Table 1

Factors Examined in this Study (Hoffmann et al., 2014)

Individual Attributes

Peers and Community

Institutional Structures

and Supports

Demographics

Education and Experience

Intrinsic Motivations

Personal Commitment to Research

Personality Traits

Collaboration

Community

Mentoring

Peer Support

Extrinsic Motivations

Institutional Supports

Methods

This quantitative study used an online survey for data collection. The online survey was based on previously published research that identified potential factors that may contribute to librarians’ research productivity. The knowledge resulting from this first phase of the research was conducted via content analysis, and the results are described in a previous publication (Hoffmann et al., 2014). We considered the survey instruments used in those studies found via the content analysis to inform the development of survey questions. Building from the results of the first phase of the research, survey questions captured participants’ research outputs and explored factors related to three categories: Individual Attributes (including demographics), Peers and Community, and Institutional Structures and Supports.

In developing the survey, we made several key decisions to address the research questions. As noted above, we wanted to draw on previous research, identified in the first phase of our project, to give the survey a solid foundation in the existing literature on research productivity. We wanted to determine relationships between factors and research productivity outputs, rather than simply describing participants’ research environments, so we designed questions with binary yes or no answers, which could easily be used to calculate statistical measures. We also decided to focus on what individual participants did, rather than what was available to them; for example, we asked “Did you take a sabbatical or other research leave?” rather than “Do you have the option to take a sabbatical or other research leave?” We included a question for open-ended comments so that participants could elaborate on answers or add other factors that they felt we had not addressed, since we anticipated that the yes or no answers might leave participants feeling that the complexities of their situations were not captured.

We iteratively pre-tested the survey instrument with twelve librarians who understood the topic, but were not part of the specific population we intended to survey (Canadian university librarians), made adjustments based on feedback, and then re-tested the questions. The survey instrument is provided in Appendix A.

Potential participants were all librarians who worked at the 75 Canadian Research Knowledge Network (CRKN) member institutions, which are listed in Appendix B. We chose these institutions because they are the largest grouping of Canadian university academic librarians, and could therefore provide the largest sample of librarians who are likely to have research as part of their job responsibilities. Because our population included both English- and French-speaking universities, our survey and recruitment materials were professionally translated into French.

We mined the public websites of each CRKN member institution to obtain email addresses of potential participants. Each individual received an email invitation to participate in the study. We also recruited through listservs, Facebook, and Twitter. We emailed the study invitation to 1,683 potential participants in April 2016. We received “mail undeliverable” messages from 30 email addresses, so 1653 potential participants received the invitation.

We asked participants to detail the number of research outputs they had in the past five years. Most of the existing literature on research productivity has focused on peer-reviewed journal articles as the measure of research output. Based on our understanding of research conducted by librarians, we felt that it was also appropriate to include conference presentations, posters, non-peer-reviewed articles, and books. The types of research output that we included all had some aspect of vetting and featured a dissemination process that the researcher needed to follow. As such, we did not include blogs or other self-posted forms of dissemination. In a comment field, we invited participants to elaborate on non-traditional forms of dissemination, such as blogs. Finally, we asked participants to only note their research output that was related to LIS. While some Canadian academic librarians work at institutions where they are explicitly allowed to research in any discipline, including creative works, others are at institutions that explicitly state that their research must be relevant to librarianship, and we wanted our survey to focus on the kind of research that we all have in common. Importantly, participants self-reported their research output, we did not ask for publication details, and so it was the participants themselves who determined whether their output was related to LIS. We did not ask participants to indicate their level of involvement or whether they were sole or a co-author on works reported.

Table 2

Weights for Each Type of Research Output

Output type	Weight
Poster	0.5
Presentation	1
Conference proceeding	1
Non-peer-reviewed article	3
Book chapter	5
Edited book	6
Peer-reviewed article	9
Authored book	10

In our analysis, we used a weighted output score to have one overall measure of productivity for each participant, and to account for the fact that not all outputs are equal. The weights for each type of output, noted in Table 2, were reached via a paired comparison analysis. In this process, we compared each type of output against every other type of output and then we assessed the relative potential impact and contribution of each pair in relation to dissemination. Once each pair had been weighted, we added the relative weights to arrive at an overall weight for each type of output. To calculate the weighted output score for each survey participant, we multiplied the overall weights by the number of research outputs of each type to arrive at a weighted output score. For example, if a participant gave their research output as two posters and two presentations, their overall weighted output score was three.

Results

We received 556 responses to the survey. After removing incomplete responses, we had 453 responses for a 27% response rate, representing 93% of the CRKN member institutions. Table 3 and Figures 1 and 2 summarize demographic characteristics of our survey respondents. Our participants comprised a representative sample of Canadian academic librarians, as compared to the 2015 census carried out by the Canadian Association of Professional Academic Librarians (CAPAL, 2016).

Table 3

Overview of Survey Respondents

		%
Gender	Female	78.5
	Male	21.5
	Other	< 1
Language	English	89
	French	11
Workplace category	Undergraduate	19
	Comprehensive	33
	Medical / Doctoral	49

Research Output

Participants disseminated a range of research output over the past five years, from none at all to multiple types and numbers of output. Presentations made up nearly half of the total reported output. Most of the participants’ output consisted of presentations, non-peer-reviewed and peer reviewed journal articles, and posters – these four types accounted for 89.5% of the total output. Table 4 summarizes participants’ reported research output.

As described above in the Methods section, we calculated a weighted output score for each participant. The distribution of participants’ weighted output scores is shown in Figure 3. The mean score was 21. There were 53 participants who reported no output of any kind, many participants had very low weighted output scores, and a small number of participants had very high weighted output scores. We used a box plot in SPSS to identify extreme values. All weighted output scores above 67 were identified as outliers and removed from the analysis. As well, we decided to focus our analysis on those participants who had demonstrated some regular engagement in research and therefore we set a lower limit for a weighted output score of three, and removed all participants with weighted output scores below that.

