This portion of the T&F OA survey supports arguments that scholars as a group do not support the Creative Commons - Attribution Only license (CC-BY), but rather when using CC licenses tend to prefer more restrictive licenses, with CC-Attribution-Noncommercial-NoDerivatives (CC-BY-NC-ND) being the most popular option. There was strong support for text and data mining. There was an interesting difference in reaction to pre-approving translations (largely positive) and adapations as a whole (largely negative), suggesting the possibility of a more nuanced approach such as ND with preapproval of translations outside the CC license per se. Attribution is taken as a given; further research into the question of attribution might be merited as attribution may not be advisable in the case of research data and the norms for attribution can vary, for example with scholarship and Wikipedia. This portion of the survey indicates support for Taylor and Francis traditional practices (Exclusive License to Publish and Copyright Transfer), which is not surprising considering the survey pool (scholars connected with T&F) and high probability of bias in these responses.
This is the third post in my Taylor and Francis Open Access Survey critique series covering p. 8 - 10 of the results of this survey, on Attitudes and values regarding the dissemination of your research and Licenses. Please see the main post above for overall critique of this survey. In brief, social science research intended to inform public policy should not be conducted by a commercial entity with a stake in the outcome (this is fox researches hen research), and the survey as a whole is significantly flawed methodologically. The following comments should be interpreted in the light of this overall critique.
Taylor & Francis questions are in bold.
What are your attitudes and values regarding the dissemination of your research? p. 8 (14,533 respondents)
It is acceptable for ... without my prior knowledge or permission, provided I receive credit as the original author.
Comment: this question assumes attribution as a given. For further research, it should be noted that there are reasons to question the assumption of attribution. For example, with data sharing public domain (not attribution) may be preferable, and some scholars are experimenting with contributing to Wikipedia, which uses attribution but in a different way (authors are anonymous).
The strongest support was for my work to be re-used for non-commercial gain, with 64% either agreeing or strongly agreeing. This was followed by others to use my work in text - or data-mining with 48% agreeing or strongly agreeing. Translations and including works in an anthology received only slightly less support.
Respondents reacted negatively to others to use my work for commercial gain with 67% agreeing or strongly agreeing and others to adapt my work with 50% disagreeing or strongly disagreeing.
Licenses (p. 9) (13,143 respondents)
Please indicate in each case if you would be willing to sign the license when publishing your research (Response options: Yes, always; Only in certain circumstances; No, never).
Response options included a variety of Creative Commons licenses, with a significant omission: the Sharealike option, along with License to Publish and Copyright Transfer. The most popular response was Creative Commons - Attribution - Noncommercial - NoDerivatives CC-BY-NC-ND with 71% responding Yes, always.
Exclusive License to Publish and Copyright Transfer were only slightly less popular. I am inclined to discount these responses because of the obvious strong likelihood of social desirability and acquiescence biases, that is, this is a T&F survey which is likely to elicit positive responses to T&F practices (see the main post in this series for detailed explanation and references); also the pool of respondents are obviously a group that continues to work with this traditional publisher ("Yes, always" actually defines the group), and respondents in favor of the publisher's practices may have been more likely to respond.
The Creative Commons - Attribution only license (CC-BY) received the lowest positive response, with 15% indicating "Yes, always", 59% indicating "Only in certain circumstances" and 26% "No, never".
License preferences (p. 10) (12,882 responses)
Please choose your most preferred, and your second most preferred, of the above licences.
Please choose your least preferred of the above licences.
Responses show a similar pattern to the questions on the page above. The most popular option was CC-BY-NC-ND. CC-BY elicited a strong negative reaction.
The response bias discussed above should be taken into account in interpreting the apparent popularity of traditional Taylor & Francis practices (Exclusive License to Publish, Copyright Transfer). That is, the survey population consists of scholars who work with Taylor & Francis; scholars who favor T&F practices may have been more likely to respond to the survey; and of respondents, there is a strong likelihood of social desirability and acquiescent response biases.
This section of the survey supports the argument that many scholars do not support the Creative Commons - Attribution Only license (CC-BY), but rather when using CC licenses prefer more restrictive licenses, in this case with CC-BY-NC-ND being the top preference. Other evidence supporting this argument comes from Nature Publishing Group's analysis of author choices for Scientific Reports, referenced from here, which offers a range of CC licenses and found that CC-BY is selected by only 5% of scholars.
This is important because many open access advocates equate open access with the CC-BY license. I argue that the similarity between CC-BY and the open access definition of the Budapest Open Access Initiative is superficial in nature, that CC-BY has both positive and negative implications for scholarship, and as a default has significant loopholes which would present an ongoing danger to open access. My work in this area can be found in my Open Access and Creative Commons critique series.
The detailed questions about re-use are somewhat useful in indicating what kinds of re-use scholars are likely to be willing to grant. Noncommercial, text and data-mining, translations and including works in anthologies are all supported by many scholars. The difference between adaptions (negative reaction) and translations (positive reaction) is interesting, as this suggest that while scholars do not necessarily want to grant blanket rights for all kinds of derivatives, many would be okay with blanket permissions for translations. This suggests that a more nuanced approach might be worthwhile, for example perhaps NoDerivs with pre-approval of translations handled outside of the CC license per se.
The results of this survey also show that respondents to a T&F survey tend to respond favorably to T&F traditional practices. This is evidence of ongoing support for T&F, which may not be necessary given that authors and editorial boards continue to work with this publisher; beyond that, given the high probability of bias, this finding should be largely discounted.
This post is part of the Taylor and Francis Open Access Survey critique series.