Evidence Summary

Some LIS Faculty Indicate Reservations about Open Access

A Review of:

Peekhaus, W., & Proferes, N. (2015). How library and information science faculty perceive and engage with open access. Journal of Information Science, 41(5), 640-661. http://dx.doi.org/10.1177/0165551515587855

Reviewed by:

Richard Hayman

Assistant Professor & Digital Initiatives Librarian

Mount Royal University

Calgary, Alberta, Canada

Email: rhayman@mtroyal.ca

Received: 29 May 2016 Accepted: 6 Aug. 2016

2016 Hayman. This is an Open Access article distributed under the terms of the Creative Commons‐Attribution‐Noncommercial‐Share Alike License 4.0 International (http://creativecommons.org/licenses/by-nc-sa/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly attributed, not used for commercial purposes, and, if transformed, the resulting work is redistributed under the same or similar license to this one.

Abstract

Objective – To examine the awareness of, attitudes toward, and engagement with open access (OA) publishing, based on rank and tenure status among library and information science (LIS) faculty in North America.

Design – Web-based survey distributed via email.

Setting – Accredited library and information science (LIS) programs in North America.

Subjects – 276 professors and professors emeriti.

Methods – Researchers collected email addresses for 1,017 tenure-track, tenured, and emeriti professors from the public websites of the LIS programs. Researchers sent an email invitation to participate in the survey by accessing a URL, with the survey itself delivered using Qualtrics software. The survey included 51 total questions, some with additional sub-questions, and most items used Likert-type rating scale. The researchers analysed the data using SPSS software, and indicated using chi-square tests to measure significance, with a stated intent to get beyond the descriptive statistics commonly seen in other publications.

Main Results – This study’s results draw on 276 completed responses, for a response rate of 27%. Researchers reported that 53% of respondents had some experience with publishing in a peer-reviewed OA format. When asked whether they agreed that scholarly articles should be free to access for everyone, pre-tenure assistant professors were most likely to agree (74%), followed by tenured associate professors (62%), full professors (59%) and then emeriti professors (8%). However, they found less likelihood that associate professors would have actually published in an OA format, highlighting a “disconnect between beliefs about accessibility of research and actual practice with open access” (p. 646). Researchers also discovered a connection between faculty awareness of institutional and disciplinary repositories and faculty publishing in OA journals, though a relatively low number (35%) had deposited their output in a repository within the previous year. That increases to 50% of respondents when timeframe is ignored.

Faculty who had never published in OA journals ranked several barriers to doing so, barriers common across disciplinary boundaries. These include objections to paying OA fees; perceptions of slow time to publish, low research impact, and venue prestige when compared to traditional subscription journals; an inability to identify an appropriate OA journal; and an inability to pay OA fees. However, the researchers note that a majority of these respondents who had never published in an OA format would do so if these barriers were removed. Those participants who had some previous experience with OA were more likely to have positive perceptions of OA journal quality and impact, as well as the overall publishing experience, as compared to publishing in traditional journals.

As in other disciplines, LIS faculty are conscious of the connection between OA and tenure and promotion processes. For example, this study reveals that non-tenured faculty are more likely to agree that publishing in OA venues may affect their career progress. Researchers report uncertainty about OA even among tenured LIS faculty. Of all respondents, only 34% agreed that a tenure or promotion committee might consider an OA publication on par with a traditional publication, while 44% of respondents were of the opinion that an OA publication would be treated less favourably than a traditional journal. A mere 1% of respondents believed that an OA publication would be treated more favourably within the tenure and promotion process. Despite this unfavourable perception of OA, the researchers report that 38% of respondents planned to publish in an OA journal regardless of whether their tenure and promotion committees might treat that OA publication unfavourably.

Conclusion – The researchers report a connection between publishing in an OA journal and academic rank, with full professors more likely to publish OA or to have previous experience in publishing in an OA journal as compared to assistant professor colleagues, who perceive publishing in OA as a potential impediment to career progress. The researchers note that there is significant opportunity for LIS faculty involved in tenure and promotion committees to consider and clarify how OA publications are treated, and the impact of OA publishing with regard to career progress. Moreover, given the levels of uncertainty and equivocacy among faculty respondents as a whole regarding certain aspects of OA, the perceptions around quality and rigour, there is room for further research into LIS professors’ perceptions and attitudes toward open access, and how these change over time.

Commentary

As a descriptive analysis, this study adds new knowledge to the conversation about OA engagement. By including their survey instrument as an appendix, alongside their coding key for collapsing certain Likert-like response categories, the authors have contributed a new tool for measuring OA engagement by faculty that can be further adapted for future research. The tool could be improved by providing justification or rationale for collapsing the Likert scale in the manner as done in this study. To aid instrument validity (Glynn, 2006), the authors have pre-trialled and adjusted their survey tool before distribution to participants, and the tool itself was adapted from an instrument used in previous research. However, the researchers do not describe if or how the instrument may have been tested for reliability or internal consistency (e.g., Cronbach’s alpha).

The researchers state their intent to move beyond descriptive statistics toward inferential analysis, but their design and analysis may be problematic. They acknowledge using a non-random (i.e., non-probability) sample, but the authors do not define their intended sampling approach except to state they surveyed “all North American faculty members (excluding Puerto Rico)” and excluding adjunct professors (p. 644). While they discuss the representativeness of participant characteristics within the response rate and their “overall sample” (p. 647), it is unclear whether they intended to achieve a census given the population of LIS faculty invited to participate, and it appears that this research instead draws on a non-probability sample (e.g., convenience sampling). Ultimately the researchers do not clearly identify the survey sampling method, and identifying the specific type of probability or non-probability sample used would have been valuable.

Moving into results analysis, while there is still debate in the literature regarding best practices, it is generally understood that inferential analysis requires a random sample to ensure that the population being studied is properly represented (López, X., Valenzuela, J., Nussbaum, M., & Tsai, 2015). Further, though the researchers indicate that the distribution of participants from the subgroups of assistant, associate, full professors matches “fairly closely” to their distribution among the larger population, there are discrepancies here that require more detail. As Lopez et al note, “if the conclusions of the study involve generalising for subgroups, then the sample size should be representative at the subgroup level”, and also that confidence intervals for calculating these should be explicitly stated (p. 107). Given the low response rate, and the lack of probability sampling, this study is open to self-selection sampling bias and should not be treated as generalizable. Thus, while the descriptive analysis provided is interesting, the inferential results are problematic, leading this author to focus solely on the implications of the descriptive rather than the inferential results.

Takeaways from this study highlight that LIS faculty who participated in the survey reflect similar attitudes toward and engagement with open access as their faculty colleagues from other disciplines. This is important as these attitudes and perceived constraints may affect uptake of OA within the LIS discipline and beyond. The participants from ALA-accredited graduate programs in North America are the educators responsible for instructing new librarians on the fundamental principles and practices of information access and availability. If these educators have reservations about the benefits, challenges, and impacts of open access, this could affect those responsible for encouraging OA practice for colleagues both within and beyond LIS. This suggests that ongoing study of researchers’ attitudes and perceptions toward open access, both within and beyond the discipline, is necessary.

References

Glynn, L. (2006). A critical appraisal tool for library and information research. Library Hi Tech, 24(3), 387-399. http://dx.doi.org/10.1108/07378830610692154

López, X., Valenzuela, J., Nussbaum, M., & Tsai, C. (2015). Some recommendations for the reporting of quantitative studies. Computers & Education, 91, 106-110. http://dx.doi.org/10.1016/j.compedu.2015.09.010