Figure 1

Age ranges of participants.

Figure 2

Number of years since participants completed their MLIS (or equivalent).

Table 4

Participants’ Reported Research Output Over the Past Five Years (2011-2016)

Output type	Min.^a	Max.^a	Mean	Median	St. dev.	Total number reported	% of output reported
Presentation	0	27	4.1	3	4.7	1846	47.7
Non-peer-reviewed article	0	36	1.3	0	3.4	609	15.7
Peer-reviewed article	0	14	1.1	0	2.0	492	12.7
Poster	0	10	1.0	0	1.6	462	11.9
Conference proceeding	0	10	.6	0	1.4	283	7.3
Book chapter	0	3	.3	0	.5	116	3
Authored book	0	3	.1	0	.3	34	.1
Edited book	0	3	.1	0	.3	27	.1
Totals						3869	100

Figure 3

Histogram of participants’ weighted output scores.

Since the distribution of weighted output scores does not approximate a normal distribution, we used non-parametric statistical tests to examine the relationship between weighted output score and the identified factors. We used the Mann-Whitney U test with variables that have two nominal groups, the Kruskal-Wallis ANOVA for variables with more than two groups, and Spearman’s rho for correlations of ordinal variables. For the Mann-Whitney and Kruskal-Wallis tests, the null hypothesis is that there is no difference in the distributions; when the null hypothesis is rejected, the difference in the distributions is found to be significant at the .05 level.

Effect of Factors on Research Productivity

The questions in the survey addressed eleven factors (Table 1) that made up the three overarching categories: Individual Attributes, Peers and Community, and Institutional Structures and Supports. Each question mapped to one of the factors, as shown in Appendix A. Some questions or factors straddle more than one category; however, for simplicity, each question was mapped to one factor within one category (Hoffmann et al., 2014). To confirm mappings for the yes or no questions, we ran correlations of the responses. In our previous research we had identified a single factor of Motivations for Research, which we further refined into two factors, Intrinsic Motivations and Extrinsic Motivations, as we analyzed the correlations.

To analyze the effects of our identified factors on research productivity, we tested at three levels: the three overarching categories in aggregate, selected factors within those categories,^{^[1]} and the individual questions that formed the components of the factors. For each of those three levels, we tested against two measures of research productivity: weighted output score and number of peer-reviewed journal articles.

All three categories were significant, both for weighted output score and number of peer-reviewed articles. In other words, many elements contribute to librarians’ research productivity. There was no single category – not Individual Attributes, nor Peers and Community, nor Institutional Structures and Supports – that emerged as being clearly more important than the others, but rather all three were significantly correlated with research output. However, there were noteworthy

findings within the eleven factors which made up the three categories, especially when we tested against different measures of research productivity.

Within the factor of Demographics, only age was significant, and it was only significant when tested against the weighted output score; there was no difference in the distribution of number of peer-reviewed articles over various age ranges. The significance in age came from lower weighted output scores for participants in the age ranges 55-59 and 60-64, as shown in Table 5. The other components of Demographics evaluated were gender, marital status, and whether a respondent cared for dependents. None of those were significant, neither for weighted output score nor number of peer-reviewed articles.

Table 5

Median Research Productivity for Age Range

		Weighted output score			Peer-reviewed articles
Age range	N	Min.	Max.	Median	Min.	Max.	Median
25 - 29	7	7	53	13.5	0	4	0
30 - 34	31	3	52	15.0	0	2	0
35 - 39	58	3	66.5	16.5	0	4	1
40 - 44	57	3	60	17.0	0	4	1
45 - 49	50	3	54.5	20.5	0	4	1
50 - 54	37	3	59	17.0	0	4	1
55 - 59	22	3	61	10.5	0	6	0
60 - 64	15	3	67	9.0	0	2	0
65 - 70	5	3	45	20.0	0	5	0

Table 6

Median Research Productivity for Institutions Where Participants Obtained their MLIS

		Weighted output score			Peer-reviewed articles
Institution^a	N	Min.	Max.	Median	Min.	Max.	Median
University of Alberta	26	3	60	25.0	0	4	1
University of British Columbia	33	4	55	16.0	0	4	1
Dalhousie University	29	3	61	14.0	0	5	0
McGill University	34	3.5	59	21.5	0	4	1
Université de Montréal	28	3	46.5	11.75	0	6	0
University of Toronto	56	3	67	17.25	0	5	0
U.S. Programs^b	14	3	48	11.75	0	2	0
Western University	96	3	66.5	17.0	0	2	1

Within the factor of Education and Experience, there were three significant components for both weighted output score and for number of peer-reviewed articles. Participants who had received research training after completing their MLIS and those who were currently working on an additional advanced degree were more likely to have higher research output. Also, there was a statistically significant difference between institutions from which participants received their MLIS. Table 6 presents median scores by institution for both measures of research productivity. Specifically, post-hoc Mann-Whitney tests indicate that graduates of Université de Montréal had significantly lower output (both weighted output score and number of peer-reviewed articles) than graduates of McGill, Alberta, British Columbia, or Western. Graduates from U.S. programs also reported a significantly lower number of peer-reviewed articles than those from McGill, Alberta, British Columbia, or Western, and they had significantly lower weighted output scores than participants from McGill or Alberta. None of the other aspects of Education and Experience were significant, for either measure of research productivity.

