depositing their research into electronic repositories. Although notably low among respondents is use of electronic subject repositories, which points to the need to create more and publicise more vigorously LIS-specific electronic repositories.

Those who have published an article in an open-access journal tend to assess publications in such venues more favourably than those without such publishing experience, are more predisposed to believe that tenure and/or promotion committees would evaluate such publications favourably and are more likely to publish again in an open-access journal within a year. Those faculty in LIS schools who have engaged with open-access publishing are also less likely than their colleagues who have not done so to believe that such publications will adversely affect their career opportunities, limit the impact of their work or be cited less frequently than articles in a traditional subscription-based journal. Similarly, those respondents with open-access publishing experience are more likely than those without such experience to believe that open-access journals offer wider circulation and larger readership by researchers than subscription-based journals. These same relationships were found for the variable knowledgeability about the various types of open access. As might also be expected, those respondents who assess open-access articles as being comparable in quality with toll-access articles express less concern about the impact of open-access publishing on their prospects for achieving tenure and/or promotion and winning research grants.

A majority of respondents believe that either there should not be any fees to publish in open-access journals (38%) or that fees should not exceed $500 (22%). That having been said, it is worth reiterating that willingness to pay article processing charges has increased significantly, with respondents in 2018 less likely than expected to reject fees. Although survey participants are almost evenly split about whether they would or would not pay an article processing charge to publish in a hybrid open-access journal, respondents in 2018 are significantly more likely than expected to accept such fees, which is a reversal from 2013. This finding gives some cause for concern since the hybrid model provides publishers a mechanism to collect double revenue streams from authors through article processing charges and from readers (or their proxies in the form of libraries) through subscriptions – what has become known colloquially as ‘double-dipping’. As such, this model of publishing is particularly threatening to the long-term financial sustainability of the scholarly communication system.

Although there has clearly been an uptake in engagement with open-access publishing in the previous 5 years, and similar to results from the 2013 iteration of this survey, there remains a fair amount of uncertainty among faculty in LIS schools when it comes to their perceptions of open-access journals. Thus, while in general it is clear that experience with and knowledgeability about open access is associated with a number of the perceptions about open-access journals, faculty in LIS schools remain somewhat equivocal about open-access publishing. Although here too, in most of the relationships found experience with and knowledgeability about open-access publishing are associated with less uncertainty.

6. Conclusion

As a follow-up to the survey conducted in 2013 about LIS faculty experience with and attitudes towards open-access publishing, this study sought to both map current perceptions of and engagement with open access and to track any changes in these that may have occurred in the 5 years between the two studies. The results from this study help fill a gap in the extant literature about open access specific to faculty in LIS schools. Comparisons between the two data sets reveal that engagement with open-access publishing has increased among faculty in LIS schools between 2013 and 2018. Some of the concerns that have historically plagued open-access journals are dissipating and several of the benefits typically attributed to open-access journals have increased marginally between 2013 and 2018, although there remains a substantial amount of uncertainty among faculty in LIS schools with regard to perceptions and assessments of open-access publishing. That having been said, those who have actually published in an open-access journal or are more knowledgeable about open access tend to be less apprehensive and uncertain about such concerns and are more likely to publish in an open-access venue within a year. Thus, the long-term sustainability of this model of scholarly communication would seem to be dependent, in part at least, on developing strategies to induce non-open-access authors to engage with it. One such strategy might be open-access mandates, which appear to be acceptable to an overwhelming majority of respondents. Indeed, this suggestion assumes added weight when considered in concert with the relationship found between open-access publishing experience and institutional open-access initiatives. Given that an overwhelming majority of respondents agree to some extent that LIS scholars should be at the forefront of efforts to expand open access to research, faculty in LIS schools not already subject to institutional mandates may want to assume more of a leadership role in the academy and commit to their own open-access mandates.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Wilhelm Peekhaus

Notes

References

Campbell

Introduction: overview of academic and professional publishing. In: Campbell

Pentz

Borthwick

(eds) Academic and professional publishing. Oxford: Chandos, 2012, pp. 1–14.

