Access to scientific literature by the conservation community

Access to the scientific literature is perceived to be a challenge to the biodiversity conservation community, but actual level of literature access relative to needs has never been assessed globally. We examined this question by surveying the constituency of the International Union for Conservation of Nature (IUCN) as a proxy for the conservation community, generating 2,285 responses. Of these respondents, ∼97% need to use the scientific literature in order to support their IUCN-related conservation work, with ∼50% needing to do so at least once per week. The crux of the survey revolved around the question, “How easy is it for you currently to obtain the scientific literature you need to carry out your IUCN-related work?” and revealed that roughly half (49%) of the respondents find it not easy or not at all easy to access scientific literature. We fitted a binary logistic regression model to explore factors predicting ease of literature access. Whether the respondent had institutional literature access (55% do) is the strongest predictor, with region (Western Europe, the United States, Canada, Australia and New Zealand) and sex (male) also significant predictors. Approximately 60% of respondents from Western Europe, the United States, Canada, Australia and New Zealand have institutional access compared to ∼50% in Asia and Latin America, and ∼40% in Eastern Europe and in Africa. Nevertheless, accessing free online material is a popular means of accessing literature for both those with and without institutional access. The four journals most frequently mentioned when asked which journal access would deliver the greatest improvements to the respondent’s IUCN-related work were Conservation Biology, Biological Conservation, Nature, and Science. The majority prefer to read journal articles on screen but books in hard copy. Overall, it is apparent that access to the literature is a challenge facing roughly half of the conservation community worldwide.


Introduction
whether the contact had an indicated language of preference in IUCN's customer relationship 153 management (CRM) system; those without a preference received the English-language version 154 by default. We sought to be inclusive of all who had any need for scientific literature in their 155 IUCN-related work and did not seek to limit the survey to those of particular roles or 156 backgrounds. Therefore, our survey results likely include some responses from individuals who 157 work in areas other than biodiversity conservation and require other types of literature e.g. legal 158 or management literature. We sent a reminder on 10 August 2016 and the survey was closed on 159 12 August 2016. The survey was wholly voluntary.

161
We aggregated results by country according to the UN regional groups-Africa, Asia-Pacific, 162 Eastern Europe, Latin America and the Caribbean, and Western Europe and Others (which 163 includes the United States, Canada, Australia and New Zealand). While a range of other national 164 socio-economic parameters (e.g. GDP, income equality, education of girls and boys) could be 165 included, we chose to select these regional groupings to reflect political and social as well as hard + hard = 0, easy + very easy = 1) and fitting a binary logistic regression model to the full 173 rank dataset of 1,970 respondents who answered all questions under consideration in the model. 174 We began with consideration of five variables suspected likely to influence ease of literature In total, we received 2,285 responses to our survey. This represents 11% of the IUCN 196 constituency to whom the survey was directly distributed, although it is difficult to give a precise 197 return rate given that the actual number of potential participants is unknown. Anecdotal email 198 responses suggest that some Member focal points erroneously thought the survey should be filled 199 out on behalf of the entire organization. Also, our results will be biased against those who did not The survey revolved around the question, "How easy is it for you currently to obtain the 208 scientific literature you need to carry out your IUCN-related work?" Roughly half (49%) of all 209 2,004 respondents to this question find it not easy or not at all easy to access scientific literature 210 ( Figure 2).

212
Overall, 47% of the 2,004 respondents to the question reported having no institutional access to 213 scientific literature online, which correlates greatly to ease of access to literature. Among those 214 with online institutional access, 72% found it easy to obtain access to required literature. By 215 contrast, a similar percentage (74%) of those reporting no institutional access found it difficult to 216 access scientific literature ( Figure 3).

218
Not surprisingly, then, institutional access was the primary explanatory variable predicting ease 219 of access. Exponentiating the model coefficient shows that institutional access increased the odds 220 of easy access to literature by a factor of 6.86; it would seem that affiliation with an institution 221 with a library greatly increases the odds of easy access to scientific literature. Being male and 222 being based in the Western Europe and Others Group were also significant predictors of ease of 223 access.

225
Respondents to our survey were based in 170 countries, allowing us to examine variation across 226 the five United Nations regional socio-geographical groupings. Nearly half of respondents 227 belonged to the Western Europe and Others Group (Figure 4). The two socio-geographic areas 228 with the greatest difficulty in obtaining scientific literature were Africa and Eastern Europe, with 229 63% of respondents from Africa and 57% of respondents from Eastern Europe reporting that accessing scientific literature as not easy or not at all easy ( Figure 5). Not surprisingly, these two 231 regions also reported the least online institutional access to scientific literature ( Figure 6). This 232 supports our model findings that being based in a country in the Western Europe and Others 233 group as opposed to one in Africa increased the odds of easy access by a factor of 1.73, as shown 234 by exponentiating the region coefficient. Other regions were not significant predictors. A 235 Tukey's post hoc test showed regional differences between Africa and Western Europe and 236 Others (p = 0.005), but no significant differences between all other pairwise combinations of 237 regions.

239
More than twice as many men (1,556 respondents) as women (710 respondents) took the survey.  Commission did not emerge as a significant predictor of ease of access in our model.

255
Institutional access to the scientific literature did vary though, from 60% among those whose 256 specialisation includes environmental law to 42% among those whose expertise includes 257 protected areas (Table 2).

259
Overall, 433 of all survey respondents reported being an employee of IUCN itself or an IUCN 260 Member organization, which we used to assess variation by sector. However, as respondents as a 261 whole were not specifically asked to identify their work sector or employer, this partial snapshot   identify those journals to which conservationists perceive that access would benefit their work 373 most greatly. Some respondents listed more than one journal: in such cases, scores were divided 374 among the journals listed (e.g. if four journals were listed, these were scored 0.25 each).

376
In total, 235 journals were mentioned by respondents, including ten listed as most desired more 377 than ten times. These included six specialist conservation journals (Conservation Biology, In addition to preferred journals, conservation professionals also have different preferred reading 391 formats between books and journal articles. To discern this difference, we asked in Question 8, 392 "In what format do you prefer to read scientific literature?" where the choices were "I prefer 393 reading on a screen", "I prefer printing out to read," and "I prefer the original hard copy." When 394 reading articles from scientific journals, the majority (59%) prefer reading on screen, but for 395 books, the majority (59%) prefer to read the original hard copy. The preference for electronic 396 journals has been noted elsewhere (Kaur, 2012). Much concern has been raised about the challenges to the scientific process faced by Africa,