The other nine factors were comprised of the yes or no questions, and we tested both the factors and the individual components. When we tested the nine remaining factors against the weighted output score, all of them were significant. However, when we tested the factors against the number of peer-reviewed articles, Intrinsic Motivation was no longer a significant factor, and there were many fewer components that were significant on their own. Tables 7, 8, and 9 show the significant components for the categories of Individual Attributes, Peers and Community, and Institutional Structures and Supports, respectively.

Within the Individual Attributes category (Table 7) there is a lot of variation in which components are significant when tested against weighted output scores or number of peer-reviewed articles. When peer-reviewed articles was used as the measure of research productivity, none of the Intrinsic Motivation components are significant and only 8 out of 29 components in the category are significant.

The Peers and Community category (Table 8) shows less variation in which components are significant when tested against weighted output scores or number of peer-reviewed articles. Most components in the Peer Support factor are not significant, but most components of the other factors are significant. Within the Collaboration factor, the component “I have done research on my own” is one where answering “No” meant higher collaboration; however, participants who answered “Yes” were more likely to have higher weighted output scores.

The Institutional Structures and Supports category (Table 9) also shows little variation. Interestingly, the component “I do research only because it is a requirement of my job” is the only one that is not significant against weighted output score and is significant against number of peer-reviewed articles, and those who answered “Yes” were more likely to have produced lower numbers of articles.

Table 7

Components of the Individual Attributes Category and Their Significance as Determined by the

Mann-Whitney U Test, Significant at the .05 Level

Factors and Components	Weighted output score	Number of peer-reviewed articles
Intrinsic Motivations
I do research to contribute to more informed decision making in librarianship.	–	–
I do research to contribute to better library services.	–	–
I do research for my personal interest.	–	–
I do research for professional growth.	–	–
I do research to contribute to greater library visibility on campus.	–	–
I do research to advance my career.	significant	–
I do research to build stronger relationships with faculty members.	significant	–
I do research to build a professional reputation for myself.	significant	–
I do research to contribute to a stronger profession.	significant	–
Personal Commitment to Research
I always have a research project that I’m working on.	significant	significant
I schedule dedicated time for research.	significant	significant
I am currently working on a research project.	significant	significant
I have participated in activities that support LIS research (e.g. peer review, editor of a journal, providing writing assistance to a colleague, etc.).	significant	significant
I do research that is meaningful to my practice.	significant	–
I consider research to be a priority.	significant	–
I believe it is important for librarians to contribute to the profession via research.	significant	–
I read research literature on a regular basis.	significant	–
I work on research outside of regular work hours.	significant	–
I have used personal funds to support my research and dissemination (e.g.: personal professional development funds or self funded).	significant	–
Personality Traits
I can achieve my research goals.	significant	significant
I am confident that I have the ability to do research.	significant	significant
I finish the research projects that I start.	significant	significant
I can easily identify questions that could be answered through research.	significant	significant
I enjoy speaking with colleagues about my research.	–	–
I enjoy presenting at conferences.	–	–
I do research to satisfy my curiosity.	–	–
Publishing gives me a personal sense of satisfaction.	significant	–
I enjoy doing research.	significant	–
I enjoy writing for publication.	significant	–

Table 8

Components of the Peers and Community Category and Their Significance as Determined by the Mann-Whitney U Test, Significant at the .05 Level

Factors and components		Weighted output score	Number of peer-reviewed articles
Collaboration
I have done research with other people (co-researchers) at my institution.	significant		significant
I have done research with other people (co-researchers) from other institutions.	significant		significant
I have done research on my own.	significant		–
Community
I feel like I belong to a research community.		significant	significant
I have consulted with an expert to get help on a specific aspect of my research.		significant	significant
I have a network of peers at my institution with whom I talk about research.		significant	significant
I know people who have similar research interests to mine.		significant	significant
I attend conferences in order to connect with others who have similar research interests.		–	–
I have a network of peers from other institutions with whom I talk about research.		significant	–
Professional associations are a source of research community for me.		significant	–
Mentoring
I have been mentored in relation to research activities.		significant	significant
I have mentored others in relation to their research activities.		significant	significant
Peer Support
I have participated in a peer support group related to research.		significant	significant
I ask my colleagues for feedback on my research.		–	–
I have participated in a journal club.		–	–
I have participated in a writing group.		–	–

Table 9

Components of the Individual Attributes Category and Their Significance as Determined by the Mann-Whitney U Test, Significant at the .05 Level

Factors and components	Weighted output score	Number of peer-reviewed articles
Extrinsic Motivations
I have received merit increments or promotion due to my research activities.	significant	significant
I am (formally or informally) expected to participate in research and scholarship.	–	–
I do research only because it is a requirement of my job.	–	significant
Institutional Supports
I have received funding for my research.	significant	significant
I have hired a research assistant to help with research tasks.	significant	significant
I have taken a sabbatical or other kind of leave to work on a research project.	significant	significant
I have space where I am able to work effectively on my research.	significant	significant
I have time to do research within my job.	significant	significant
I am encouraged and supported by my library to do research.	–	–

Finally, we examined participants’ open-ended comments. Most comments corresponded to one of the factors that we had identified, especially to elements of Institutional Structures and Supports: time and perceived institutional support. Within comments about time, some participants specifically said that they did not have time for research because they had an administrative role, which is an area we did not explore. An unexpected theme emerged around precarious employment; participants who were in contract or part-time positions described an inability to plan for research (for example, because of the time needed to submit research ethics applications) and less supported by their institutions to do research.

Discussion

Returning to our research questions, we found that the three categories – Individual Attributes, Peers and Community, and Institutional Structures and Supports – all had a positive effect on librarians’ research productivity. This is an important finding, since it reinforces that many elements contribute to librarians’ research productivity. Figure 4 is a visual representation of how the three categories work together. An environment that embraces all three areas, by encouraging individual attributes, fostering peer and community interaction, and providing institutional supports, will be likely to promote research productivity among librarians.

