Computational social science: On measurement

doi:10.1126/science.abe8308

GSTDTAP > 气候变化

DOI	10.1126/science.abe8308
	Computational social science: On measurement
	Angela Xiao Wu; Harsh Taneja; danah boyd; Paul Donato; Matthew Hindman; Philip Napoli; James Webster
	2020-12-04
发表期刊	Science
出版年	2020
英文摘要	In their Policy Forum “Computational social science: Obstacles and opportunities” (28 August, p. [1060][1]), D. M. J. Lazer et al. propose ethical data infrastructures for computational social science research. Concentrating on access to platform trace data, they dismiss third-party market data from such companies as Nielsen and comScore because of “opaque” methods and high cost. We believe both have virtues, but their proper use requires a keener appreciation of each measurement regime. All data result from measurement processes designed and executed to serve a given institutional context ([ 1 ][2], [ 2 ][3]). Platforms profit from shaping usage and they measure toward that end. Using their trace data to understand human conduct remains problematic as long as platforms are themselves opaque about their methods for managing user behavior ([ 3 ][4]). Social Science One and Twitter's COVID-19 application programming interface may be productive precedents of platform data provision, but computational social science should reckon with the effects of platform measurement. Unlike platforms, third-party measurement firms are not invested in how users behave. As with public-sector data (such as the U.S. Census), third-party measurement is periodically audited ([ 4 ][5]). Its procedures and consequences are constantly appraised by actors with competing interests ([ 5 ][6]). Serving industries, policy-makers, and academics, third-party market research has invested for decades in refining what Lazer et al. aspire to: “an administrative infrastructure… enforcing compliance with privacy and ethics rules,” which aligns “with critical research norms” including “transparency, reproducibility, replication, and consent” ([ 3 ][4], [ 6 ][7], [ 7 ][8]). Third-party measurement firms such as Nielsen and comScore supply data to a broad subscriber base of advertising agencies and content publishers, which lowers data costs. Academic institutions worldwide may access numerous such third-party datasets via Wharton Research Data Services and Chicago Booth, brokers that partner with third-party firms for this purpose. Meanwhile, public data can be cost prohibitive (such as CDC's National Death Index). What ensures data's “public accountability” is not a public-sector origin but how the measurement regime is institutionally arranged ([ 3 ][4]). In addition to expanding data collaborations and data infrastructures, attention to the measurement regimes of “found data” and reflexive triangulation across data sources are indispensable to development of computational social science. 1. [↵][9]1. T. M. Porter , Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton University Press, Princeton, NJ, 1995). 2. [↵][10]1. W. N. Espeland, 2. M. L. Stevens , Arch. Eur. Sociol. 49, 401 (2008). [OpenUrl][11] 3. [↵][12]1. A. X. Wu, 2. H. Taneja , New Media Society 10.1177/1461444820933547 (2020). 4. [↵][13]1. P. M. Napoli, 2. A. B. Napoli , First Monday 24, 10.5210/fm.v24i12.10124 (2019). 5. [↵][14]1. N. Anand, 2. R. A. Peterson , Org. Sci. 11, 270 (2000). [OpenUrl][15] 6. [↵][16]Advertising Research Foundation, Member Code of Conduct (2019); . 7. [↵][17]ESOMAR, The ICC/ESOMAR Code (2020); [www.esomar.org/what-we-do/code-guidelines][18]. [1]: http://www.sciencemag.org/content/369/6507/1060 [2]: #ref-1 [3]: #ref-2 [4]: #ref-3 [5]: #ref-4 [6]: #ref-5 [7]: #ref-6 [8]: #ref-7 [9]: #xref-ref-1-1 "View reference 1 in text" [10]: #xref-ref-2-1 "View reference 2 in text" [11]: {openurl}?query=rft.jtitle%253DArch.%2BEur.%2BSociol.%26rft.volume%253D49%26rft.spage%253D401%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [12]: #xref-ref-3-1 "View reference 3 in text" [13]: #xref-ref-4-1 "View reference 4 in text" [14]: #xref-ref-5-1 "View reference 5 in text" [15]: {openurl}?query=rft.jtitle%253DOrg.%2BSci.%26rft.volume%253D11%26rft.spage%253D270%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [16]: #xref-ref-6-1 "View reference 6 in text" [17]: #xref-ref-7-1 "View reference 7 in text" [18]: http://www.esomar.org/what-we-do/code-guidelines
领域	气候变化 ; 资源环境
URL	查看原文
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.173/C666/handle/2XK7JSWQ/305816
专题	气候变化资源环境科学
推荐引用方式 GB/T 7714	Angela Xiao Wu,Harsh Taneja,danah boyd,et al. Computational social science: On measurement[J]. Science,2020.
APA	Angela Xiao Wu.,Harsh Taneja.,danah boyd.,Paul Donato.,Matthew Hindman.,...&James Webster.(2020).Computational social science: On measurement.Science.
MLA	Angela Xiao Wu,et al."Computational social science: On measurement".Science (2020).