社会科学领域引用最多的论著是哪些（基于Google Scholar的分析）?

Posted 2021-05-02 重建巴别塔

我们已经先后推出了几份书单，包括：

1.

2.

3.

4.

5.

6.

7.

以后也许还会整理一些其它的书单。

除书单外，我们还推送了一些读书心得，大家也可以参考，目前推送的读书笔记包括：

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

今天我们推送一篇文章，作者是伦敦政治经济学院副教授Elliott Green。他基于Google Scholar，分析了社会科学领域高被引文献及其分布特征。

Elliott Green首先选取了Google Scholar收录的被引次数超过20000的出版物，并将期刊论文和专著分开，因为不同领域的科研人员出版专著或者期刊论文的偏好和习惯有所差异。

随后，作者将文献分成了以下一些类别：人类学、经济学、教育学、地理学、语言学、管理学、哲学、政治学和心理学，并将自然科学和人文文献剔除。

在决定一个出版物是否属于社会科学的时候，作者查阅了原文作者的学位和机构以及期刊类别（如果是期刊论文的话）等信息。

顺便提及的是，Nature期刊曾在2014年发表了一篇论文，该论文研究了世界上被引量最高的100篇论文。然而，不同学科获得被引量有显著差异，相对于自然科学而言，社会科学的被引量一般较低。但这并不能说明社会科学的研究质量不如自然科学。

表1 社会科学领域中被引用最多的25本书籍

排序	书名	作者	出版年份	引用数	学科
1	The  Structure of Scientific Revolution 科学革命的结构	Thomas  Kuhn	1962	81311	哲学
2	Diffusion  of Innovation 创新的扩散	Everett  Rogers	1962	72780	社会学
3	Pedagogy  of the Oppressed 被压迫的教育学	Paulo  Freire	1968/1970	72359	教育学
4	Competitive  Strategy 竞争战略	Michael E  Porte	1980	65406	经济学
5	imagined  Communities 想像的共同体	Benedict Anderson	1983	64167	政治科学
6	Mind in  Society 社会中的心灵	LS  Vygotsky	1978	63809	心理学
7	Discipline and Punish 规训与惩罚	Michael  Foucault	1976/1977	60700	哲学
8	A theory of Justive正义论	John Rawls	1971	58594	政治科学
9	Social Foundations of Thought and Action思想和行为的社会基础:社会认知理论	Albert Bandura	1986	55324	心理学
10	The Interpretation of Cultures 文化的解释	Clifford Geertz	1973	48984	人类学
11	The History of Sexuality 性史（三卷本）	Michael  Foucault	1978-1986	47955	哲学
12	Situated  Learning: Legitimate Peripheral Participation情境学习: 合法的边缘性参与	Jean Lave和Etienne Wenger	1997	47627	教育学
13	The Fifth Discipline  第五项修炼	Peter M  Senge	1992	43876	管理学
14	Institution  ,Iinstitution Change and Economic Performance制度、制度变迁与经济绩效	Douglass CNorth	1990	43411	经济学
15	Culture's  Consequences: Comparing Values, Behaviors, Institutions and Organizations  Across Nations 文化的重要性	Geert Hofstede	1980	42144	管理学
16	The  presentation of the self in Everyday life 日常生活中的自我呈现	Erving Goffman	1959	40573	社会学
17	Das  Capital  资本论	Karl Max	1867-1894	40237	经济学
18	Distinction：A social Critique of the Judgment of  Taste 区隔	Pierre Bourdieu	1984	39729	社会学
19	The Social  Construction of Reality 现实的社会建构	Peter L. Berger 和Thomas Luckmann	1966	38845	社会学
20	Metaphors  we live by 我们赖以生存的隐喻	George Lakoff和 Mark Johnson	1980	38723	语言学
21	Stress ,Appraisal  and Coping 压力、评价与应对	Richard S Lazarus和Susan Folkman	1984	38665	心理学
22	Community  of Practice 实践社区	Etienne Wenger	1999	37775	心理学
23	The  Economic Institution of Capitalism 资本主义的经济制度	Oliver Williamson	1985	37651	经济学
24	Motivation  and Personality 激励与个性	Abraham Maslow	1954	37614	心理学
25	Attachment  依恋理论	John Bowlby	1969	37318	心理学

从上表中可以看出没有哪一门学科在这25本书中占绝对优势，前6本书就分别来自于社会科学中的不同学科。

如果把25扩展到前50本书，我们可以看一下各个学科中所占到的数目（如图1），其中经济学最多，占有11本，其次是社会学和心理学，分别是9本和8本。

图1 被引用最多的50本书籍不同学科的分布

如果以图书出版的年份来看，大部分的图书都出版于上个世纪的80和90年代（如图2）。其中21世纪只有一本书进入榜单，就是罗伯特.普特南（2001）的《独自打保龄球》。

