Studying 5.6 million biomedical science articles published over three decades, we reconcile conflicts in a long-standing interdisciplinary literature on scientists' life-cycle productivity by controlling for selective attrition and distinguishing between research quantity and quality. While research quality declines monotonically over the career, this decline is easily overlooked because higher ability authors have longer publishing careers. Our results have implications for broader questions of human capital accumulation over the career and federal research policies that shift funding to early-career researchers-while funding researchers at their most creative, these policies must be undertaken carefully because young researchers are less able on average.

Using unique, new, matched UMETRICS data on people employed on research projects and Author-ity data on biomedical publications, this paper shows that National Institutes of Health funding stimulates research by supporting the teams that conduct it. While faculty-both principal investigators (PIs) and other faculty-and their productivity are heavily affected by funding, so are trainees and staff. The largest effects of funding on research output are ripple effects on publications that do not include PIs. While funders focus on research output from projects, they would be well advised to consider how funding ripples through the wide range of people, including trainees and staff, employed on projects.

We study the effects of peer gender composition in STEM doctoral programs on persistence and degree completion. Leveraging unique new data and quasi-random variation in gender composition across cohorts within programs, we show that women entering cohorts with no female peers are 11.7 percentage points less likely to graduate within 6 years than their male counterparts. A 1 standard deviation increase in the percentage of female students differentially increases women’s probability of on-time graduation by 4.4 percentage points. These gender peer effects function primarily through changes in the probability of dropping out in a PhD program’s first year. © 2022 The University of Chicago. All rights reserved.

In several author name disambiguation studies, some ethnic name groups such as East Asian names are reported to be more difficult to disambiguate than others. This implies that disambiguation approaches might be improved if ethnic name groups are distinguished before disambiguation. We explore the potential of ethnic name partitioning by comparing performance of four machine learning algorithms trained and tested on the entire data or specifically on individual name groups. Results show that ethnicity-based name partitioning can substantially improve disambiguation performance because the individual models are better suited for their respective name group. The improvements occur across all ethnic name groups with different magnitudes. Performance gains in predicting matched name pairs outweigh losses in predicting nonmatched pairs. Feature (e.g., coauthor name) similarities of name pairs vary across ethnic name groups. Such differences may enable the development of ethnicity-specific feature weights to improve prediction for specific ethic name categories. These findings are observed for three labeled data with a natural distribution of problem sizes as well as one in which all ethnic name groups are controlled for the same sizes of ambiguous names. This study is expected to motive scholars to group author names based on ethnicity prior to disambiguation.

How can we evaluate the performance of a disambiguation method implemented on big bibliographic data? This study suggests that the open researcher profile system, ORCID, can be used as an authority source to label name instances at scale. This study demonstrates the potential by evaluating the disambiguation performances of Author-ity2009 (which algorithmically disambiguates author names in MEDLINE) using 3 million name instances that are automatically labeled through linkage to 5 million ORCID researcher profiles. Results show that although ORCID-linked labeled data do not effectively represent the population of name instances in Author-ity2009, they do effectively capture the 'high precision over high recall' performances of Author-ity2009. In addition, ORCID-linked labeled data can provide nuanced details about the Author-ity2009's performance when name instances are evaluated within and across ethnicity categories. As ORCID continues to be expanded to include more researchers, labeled data via ORCID-linkage can be improved in representing the population of a whole disambiguated data and updated on a regular basis. This can benefit author name disambiguation researchers and practitioners who need large-scale labeled data but lack resources for manual labeling or access to other authority sources for linkage-based labeling. The ORCID-linked labeled data for Author-ity2009 are publicly available for validation and reuse.