While our findings unquestioningly show that all three categories have a significant effect on research productivity, they also show that within those categories, there are components that are particularly interesting or that warrant further examination.

For a female-dominated profession such as librarianship, it is both interesting and encouraging that gender did not have a significant effect on research productivity, especially since gender has been shown to be significant for academics generally (Aiston & Jung, 2015).

Within the factor of Education and Experience, it is not surprising that graduates of the Université de Montréal have significantly lower research output; until 2007 when the University of Ottawa’s program started, Montréal’s was the only Canadian MLIS program offered in French, and many of the participants who commented in French said that research was not part of their job responsibilities. Since librarians have long bemoaned the inadequacy of research training received in MLIS programs (Black & Leysen, 1994), it is perhaps also not surprising that having received such training was not significant. However, it is encouraging that participants who received research training after completing their MLIS and those who were working on advanced degrees reported significantly higher productivity. Further examining these components may help to understand how such experiences can best help librarians in their research endeavours.

A close examination of the Institutional Structures and Supports category shows that three components are not significant for weighted output score:

· I am encouraged and supported by my library to do research.

· I am (formally or informally) expected to participate in research and scholarship.

· I do research only because it is a requirement of my job.

However, the last of these components is significant when tested against number of peer-reviewed articles; those who answered “No” are more likely to have higher numbers of articles. In other words, participants who do research for reasons other than a job requirement publish more peer-reviewed articles. This suggests that intrinsic rewards might be stronger motivators than institutional expectations, although the components that we examined for Intrinsic Motivations were not significant when tested against number of peer-reviewed articles. As such, more investigation is needed into what motivates librarians to do research, especially for publication of peer-reviewed articles.

Figure 4

Three categories contributing to librarians’ research productivity.

In considering these results, it is evident that some librarians who feel that they have the expectation and support to do research are productive researchers, while others are not. Similarly, some librarians are productive researchers despite feeling unsupported and not being expected to do research. It is reasonable to imagine that institutional expectations are “powerful motivators” (Hollister, 2016, 369) and yet these findings do not support that.

This is a provocative finding when considered together with the open-ended comments that participants provided, where they overwhelmingly expressed a desire to experience a supportive institutional environment for research, and frustration with library environments that were not supportive or that conveyed mixed messages. For example, participants said that research was “an unfunded mandate,” that “management … values our research activity as long as our ‘regular’ work doesn’t suffer,” and that “we are not encouraged to devote much time to research, yet we are expected to in order to obtain continuing status, prestige, annual report grades [sic], etc.”

Institutional factors therefore warrant more study. What do librarians mean by “feeling supported” to do research? What does it mean that so many participants wanted to feel supported and expressed frustration with a perceived lack of support, and yet these factors did not have a significant effect on research output? What role do institutional expectations play as motivators for producing research? In our previous study (Hoffmann et al., 2014), we had identified a factor of Positive Organizational Climate, which we did not explore in this study. The current research points to the need to investigate organizational climate in order to gain a fuller understanding of librarians’ research productivity.

We also see some striking differences in factors and components that are significant when tested against weighted output score, but are not significant when tested against number of peer-reviewed articles. This is especially evident in the Individual Attributes category; Intrinsic Motivation is only a significant factor when considered against weighted output score, and there is considerable variation in which components are significant.

These findings suggest that, as a collective, librarians must first consider what type of research output they value, in order to have the clearest possible understanding of the factors that will foster their productivity. If librarians want to encourage dissemination of peer-reviewed articles, they may want to focus on a narrower range of factors in order to foster that research output. While peer-reviewed articles are the standard measure of productivity in many disciplines, and by extension, they are considered the goal output for researchers in those disciplines, it is not clear that they are the primary desired research output for librarians. This is seen in the number of peer-reviewed articles reported by our participants, only 12.7% of the total reported research output. This is also supported by Shaw and Szwajcer’s findings (2016) that only 32% of their sample of conference presentations were also published as peer-reviewed articles. In contrast, Tsafe, Chiya, and Aminu’s (2016) analysis of Nigerian librarians found that 69% of total output was journal articles, although they did not distinguish between peer-reviewed and non-peer-reviewed articles, perhaps indicating that dissemination preferences vary by geography.

This apparent tension around the type of research outputs that are highly valued may be related to uncertainty about the value attributed to research that is closely tied to the practice of librarianship. Again looking at the Individual Attributes category, several of the components that are not significant when tested against peer-reviewed articles reflect a focus on professional engagement:

· I do research to build stronger relationships with faculty members.

· I do research to build a professional reputation for myself.

· I do research to contribute to a stronger profession.

· I believe it is important for librarians to contribute to the profession via research.

· I do research that is meaningful to my practice.

Participants who focus on peer-reviewed articles may therefore be less motivated to tie their research to their practice. Or it may be that other venues are perceived to be better for disseminating research that is related to practice. In open-ended comments, participants again expressed uncertainty around this element. For example, “there can be pressure from within your library to do certain types of research (very practice-oriented to your specific library), which might not align with your personal research interests,” and “... I mentioned that I was not highly interested in research but enjoyed presenting at conferences and feel that research and conference presenting are different, my supervisor sees these two as research.”

As mentioned above, Canadian academic librarians have been working to develop a research culture for themselves. As this continues, it will be helpful to include conversations about how we, as a profession, want to value and promote various types of research output, what we mean by research that is tied to practice, and how we value that research. Further study of these questions may also result in more informed conversations.

