Big data challenges for the social sciences: from society and opinion to replications

Posté le par

Author : Dominique Boullier

Big Data dealing with the social produce predictive correlations for the benefit of brands and web platforms. Beyond “society” and “opinion” for which the text lays out a genealogy, appear the “traces” that must be theorized as “replications” by the social sciences in order to reap the benefits of the uncertain status of entities’ widespread traceability.

High frequency replications as a collective phenomenon did exist before the digital networks emergence but now they leave traces that can be computed. The third generation of Social Sciences currently emerging must assume the specific nature of the world of data created by digital networks, without reducing them to the categories of the sciences of “society” or “opinion”.

Examples from recent works on Twitter and other digital corpora show how the search for structural effects or market-style trade-offs are prevalent even though insights about propagation, virality and memetics could help build a new theoretical framework.

URL : http://arxiv.org/abs/1607.05034

Research data management in social sciences and humanities: A survey at the University of Lille (France)

Posté le par

Authors : Joachim Schöpfel, Hélène Prost

The paper presents results from a campus-wide survey at the University of Lille (France) on research data management in social sciences and humanities.

The survey received 270 responses, equivalent to 15% of the whole sample of scientists, scholars, PhD students, administrative and technical staff (research management, technical support services); all disciplines were represented.

The responses show a wide variety of practice and usage. The results are discussed regarding job status and disciplines and compared to other surveys. Four groups can be distinguished, i.e. pioneers (20-25%), motivated (25-30%), unaware (30%) and reluctant (5-10%).

Finally, the next steps to improve the research data management on the campus are presented.

URL : Research data management in social sciences and humanities: A survey at the University of Lille (France)

Alternative location : http://libreas.eu/ausgabe29/09schoepfel/

What Motivates Authors of Scholarly Articles? The Importance of Journal Attributes and Potential Audience on Publication Choice

Posté le par

Authors : Carol Tenopir, Elizabeth Dalton, Allison Fish, Lisa Christian, Misty Jones, MacKenzie Smith

In this article we examine what motivations influence academic authors in selecting a journal in which to publish.

A survey was sent to approximately 15,000 faculty, graduate students, and postdoctoral researchers at four large North American research universities with a response rate of 14.4% (n = 2021).

Respondents were asked to rate how eight different journal attributes and five different audiences influence their choice of publication output. Within the sample, the most highly rated attributes are quality and reputation of journal and fit with the scope of the journal; open access is the least important attribute. Researchers at other research-intensive institutions are considered the most important audience, while the general public is the least important.

There are significant differences across subject disciplines and position types. Our findings have implications for understanding the adoption of open access publishing models.

URL : What Motivates Authors of Scholarly Articles? The Importance of Journal Attributes and Potential Audience on Publication Choice

Alternative location : http://www.mdpi.com/2304-6775/4/3/22

Deconstructing the Durham Statement: The Persistence of Print Prestige During the Age of Open Access

Posté le par

Author : Sarah Reis

In the seven years following the promulgation of the Durham Statement on Open Access to Legal Scholarship, law journals have largely responded to the call to make articles available in open, electronic formats, but not to the call to stop print publication and publish only in electronic format.

Nearly all of the flagship law reviews at ABA-accredited institutions still insist on publishing in print, despite the massive decline in print subscribers and economic and environmental waste.

The availability of a law journal in print format remains a superficial indicator of prestige and quality to law professors, student editors, and law school administrations. A shift from print publication to electronic-only publication is not as simple as having a law journal merely cancel its print runs, but rather requires several fundamental changes to the publication process.

Many law journals must also greatly improve their websites before electronic-only publication can truly replace print publication. The Durham Statement was drafted by law library directors from top law schools across the country.

Law librarians today must assist in facilitating the transition if we ever expect to see a world of electronic-only publication of law journals.

This paper argues that the Harvard Law Review, Yale Law Journal, and Stanford Law Review must be the first law reviews to transition to electronic-only publication, after which other law journals will follow suit.

URL : http://ssrn.com/abstract=2785307

The Miracle of Peer Review and Development in Science: An Agent-Based Model

Posté le par

Authors : Simone Righi, Károly Takács

It is not easy to rationalize how peer review, as the current grassroots of science, can work based on voluntary contributions of reviewers. There is no rationale to write impartial and thorough evaluations.

Consequently, there is no risk in submitting low-quality work by authors. As a result, scientists face a social dilemma: if everyone acts according to his or her own self-interest, low scientific quality is produced. Still, in practice, reviewers as well as authors invest high effort in reviews and submissions.

We examine how the increased relevance of public good benefits (journal impact factor), the editorial policy of handling incoming reviews, and the acceptance decisions that take into account reputational information can help the evolution of high-quality contributions from authors.

High effort from the side of reviewers is problematic even if authors cooperate: reviewers are still best off by producing low-quality reviews, which does not hinder scientific development, just adds random noise and unnecessary costs to it.

We show with agent-based simulations that tacit agreements between authors that are based on reciprocity might decrease these costs, but does not result in superior scientific quality. Our study underlines why certain self-emerged current practices, such as the increased importance of journal metrics, the reputation-based selection of reviewers, and the reputation bias in acceptance work efficiently for scientific development.

Our results find no answers, however, how the system of peer review with impartial and thorough evaluations could be sustainable jointly with rapid scientific development.

URL : http://arxiv.org/abs/1607.02574

Science Concierge: A Fast Content-Based Recommendation System for Scientific Publications

Posté le par

Authors : Titipat Achakulvisut, Daniel E. Acuna, Tulakan Ruangrong , Konrad Kording

Finding relevant publications is important for scientists who have to cope with exponentially increasing numbers of scholarly material. Algorithms can help with this task as they help for music, movie, and product recommendations.

However, we know little about the performance of these algorithms with scholarly material. Here, we develop an algorithm, and an accompanying Python library, that implements a recommendation system based on the content of articles.

Design principles are to adapt to new content, provide near-real time suggestions, and be open source. We tested the library on 15K posters from the Society of Neuroscience Conference 2015.

Human curated topics are used to cross validate parameters in the algorithm and produce a similarity metric that maximally correlates with human judgments. We show that our algorithm significantly outperformed suggestions based on keywords.

The work presented here promises to make the exploration of scholarly material faster and more accurate.

URL : Science Concierge: A Fast Content-Based Recommendation System for Scientific Publications

DOI : http://dx.doi.org/10.1371/journal.pone.0158423

Men set their own cites high: Gender and self-citation across fields and over time

Posté le par

Authors : Molly M. King, Carl T. Bergstrom, Shelley J. Correll, Jennifer Jacquet, Jevin D. West

How common is self-citation in scholarly publication and does the practice vary by gender? Using novel methods and a dataset of 1.5 million research papers in the scholarly database JSTOR published between 1779-2011, we find that nearly 10% of references are self-citations by a paper’s authors.

We further find that over the years between 1779-2011, men cite their own papers 56% more than women do. In the last two decades of our data, men self-cite 70% more than women. Women are also more than ten percentage points more likely than men to not cite their own previous work at all.

Despite increased representation of women in academia, this gender gap in self-citation rates has remained stable over the last 50 years. We break down self-citation patterns by academic field and number of authors, and comment on potential mechanisms behind these observations.

These findings have important implications for scholarly visibility and likely consequences for academic careers.

URL : https://arxiv.org/abs/1607.00376