除了这本以外，20年出版的书还有1999年阿玛蒂亚.森的以《自由看待发展》（1999）和爱丁纳·温格的《实践社区》（1999）。

在20世纪之前出版的书有2本，分别是《国富论》和《资本论》。

图2 被引用最多的50本书籍出版年代的分布

作者还分析了1950年之前出版的被引用最多的10本书籍（见表2），其中经济学占到了6本，社会学2本，心理学和教育学各1本。

表2 1950年之前出版的被引用最多的10本书籍

排序	书名	作者	出版年份	引用数	学科
1	Das Capital  资本论	Karl Max	1867-1894	40237	经济学
2	The Wealth of  Nations 国富论	Adam Smith	1776	36331	经济学
3	Capitalism,  Socialism and Democracy 资本主义、社会主义和民主	Joseph Schumpeter	1942	33025	经济学
4	Mind,Self and  Society 心灵、自我与社会	George Herbert Mead	1934	32721	社会学
5	The Thoery of  Economic Development 经济发展的理论	Joseph Schumpeter	1911/1934	31145	经济学
6	The General Theory  of Employment, Interest and Money就业、利息和货币通论	John Maynard Keynes	1936	29131	经济学
7	Theory of games and economic behavior 博弈论与经济行为	John von Neumann and  economist Oskar Morgenstern	1944	25969	经济学
8	The Protestant Ethic and the Spirit of Capitalism新教伦理与资本主义精神	Max Weber	1905/1930	25276	社会学
9	Democracy and Education民主与教育	John Dewey	1916	25159	教育学
10	The Organization of  Behavior: A Neuropsychological Theory行为组织：神经心理学理论	Donald.O Hobb	1949	22518	心理学

表3 被引用最多的10本方法类书籍

对于“方法论”方面的专著而言，其被引次数比学科领域的专业著作更高，最高可达13万多次。

表4 被引用最多的10篇期刊文章（不包括方法类文章）

参考文献

1.社会科学领域影响力最大的专著

http://toutiao.manqian.cn/wz_bHIue6oMSW9.html

2.Green, E. (2016). What are the most-cited publications in thesocial sciences (according to Google Scholar)? Retrieved from http://blogs.lse.ac.uk/impactofsocialsciences/2016/05/12/what-are-the-most-cited-publications-in-the-social-sciences-according-to-google-scholar/（见附录）

附录：What are the most-cited publications in the social sciences (according to Google Scholar)?

• 
  

Drawing on citation data that spans disciplines and time periods, Elliott Green has identified the most cited publications in the social sciences. Here he shares his findings on the 25 most cited books as well as the top ten journal articles. The sheer number of citations for these top cited publications is worth noting as is the fact that no one discipline dominates over the others in the top 20, with the top six books all from different disciplines.

There has been an increasing focus on citations as a measure of academic productivity in recent years, in part due to the increased ease of access to data from Google Scholar since it was launched in 2004. While there has thus been a large amount of interest in how scholars can obtain higher numbers of citations, there has been comparatively little attention to examining what publications actually obtain huge citation counts. There has been one study from Nature in 2014 of the top 100 most cited publications using data from Google Scholar, which included papers across all academic disciplines. Due to higher citation counts in the natural sciences, the list thus included very few publications from the social sciences (with only one spot in the top ten), of which a large number were actually methodology textbooks.

As such it is worth constructing a similar list, but just for the social sciences. I used Google Scholar to create a list of all publications which drew more than 20,000 citations, which I sub-divided into books and journal articles due to the fact that some disciplines tend to publish more in one format than the other, as well as methodology publications (including econometrics and statistics as well as qualitative methodology) due to the fact that methodology publications tend to be cited as if they were textbooks and can thus attract much higher citation counts than other publications.

Image credit: Archives New Zealand (CC BY-SA)

I included the following disciplines in my list: Anthropology, Economics, Education, Geography, Linguistics, Management, Philosophy, Political Science and Psychology. I thus excluded publications from both the humanities and the natural sciences, which in some borderline cases meant that I had to decide whether a publication belonged in the list due to its subject matter. In such cases I tended to decide based on the author’s degrees and/or professional title in order to maintain some degree of objectivity: thus I included Judith Butler’s Gender Trouble (1990) as she has a PhD in philosophy despite holding a professorship in a Literature department, but I did not include Homi Bhabha’s The Location of Culture (1994) as his PhD is in English literature. On the natural sciences side I similarly chose to exclude Lofti Zadeh’s 1965 article “Fuzzy Sets” despite its use in the social sciences as Zadeh holds a PhD in Engineering. In cases of books with multiple volumes such as Karl Marx’s Kapital I combined the citations for all volumes, and I confronted the problem of works published in multiple languages by only including citations of English-language editions for publications initially published in English, and citations of both the original language and English-language editions for works published initially in other languages. (In all but one cases this algorithm captured the two highest amounts of citations for foreign language publications. The one exception was Paulo Freire’s Pedagogy of the Oppressed (1968), whose Spanish-language edition has more citations than its original Portuguese-language edition.)