Overall, this study confirms that the categories and factors we identified in our previous research are relevant and important. The issue of precarious employment in academia has gained attention in recent years, although challenges with contract or part-time work are not new (Feldman & Turnley, 2004), and this may be an additional element to explore in future studies.

The tool that we have developed may be useful for examining research productivity in the future, perhaps as research culture becomes more finely tuned for Canadian academic librarians. It may also be useful for surveying other populations beyond Canada in order to see if there is any variation in which factors are significant and to see what more we can learn about librarians’ research output and productivity.

There are several limitations to our chosen study design. Our study participants were self-selected, so the results reflect a self-selection bias. As well, we were not able to control participants’ responses to ensure that they were replying as we intended. For example, questions about research output asked participants to provide counts for the last five years, but it is possible that someone reported publications over their career. The questions with bivariate variables (yes or no answers) were helpful for our analysis, but also limited the level of detail in the responses and restricted the scope of possible statistical tests that we could run. Because we focused our analysis on participants who had some regular engagement with research, as determined by weighted output scores between 3 and 67, we do not know what factors are significant for the participants excluded from analysis. Finally, the quantitative approach of this study means that we are not able to capture the full complexity of individual factors; though we can identify which factors are statistically significant, but we cannot explain why this is the case. We are also not able to take into account the context of the individuals who participated in the survey; for example, whether a particular situation in someone’s institution or personal life has affected their research productivity.

Conclusions

The findings from this quantitative study contribute to a greater understanding of librarians’ research productivity and the factors that contribute to research success. While we might have hoped for the findings to reveal a ‘magic bullet’ for research output, they instead reveal even more complexity. Research productivity is significantly impacted by individual qualities, by interaction and support from peers and community, and by strong institutional supports. These findings suggest that librarians and library administrators focus on all three of these areas in order to promote research productivity.

At the same time, these findings raise additional questions and highlight aspects where more investigation is needed. Our participants’ expressed desire for supportive institutional climates is in tension with the finding that feeling supported by one’s institution and feeling expected to do research are not significantly related to research output. Further examination of librarians’ motivations for doing research, and of the interplay between intrinsic and extrinsic motivations, may help to illuminate the role of a supportive institutional climate.

Also related to motivations, our findings suggest that it will be important to explore questions about the value of research that is closely related to practice and the value of various types of research output. This may help to build a more cohesive research culture and may also help reveal factors that are key for different types of research, since we saw that the significance of some factors, and of some components within factors, varied depending on the measure of research productivity that we examined.

We plan to further extend this research with qualitative exploration into one or more of the areas above. We hope that this study’s findings will prompt others to also explore librarians’ research from the perspective of enablers rather than barriers.

Acknowledgements

The authors wish to thank Kristi Thompson, Data Librarian at the University of Windsor, for advice and assistance on data analysis. We also thank our 12 colleagues who pre-tested our survey instrument and provided valuable feedback.

References

Aiston, S. J., & Jung, J. (2015). Women academics and research productivity: An international comparison. Gender and Education, 27(3), 205-220. https://doi.org/10.1080/09540253.2015.1024617

Baro, E. E., & Ebhomeya, L. (2012). A comparative study of the publication output of librarians and academics in universities in the south-south zone of Nigeria. Journal of Scholarly Publishing, 43(2), 200-219. https://doi.org/10.1353/scp.2012.0009

Black, W. K., & Leysen, J. M. (1994). Scholarship and the academic librarian. College & Research Libraries, 55(3), 229-241. https://doi.org/10.5860/crl_55_03_229

Brew, A., Boud, D., Namgung, S. U., Lucas, L., & Crawford, K. (2016). Research productivity and academics’ conceptions of research. Higher Education, 71(5), 681-697. https://doi.org/10.1007/s10734-015-9930-6

Brown, J. M. (2001). Time and the academic librarian. portal: Libraries and the Academy, 1(1), 59-70. https://doi.org/10.1353/pla.2001.0009

Campbell, K., Ellis, M., & Adebonojo, L. (2012). Developing a writing group for librarians: The benefits of successful collaboration. Library Management, 33(1/2), 14-21. https://doi.org/10.1108/01435121211203284

Canadian Association of Professional Academic Librarians. (2016). 2016 census of Canadian academic librarians. User guide and results summary. Retrieved from http://capalibrarians.org/wp/wp-content/uploads/2016/12/Census_summary_and_user_guide_December_16_2016.pdf

Canadian Association of Research Libraries. (2017). Librarians’ Research Institute. Retrieved from http://www.carl-abrc.ca/strengthening-capacity/workshops-and-training/librarians-research-institute/

Centre for Evidence Based Library and Information Practice (2017). Vision and mission. Retrieved from http://library.usask.ca/ceblip/about/vision--mission.php

Carson, P., Colosimo, A. L., Lake, M., & McMillan, B. (2014). A “Partnership” for the professional development of librarian researchers. Partnership: The Canadian Journal of Library and Information Practice and Research, 9(2). https://doi.org/10.21083/partnership.v9i2.3037

Cirasella, J., & Smale, M. A. (2011). Peers don’t let peers perish: Encouraging research and scholarship among junior library faculty. Collaborative Librarianship, 3(2), 98-109. Retrieved from http://digitalcommons.du.edu/collaborativelibrarianship/vol3/iss2/5

Edwards, P. M., Jennerich, E. Z., & Ward, J. L. (2009). Supporting a culture of library research at the University of Washington at Seattle. In E. Connor (Ed.), An introduction to staff development in academic libraries (pp. 77-89). New York, NY: Routledge.

