International collaboration does not have greater epistemic authority

Research output: Contribution to journalArticleScientificpeer-review

Abstract

The consistent finding that internationally coauthored papers are more heavily cited has led to a tacit agreement among politicians and scientists that international collaboration in scientific research should be particularly promoted. However, existing studies of research collaboration suffer from a major weakness in that the Thomson Reuters Web of Science until recently did not link author names with affiliation addresses. The general approach has been to hierarchically code papers into international paper, national paper, or local paper based on the address information. This hierarchical coding scheme severely understates the level and contribution of local or national collaboration on an internationally coauthored paper. In this research, I code collaboration variables by hand checking each paper in the sample, use two measures of a paper's impact, and try several regression models. I find that both international collaboration and local collaboration are positively and significantly associated with a paper's impact, but international collaboration does not have more epistemic authority than local collaboration. This result suggests that previous findings based on hierarchical coding might be misleading.
Original languageEnglish
Pages (from-to)2151-2164
JournalJournal of the American Society for Information Science and Technology
Volume60
Issue number10
Publication statusPublished - 2009

Cite this

@article{f04538282ac14229b91fd081153fb9b5,
title = "International collaboration does not have greater epistemic authority",
abstract = "The consistent finding that internationally coauthored papers are more heavily cited has led to a tacit agreement among politicians and scientists that international collaboration in scientific research should be particularly promoted. However, existing studies of research collaboration suffer from a major weakness in that the Thomson Reuters Web of Science until recently did not link author names with affiliation addresses. The general approach has been to hierarchically code papers into international paper, national paper, or local paper based on the address information. This hierarchical coding scheme severely understates the level and contribution of local or national collaboration on an internationally coauthored paper. In this research, I code collaboration variables by hand checking each paper in the sample, use two measures of a paper's impact, and try several regression models. I find that both international collaboration and local collaboration are positively and significantly associated with a paper's impact, but international collaboration does not have more epistemic authority than local collaboration. This result suggests that previous findings based on hierarchical coding might be misleading.",
author = "Z.L. He",
year = "2009",
language = "English",
volume = "60",
pages = "2151--2164",
journal = "Journal of the American Society for Information Science and Technology",
issn = "1532-2882",
publisher = "John Wiley & Sons Inc.",
number = "10",

}

International collaboration does not have greater epistemic authority. / He, Z.L.

In: Journal of the American Society for Information Science and Technology, Vol. 60, No. 10, 2009, p. 2151-2164.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - International collaboration does not have greater epistemic authority

AU - He, Z.L.

PY - 2009

Y1 - 2009

N2 - The consistent finding that internationally coauthored papers are more heavily cited has led to a tacit agreement among politicians and scientists that international collaboration in scientific research should be particularly promoted. However, existing studies of research collaboration suffer from a major weakness in that the Thomson Reuters Web of Science until recently did not link author names with affiliation addresses. The general approach has been to hierarchically code papers into international paper, national paper, or local paper based on the address information. This hierarchical coding scheme severely understates the level and contribution of local or national collaboration on an internationally coauthored paper. In this research, I code collaboration variables by hand checking each paper in the sample, use two measures of a paper's impact, and try several regression models. I find that both international collaboration and local collaboration are positively and significantly associated with a paper's impact, but international collaboration does not have more epistemic authority than local collaboration. This result suggests that previous findings based on hierarchical coding might be misleading.

AB - The consistent finding that internationally coauthored papers are more heavily cited has led to a tacit agreement among politicians and scientists that international collaboration in scientific research should be particularly promoted. However, existing studies of research collaboration suffer from a major weakness in that the Thomson Reuters Web of Science until recently did not link author names with affiliation addresses. The general approach has been to hierarchically code papers into international paper, national paper, or local paper based on the address information. This hierarchical coding scheme severely understates the level and contribution of local or national collaboration on an internationally coauthored paper. In this research, I code collaboration variables by hand checking each paper in the sample, use two measures of a paper's impact, and try several regression models. I find that both international collaboration and local collaboration are positively and significantly associated with a paper's impact, but international collaboration does not have more epistemic authority than local collaboration. This result suggests that previous findings based on hierarchical coding might be misleading.

M3 - Article

VL - 60

SP - 2151

EP - 2164

JO - Journal of the American Society for Information Science and Technology

JF - Journal of the American Society for Information Science and Technology

SN - 1532-2882

IS - 10

ER -