Attempts to find central influencers, opinion leaders, hubs, optimal seeds, or other important people who can hasten or slow diffusion or social contagion has long been a major research question in network science. We demonstrate that opinion leadership occurs only under conventional but implausible scope conditions. We demonstrate that a highly central node is a more effective seed for diffusion than a random node if nodes can only learn via the network. However, actors are also subject to external influences such as mass media and advertising. We find that diffusion is noticeably faster when it begins with a high centrality node, but that this advantage only occurs in the region of parameter space where external influence is constrained to zero and collapses catastrophically even at minimal levels of external influence. Importantly, nearly all prior agent-based research on choosing a seed or seeds implicitly occurs in the network influence only region of parameter space. We demonstrate this effect using preferential attachment, small world, and several empirical networks. These networks vary in how large the baseline opinion leadership effect is, but in all of them it collapses with the introduction of external influence. This implies that, in marketing and public health, advertising broadly may be underrated as a strategy for promoting network-based diffusion.

Machine learning for author name disambiguation is usually conducted on the training and test subsets of labeled data created for a specific task. As a result, disambiguation models learned on heterogeneous labeled data are often inapplicable for other purposes that either do not use the same labeled data or do not make use of any labeled data at all. This article explores the idea of transfer learning in a new context, author name disambiguation. We focus on cases where a disambiguation task lacking labeled training data uses models trained on labeled data generated for other tasks. For this purpose, two labeled source datasets are used for training of disambiguation models to be applied to three test target datasets that are deficient of labeled training data. Our results show that transfer learning can produce disambiguation performances similar to those achievable by traditional machine learning in which training and test datasets come from the same labeled data source. The good performance through transfer learning are possible when training source datasets have similar feature distributions as test target datasets. This study suggests that through transfer learning, rich disambiguation models in previous studies can be retained and reused across ambiguous bibliographic data from different fields and data sources, motivating further research on how to correct feature distribution differences between source and target datasets to expand the application of transfer learning in author name disambiguation beyond the model sharing explored in this research.

y Conference publications in computer science (CS) have attracted scholarly attention due to their unique status as a main research outlet, unlike other science fields where journals are dominantly used for communicating research findings. One frequent research question has been how different conference and journal publications are, considering an article as a unit of analysis. This study takes an author-based approach to analyze the publishing patterns of 517,763 scholars who have ever published both in CS conferences and journals for the last 57 years, as recorded in DBLP. The analysis shows that the majority of CS scholars tend to make their scholarly debut, publish more articles, and collaborate with more coauthors in conferences than in journals. Importantly, conference articles seem to serve as a distinct channel of scholarly communication, not a mere preceding step to journal publications: coauthors and title words of authors across conferences and journals tend not to overlap much. This study corroborates findings of previous studies on this topic from a distinctive perspective and suggests that conference authorship in CS calls for more special attention from scholars and administrators outside CS who have focused on journal publications to mine authorship data and evaluate scholarly performance.

We identify two polar life cycles of scholarly creativity among Nobel laureate economists with Tinbergen falling broadly in the middle. Experimental innovators work inductively, accumulating knowledge from experience. Conceptual innovators work deductively, applying abstract principles. Innovators whose work is more conceptual do their most important work earlier in their careers than those whose work is more experimental. Our estimates imply that the probability that the most conceptual laureate publishes his single best work peaks at age 25 compared to the mid-50 s for the most experimental laureate. Thus, while experience benefits experimental innovators, newness to a field benefits conceptual innovators.

To train algorithms for supervised author name disambiguation, many studies have relied on hand-labeled truth data that are very laborious to generate. This paper shows that labeled data can be automatically generated using information features such as email address, coauthor names, and cited references that are available from publication records. For this purpose, high-precision rules for matching name instances on each feature are decided using an external-authority database. Then, selected name instances in target ambiguous data go through the process of pairwise matching based on the rules. Next, they are merged into clusters by a generic entity resolution algorithm. The clustering procedure is repeated over other features until further merging is impossible. Tested on 26 K instances out of the population of 228K author name instances, this iterative clustering produced accurately labeled data with pairwise F1=0.99. The labeled data represented the population data in terms of name ethnicity and co-disambiguating name group size distributions. In addition, trained on the labeled data, machine learning algorithms disambiguated 24K names in test data with performance of pairwise F1=0.90-0.92. Several challenges are discussed for applying this method to resolving author name ambiguity in large-scale scholarly data.