Exner, N., & Houk, A. H. (2010). Starting the write way: Comparing two library scholarly development programs. Library Leadership & Management, 24(4), 178-182. Retrieved from https://journals.tdl.org/llm/index.php/llm/index

Fallon, H. (2012). Using a blended group learning approach to increase librarians’ motivation and skills to publish. New Review of Academic Librarianship, 18(1), 7-25. https://doi.org/10.1080/13614533.2012.654673

Feldman, D. C. & Turnley, W. H. (2004). Contingent employment in academic careers: Relative deprivation among adjunct faculty. Journal of Vocational Behavior, 64(2), 284-307. https://doi.org/10.1016/j.jvb.2002.11.003

Fitzgibbons, M., Kloda, L., & Miller-Nesbitt, A. (2017). Exploring the value of academic librarians’ participation in journal clubs. College & Research Libraries, 78(6), 774-788. https://doi.org/10.5860/crl.78.6.774

Fox, D. (2007). Finding time for scholarship: A survey of Canadian research university librarians. portal: Libraries and the Academy, 7(4), 451-462. https://doi.org/10.1353/pla.2007.0041

Hall, L. W., & McBain, I. (2014). Practitioner research in an academic library: Evaluating the impact of a support group. The Australian Library Journal, 63(2), 129-143. http://dx.doi.org/10.1080/00049670.2014.898238

Harrington, M. R., & Gerolami, N. (2014). Highs and lows: An examination of academic librarians’ collective agreements. In J. Dekker & M. Kandiuk (Eds.), In solidarity: Academic librarian labour activism and union participation in Canada (pp. 151-169). Sacramento, CA: Library Juice Press.

Hoffmann, K., Berg, S. A., & Koufogiannakis, D. (2014). Examining success: Identifying factors that contribute to research productivity across librarianship and other disciplines. Library and Information Research, 38(119), 13-28. Retrieved from http://www.lirgjournal.org.uk/lir/ojs/index.php/lir/index

Hollister, C. V. (2016). An exploratory study on post-tenure research productivity among academic librarians. Journal of Academic Librarianship, 42(4), 368-381. https://doi.org/10.1016/j.acalib.2016.04.021

Jacobs, H. L. M., & Berg, S. A. (2013). By librarians, for librarians: Building a strengths-based institute to develop librarians’ research culture in Canadian academic libraries. Journal of Academic Librarianship, 39(3), 227-231. https://doi.org/10.1016/j.acalib.2013.02.003

Kennedy, M. R., & Brancolini, K. R. (2012). Academic librarian research: A survey of attitudes, involvement, and perceived capabilities. College & Research Libraries, 73(5), 431-448. https://doi.org/10.5860/crl-276

Lessick, S., Perryman, C., Billman, B. L., Alpi, K. M., De Groote, S. L., & Babin, T. D. (2016). Research engagement of health sciences librarians: a survey of research-related activities and attitudes. Journal of the Medical Library Association: JMLA, 104(2), 166-173. https://doi.org/10.3163/1536-5050.104.2.015

McBain, I., Culshaw, H., & Hall, L. W. (2013). Establishing a culture of research practice in an academic library: An Australian case study. Library Management, 34(6/7), 448-461. https://doi.org/10.1108/LM-08-2012-0053

Meadows, K. N., Berg, S. A., Hoffmann, K., Torabi, N., & Gardiner, M. M. (2013). A needs-driven and responsive approach to supporting the research endeavours of academic librarians. Partnership: The Canadian Journal of Library and Information Practice and Research, 8(2). https://doi.org/10.21083/partnership.v8i2.2776

Mierke, J., & Williamson, V. (2017). A Framework for Achieving Organizational Culture Change. Library Leadership & Management, 31(2). Retrieved from https://journals.tdl.org/llm/index.php/llm/index

Miller, J. P., & Benefiel, C. R. (1998). Academic librarians and the pursuit of tenure: The support group as a strategy for success. College & Research Libraries, 59(3), 260-265. https://doi.org/10.5860/crl.59.3.260

O’Brien, T., & Cronin, K. (2016). Research output of academic librarians from Irish higher education institutions 2000-2015: Findings from a review, analysis, and survey. New Review of Academic Librarianship, 22(2-3), 203-224. https://doi.org/10.1080/13614533.2016.1181666

Pickton, M. (2016). Facilitating a research culture in an academic library: Top down and bottom up approaches. New Library World, 117(1/2), 105-127. https://doi.org/10.1108/NLW-10-2015-0075

Powell, R. R., Baker, L. M., & Mika, J. J. (2002). Library and information science practitioners and research. Library & Information Science Research, 24(1), 49-72. https://doi.org/10.1016/S0740-8188(01)00104-9

Sapon-White, R., King, V., & Christie, A. (2004). Supporting a culture of scholarship for academic librarians. portal: Libraries and the Academy, 4(3), 407-422. https://doi.org/10.1353/pla.2004.0044

Sassen, C., & Wahl, D. (2014). Fostering research and publication in academic libraries. College & Research Libraries, 75(4), 458-491. https://doi.org/10.5860/crl.75.4.458

Schrader, A. M., Shiri, A., & Williamson, V. (2012). Assessment of the research learning needs of University of Saskatchewan librarians: A case study. College & Research Libraries, 73(2), 147-163. https://doi.org/10.5860/crl-235

Shaw, C. E., & Szwajcer, A. L. (2016). Publication rate of presentation abstracts presented at the Canadian Health Libraries Association (CHLA/ABSC) annual meetings from 2004-2009. Performance Measurement and Metrics, 17(3), 252-262. https://doi.org/10.1108/PMM-07-2016-0034

Smigielski, E. M., Laning, M. A., & Daniels, C. M. (2014). Funding, time, and mentoring: A study of research and publication support practices of ARL member libraries. Journal of Library Administration, 54(4), 261-276. https://doi.org/10.1080/01930826.2014.924309