Budapest Open Access Initiative. Budapest: Budapest Open Access Initiative, 2002, https://www.budapestopenaccessinitiative.org/read

Peekhaus

Proferes

How library and information science faculty perceive and engage with open access. J Inf Sci 2015; 41: 640–661.

Peekhaus

Proferes

An examination of North American Library and Information studies faculty perceptions of and experience with open-access scholarly publishing. Libr Inform Sci Res 2016; 38: 18–29.

Björk B-C. Open access: are the barriers to change receding? Publications 2013; 1(1): 5–15.

Mercer

Almost halfway there: an analysis of the open access behaviors of academic librarians. Coll Res Libr 2011; 72: 443–453.

Grandbois

Beheshti

A bibliometric study of scholarly articles published by library and information science authors about open access. Inform Res 2014; 19(4): 1–23.

Campbell-Meier

Faculty adoption of open access publishing in the University System of Georgia. J Technol Knowl Shar 2015; 11(1): 1–10.

Gargouri

Lariviere

Gingras

, et al. Green and gold open access percentages and growth, by discipline. In: Proceedings of the international conference on science and technology indicators, Montreal, QC, Canada, 5–8 September 2012.

10.

Chang

Y-W.

Comparative study of characteristics of authors between open access and non-open access journals in library and information science. Libr Inform Sci Res 2017; 39(1): 8–15.

11.

Teplitzky

Phillips

Evaluating the impact of open access at Berkeley: results from the 2015 survey of Berkeley Research Impact Initiative (BRII) funding recipients. Coll Res Libr 2016; 77(5): 568–581.

12.

Wolff-Eisenberg

Rod

Schonfeld

RC.

Ithaka S+R US faculty survey 2015. New York: Ithaka S+R, 2016.

13.

Solomon

Björk

B-C.

Publication fees in open access publishing: sources of funding and factors influencing choice of journal. J Assoc Inf Sci Tech 2012; 63(1): 98–107.

14.

Nelson

Eggett

DL.

Citations, mandates, and money: author motivations to publish in chemistry hybrid open access journals. J Assoc Inf Sci Tech 2017; 68(10): 2501–2510.

15.

Gaulé

Maystre

Getting cited: does open access help?

Res Policy 2011; 40(10): 1332–1338.

16.

Frass

Cross

Gardner

Taylor & Francis open access survey. Oxford: Taylor & Francis, 2014.

17.

Rowley

Johnson

Sbaffi

, et al. Academics’ behaviors and attitudes towards open access publishing in scholarly journals. J Assoc Inf Sci Tech 2017; 68(5): 1201–1211.

18.

Xia

A longitudinal study of scholars [sic] attitudes and behaviors toward open-access journal publishing. J Am Soc Inf Sci Tec 2010; 61: 615–624.

19.

Dalton

A dissemination divide? The factors that influence the journal selection decision of Library & Information Studies (LIS) researchers and practitioners. Lib Info R 2013; 37(115): 33–57.

20.

Moore

Survey of University of Toronto faculty awareness, attitudes, and practices regarding scholarly communication: a preliminary report. Toronto, ON, Canada: University of Toronto, 2011.

21.

Aulisio

Open access publishing and social justice: Scranton’s perspectives. Jesuit High Educ 2014; 3(2): 55–73.

22.

Creaser

Open access to research outputs – institutional policies and researchers’ views: results from two complementary surveys. New Rev Acad Librarianship 2010; 16: 4–25.

23.

Warlick

Vaughan

Factors influencing publication choice: why faculty choose open access. Biomed Digital Librar 2007; 4(1), http://www.bio-diglib.com/content/pdf/1742-5581-4-1.pdf

24.

Mischo

Schlembach

MC.

Open access issues and engineering faculty attitudes and practices. J Library Administr 2011; 51: 432–454.

25.

Siler

Haustein

Smith

, et al. Authorial and institutional stratification in open access publishing: the case of global health research. PeerJ 2018; 6: e4269.

26.

Pinfield

Salter

Bath

PA.

The ‘total cost of publication’ in a hybrid open-access environment: institutional approaches to funding journal article-processing charges in combination with subscriptions. J Assoc Inf Sci Tech 2016; 67(7): 1751–1766.