I begin with the list of the top 25 most published books in Table 1. Several things are immediately apparent. First is the sheer number of citations for the most-cited books: the most cited book, Thomas Kuhn’s bookThe Structure of Scientific Revolutions, has more citations that the entire body of work of such noted social scientists as John Stuart Mill (74,807 citations), Daniel Dennett (67,625), Avinash Dixit (61,112) or Steven Pinker (58,574). Secondly, no one discipline dominates over the others in the top 20, with the top six books all from different disciplines. However, if we expand our analysis to the top 50 books, as seen in Figure 1, economics dominates over other disciplines, followed by sociology and psychology. In contrast geography is totally absent from the list, with its highest-cited book, David Harvey’s The Condition of Postmodernity (1989), coming in at 65^th on the list, and no other book in anthropology has more than 20,000 citations other than Clifford Geertz’s The Interpretation of Cultures (1973).

Third, there is a distinct inverted-U shape curve to the date of publication, with books published between 1960 and the early 1990s dominating over both older and newer books. In Figure 2 I plot the distribution of date of publication by decade for books in the top 50, which shows that the 1980s and 1990s dominate over other decades. The only books in the top 50 published within the last 20 years are Robert Putnam’sBowling Alone (2001), Amartya Sen’s Development as Freedom (1999) and Etienne Wenger’sCommunities of Practice (1999); on the other end the only two books published more than a century ago are Adam Smith’s The Wealth of Nations (1776) and Karl Marx’s Das Kapital (published initially in German between 1867 and 1894).

Indeed, given the strong trend towards recent publications it is useful just to examine the top ten most cited books published before 1950, which I list in Table 2. As in the top 50 economics dominates over other disciplines, with six of the top seven spots; in contrast there are no books published in anthropology, geography, linguistics, management, or political science before 1950 with more than 20,000 citations.

* Books originally published in a foreign language are listed twice, with the English-language edition listed second.

It is useful to compare the list in Table 1 to a list of the top 10 most cited methodology books, which can be found in Table 3. As expected, the numbers of citations are incredibly high, such that merging Tables 1 and 3 would result in methodology books taking up half of the top 10. Another point to note here is the much better representation of female authors in Table 3, where three books have at least one female co-author and one, Using Multivariate Analysis (1989), is authored by two women. In contrast the highest non-methodology ranked book authored or co-authored by a woman is Stress, Appraisal and Coping (1989) by Richard Lazarus and Susan Folkman at #19, followed by Judith Butler’s Gender Trouble at #27.

Table 3: The 10 most cited methodology books in the social sciences

Another useful exercise is to compare the top ten books in Table 1 to the second-most cited books by the same authors as a measure of consistency. One would expect the ratio to start relatively high and then decline if the gap in citations between the second-most cited books was smaller than for the most-cited books, and indeed as expected the first three authors in based on the raking in Table 1, Thomas Kuhn, Everett Rogers and Paulo Freire, all have higher ratios than the last four on the list. However, the clear outlier here is Benedict Anderson, whose second-most cited publication, his 1998 book The Spectre of Comparisons, has only 979 citations, for a ratio of 68.6 to 1. One point to observe about Anderson’sImagined Communities is that, like some other books on the list, many if not most of its citations tend to be references to its title (i.e., that nations are “imagined communities”) rather than anything actually inside the book, leading one author to write about Imagined Communities that “rarely has a critical best-seller been so popular and so ignored at the same time.”

Finally, I examine in Table 4 the top 10 most cited journal articles, excluding methodological articles. (If such articles were included they would occupy three of the top six spots, starting with E. Kaplan and P. Meier’s 1958 article “Nonparametric Estimation from Incomplete Observations” from the Journal of the American Statistical Association.) Here the list is dominated by economics and psychology, with four publications each; in contrast the top ranked article in political science is R. Axelrod and WD Hamilton’s 1981 article “The Evolution of Cooperation” in Science at #22, with no articles from anthropology, education, geography or linguistics with more than 20,000 citations. Another thing to note is the relatively low numbers compared to Table 1, where there are seventeen publications with more than 40,000 citations compared to only four journal articles.

Note: This article gives the views of the author, and not the position of the LSE Impact blog, nor of the London School of Economics. Please review our Comments Policy if you have any concerns on posting a comment below.

About the Author

Elliott Green is Associate Professor of Development Studies in the Department of International Development at the LSE.  His research on ethnic politics, decentralization and development in Sub-Saharan Africa has been published recently in such journals as Comparative Politics, Economic Development and Cultural Change and International Studies Quarterly, among others.