Spring, H., Doherty, P., Boyes, C., & Wilshaw, K. (2014). Research engagement in health librarianship: Outcomes of a focus group. Library & Information Science Research, 36(3-4), 142-153. https://doi.org/10.1016/j.lisr.2014.06.004

Stephens, J., Sare, L., Kimball, R., Foster, M., & Kitchens, J. (2011). Tenure support mechanisms provided by the Faculty Research Committee at Texas A&M University Libraries: A model for academic libraries. Library Management, 32(8/9), 531-539. https://doi.org/10.1108/01435121111187897

Tsafe, A. G., Chiya, U., & Aminu, B. A. (2016). Scholarly publications of librarians in universities in Nigeria: 2000-2012 – a bibliometric analysis. Library Philosophy and Practice. Retrieved from http://digitalcommons.unl.edu/libphilprac/1394

Tysick, C., & Babb, N. (2006). Perspectives on... Writing support for junior faculty librarians: A case study. Journal of Academic Librarianship, 32(1), 94-100. https://doi.org/10.1016/j.acalib.2005.10.005

Walters, W. H. (2016). The faculty subculture, the librarian subculture, and librarians’ scholarly productivity. portal: Libraries and the Academy, 16(4), 817-843. http://doi.org/10.1353/pla.2016.0054

Wilson, V. (2017). The C-EBLIP Fall Symposium: Librarians as researchers. Evidence Based Library and Information Practice, 12(2), 4-5. https://doi.org/10.18438/B8RS9X

Appendix A

Survey Instrument (English Version)

In the survey text below, each question is annotated with an abbreviation to indicate the factor to which it is mapped. These annotations are provided for this paper and were not included in the survey instrument.

D Demographics

EE Education and Experience

EM Extrinsic Motivations

IM Intrinsic Motivations

IS Institutional Supports

PCR Personal Commitment to Research

PT Personality Traits

COL Collaboration

COM Community

M Mentoring

PS Peer Support

Factors Influencing Research Productivity

The survey is expected to take less than 20 minutes and includes questions related to: a.) Education and Professional Experience; b.) Factors Influencing Research Productivity; c.) Demographic Information; d.) Research Outputs;

Some of the questions are simple yes or no questions and require you to choose the best answer that reflects your situation or your feelings.

The study seeks participation from Canadian academic librarians (at CKRN institutions) who are and who are not active researchers. For this study, we are using the definition of research provided by the Tri-Council Policy Statement: Research is “defined as an undertaking intended to extend knowledge through a disciplined inquiry and/or systematic investigation.”

The researchers are interested in Canadian academic librarians’ contributions to library and information studies (LIS) research. While it is recognized that librarians may undertake research outside of LIS, the researchers are gathering information in this study only on LIS-related research.

By taking this online survey I am indicating that I have read the information letter and voluntarily agree to participate in the research study.

Please remember to print a copy of the information letter for your records.

Where do you currently work? EE

Drop-down menu of the 75 CKRN institutions.

Do you have tenure (or equivalent) or are you in a tenure-track (or equivalent) position? EE

What year did you complete your MLIS degree (or equivalent)? EE

Drop-down menu of years

Where did you obtain your MLIS degree (or equivalent?) EE

Dalhousie University
McGill University
University of Alberta
University of British Columbia
Université de Montréal
University of Ottawa
University of Toronto
Western University
Other, please specify...

During your MLIS program (or equivalent), did you complete any of the following: EE

Check all that apply.

research methods course
independent research study
thesis

Since completing your MLIS (or equivalent), have you taken any formal research training? EE

Check all that apply.

university-level research course
full- or half-day research workshop
CARL Librarians' Research Institute
Thinking Qualitative Workshop Series
online non-credit research course
Other, please specify...

Do you have an advanced degree in addition to your MLIS (or equivalent)? EE

Check all that apply.

Yes, thesis-based Masters
Yes, non-thesis-based Masters
Yes, PhD
No additional degree
Other, please specify...

Are you currently working towards an additional degree? EE

Check all that apply.

Yes, thesis-based Masters
Yes, non-thesis-based Masters
Yes, PhD
No additional degree
Other, please specify…

Please indicate whether or not each statement applies to you.

(presented in random order)

PCR	I consider research to be a priority.	Yes \| No
PCR	I am currently working on a research project.	Yes \| No
PCR	I always have a research project that I’m working on.	Yes \| No
PCR	I do research that is meaningful to my practice.	Yes \| No
PCR	I believe it is important for librarians to contribute to the profession via research.	Yes \| No
PCR	I work on research outside of regular work hours.	Yes \| No
PCR	I schedule dedicated time for research.	Yes \| No
PCR	I have participated in activities that support LIS research (e.g. peer review, editor of a journal, providing writing assistance to a colleague, etc.)	Yes \| No
PCR	I have used personal funds to support my research and dissemination (e.g.: personal professional development funds or self funded).	Yes \| No
PCR	I read research literature on a regular basis.	Yes \| No
IS	I am encouraged and supported by my library to do research.	Yes \| No
IS	I have time to do research within my job.	Yes \| No
IS	I have space where I am able to work effectively on my research.	Yes \| No
IS	I have taken a sabbatical or other kind of leave to work on a research project.	Yes \| No
IS	I have hired a research assistant to help with research tasks.	Yes \| No
IS	I have received funding for my research.	Yes \| No
COM	I have a network of peers at my institution with whom I talk about research.	Yes \| No
COM	I have a network of peers from other institutions with whom I talk about research.	Yes \| No
COM	I know people who have similar research interests to mine.	Yes \| No
COM	Professional associations are a source of research community for me.	Yes \| No
COM	I attend conferences in order to connect with others who have similar research interests.	Yes \| No
COM	I feel like I belong to a research community.	Yes \| No
COM	I have consulted with an expert to get help on a specific aspect of my research.	Yes \| No
COL	I have done research with other people (co-researchers) at my institution.	Yes \| No
COL	I have done research with other people (co-researchers) from other institutions.	Yes \| No
COL	I have done research on my own.	Yes \| No
PS	I have participated in a peer support group related to research.	Yes \| No

Please indicate whether or not each statement applies to you.

(presented in random order)

PS	I have participated in a writing group.	Yes \| No
PS	I have participated in a journal club.	Yes \| No
PS	I ask my colleagues for feedback on my research.	Yes \| No
EM	I have received merit increments or promotion due to my research activities.	Yes \| No
EM	I am (formally or informally) expected to participate in research and scholarship.	Yes \| No
EM	I do research only because it is a requirement of my job.	Yes \| No
PT	I enjoy doing research.	Yes \| No
PT	I enjoy writing for publication.	Yes \| No
PT	I am confident that I have the ability to do research.	Yes \| No
PT	I can achieve my research goals.	Yes \| No
PT	I enjoy presenting at conferences.	Yes \| No
PT	I enjoy speaking with colleagues about my research.	Yes \| No
PT	Publishing gives me a personal sense of satisfaction.	Yes \| No
PT	I can easily identify questions that could be answered through research.	Yes \| No
PT	I do research to satisfy my curiosity.	Yes \| No
PT	I finish the research projects that I start.	Yes \| No
IM	I do research to advance my career.	Yes \| No
IM	I do research for my personal interest.	Yes \| No
IM	I do research to contribute to better library services.	Yes \| No
IM	I do research for professional growth.	Yes \| No
IM	I do research to build a professional reputation for myself.	Yes \| No
IM	I do research to contribute to more informed decision making in librarianship.	Yes \| No
IM	I do research to contribute to greater library visibility on campus.	Yes \| No
IM	I do research to build stronger relationships with faculty members.	Yes \| No
IM	I do research to contribute to a stronger profession.	Yes \| No
M	I have been mentored in relation to research activities.	Yes \| No
M	I have mentored others in relation to their research activities.	Yes \| No

What is your gender? D

Female
Male
Other

What month and year were you born? D

What is your marital status? D

single
married
living with partner
divorced
separated
widowed
Other, please specify...

Do you have children or adults who depend on you for care? D

Child(ren) under 18 years of age
Child(ren) over 18 years of age
Other adult dependent upon me for care
No children or dependent adult

Can you think of other factors that were not fully captured in the previous questions that have affected your research productivity? If so, please share them here.

Open text box

Thinking back over the last five years, please indicate how many times you have disseminated your LIS-related research in each of the following venues:

The researchers are interested in the research outputs of Canadian academic librarians related to library and information studies (LIS). While it is recognized that librarians may undertake research outside of LIS, do research that is not disseminated, or disseminate research in non-traditional formats, in this question the researchers are gathering information about specific ways of disseminating LIS-related research.

presented a poster at a conference (both peer reviewed and not)	drop-down 0-50
gave an oral presentation at a conference (both peer reviewed and not)	drop-down 0-50
published in conference proceedings	drop-down 0-50
published a non-peer reviewed journal article	drop-down 0-50
published a peer reviewed journal article	drop-down 0-50
published a chapter in a book (contributed chapter)	drop-down 0-50
authored a book (solo or co-author)	drop-down 0-50
edited a book (collection of contributed chapters)	drop-down 0-50

The scholarly landscape is changing and researchers are disseminating their research outputs in new ways. Please list any ways that you have disseminated your research that were not included in the previous question.

Open text box

Appendix B

Canadian Research Knowledge Network (CRKN) Member Institutions

Acadia University

Algoma University

Athabasca University

Bishop’s University

Brandon University

Brock University

Cape Breton University

Carleton University

Concordia University

Concordia University College of Alberta

Dalhousie University

École Polytechnique de Montréal

HEC Montréal

Kwantlen Polytechnic University

Lakehead University

Laurentian University

MacEwan University

McGill University

McMaster University

Memorial University of Newfoundland

Mount Allison University

Mount Royal University

Mount Saint Vincent University

Nipissing University

NSCAD University

OCAD University

Queen’s University

Royal Military College of Canada

Royal Roads University

Ryerson University

Saint Mary’s University

Simon Fraser University

St. Francis Xavier University

The King’s University College of Alberta

Thompson Rivers University

Trent University

Trinity Western University

Université de Moncton

Université de Montréal

Université de Sherbrooke

Université du Québec:

École nationale d’administration publique

École de technologie supérieure

Institut national de la recherche scientifique

Université du Québec à Chicoutimi

Université du Québec à Montréal

Université du Québec à Rimouski

Université du Québec à Trois-Rivières

Université du Québec en Abitibi-Témiscamingue

Université du Québec en Outaouais

Télé-université du Québec

Université Laval

Université Sainte-Anne

University of the Fraser Valley

University of Alberta

University of British Columbia

University of Calgary

University of Guelph

University of Lethbridge

University of Manitoba

University of New Brunswick

University of Northern British Columbia

University of Ontario Institute of Technology

University of Ottawa

University of Prince Edward Island

University of Regina

University of Saskatchewan

University of Toronto

University of Victoria

University of Waterloo

University of Windsor

University of Winnipeg

Vancouver Island University

Western University

Wilfrid Laurier University

York University

[1] We could not test Demographics or Education and Experience as factors, only the individual questions comprising those factors, because the forms of the questions did not lend themselves to being combined in aggregate.