Schubert, A. (2002). The web of scientometrics: A statistical overview of the first 50 volumes of the journal. Scientometrics, 53(1), 3-20.

Schubert, A. (2002). The web of scientometrics: A statistical overview of the first 50 volumes of the journal. Scientometrics, 53(1), 3-20.

Schoepflin and Glänzel (2001)將Scientometrics期刊上的論文主題分為六類：1. 理論(包括書目計量學理論、數學模型和書目計量學定律的公式化)、2. 案例(案例研究和實務論文)、3. 方法學與應用、4. 指標 (指標工程與資料呈現)、5. 社會學取向(書目計量學的社會學方法和科學社會學)以及6. 政策研究(包括科學政策、科學管理和一般與科技討論)等，並應用這樣的分類討論Scientometrics期刊的論文主題在1980、1989及1997等三個不同年份的比重變化。上述的主題可以在歸納成核心的書目計量學(core bibliometrics) (2, 3, 4)和背景研究(background research) (1, 5, 6)或是理論研究(1, 3, 5)和應用研究(2, 4, 6)。Schoepflin and Glänzel (2001)的研究發現案例與方法學有明顯與穩定的成長，但社會學取向和科學政策的論文數量減少。

本研究以前50卷1443篇Scientometrics期刊的論文資料進行計量分析，這1443篇論文上包含來自60個國家的1223位作者，共包含25200個參考文獻(不重複的項目共16500個)，除了382篇論文不曾被引用外，1061篇曾被引用的論文共被引用7242次。

以合作關係的分析，55.1%的論文為單一作者，只有5.4%的論文由3位以上的作者合作完成，論文的平均作者數是1.61，過去十年來合作的情形略有增加，因此科學計量學領域比較像社會科學或數學。國際合作情形較罕見，只有 7%的Scientometrics論文由一個以上的國家合作完成，但後25卷的國際合作情形是前25卷的兩倍以上。

以Price指標(the Price index)分析Scientometrics論文的25200個參考文獻，了解其引用文獻在五年內出版的比例，藉以判斷該期刊是屬於軟科學(soft sciences)或硬科學(hard sciences)，結果發現在五年內出版的文獻大約占45%。其引用的期刊來源，除了圖書資訊學(Information Science & Library Science)、電腦科學(Computer Science)和跨領域應用(Interdisciplinary Applications)外，其餘多屬普通物理(general physics)、普通化學(general chemistry)、普通醫學(general medicine)等較廣泛的領域。但也可以發現有相當高比例(13.6%)是期刊自我引用的情形，就被引用的情形來看則高達47.3%是來自期刊本身。

A total of 1443 items were published by 1223 authors from 60 countries. They contained 25200 references to about 16500 different items ...1061 Scientometrics papers received 7242 citations during the 1978-2000 period (i.e., 382 papers remained uncited).

Activity Indexes show even more spectacularly the outstanding relative activity of a few countries (first of all Hungary, but also Belgium, Bulgaria, Chile, Mexico, Netherlands and, if measured against social science standards, then P.R. China, Poland, Russia and Spain, as well). A conspicuously low relative activity is shown by Italy and Japan (not so much if social science standards are concerned); interestingly, most English-speaking countries (US, England, Australia, Canada) exhibit a lower-than-average activity.

In their paper, Schoepflin and Glänzel (2001) classified the papers published in Scientometrics into six thematic categories, and studied the change in the weight of these categories by selecting three sample years: 1980, 1989 and 1997.
They used the following categories:
1. (THEO) Bibliometric theory, mathematical models and formalisation of bibliometric laws;
2. (CASE) Case studies and empirical papers;
3. (METH) Methodological papers including applications;
4. (INDI) Indicator engineering and data presentation;
5. (SOCI) Sociological approach to bibliometrics, sociology of science;
6. (POLI) Science policy, science management and general or technical discussions.

The classification permitted to group the material in several ways: the categories can be regarded from the viewpoint of core bibliometrics (2, 3, and 4) and background research (1, 5, and 6), but also with respect to theoretical (1, 3, and 5) and applied research (2, 4, and 6).

There are two obvious developments: an impressing and steady growth of case studies (Category 2) and methodology (Category 3) and the loss of position of articles on sociological (Category 5) and science policy (Category 6) issues.

In this respect, however, scientometrics resembles rather to the social sciences(or, maybe, mathematics) than to the sciences: 55.1% of the papers published in the first 50 volumes of Scientometrics is single-authored and only 5.4% of them are multi-authored (more than 3 authors). The average number of authors per paper is 1.61. In the past decade, nevertheless, there is a slight tendency of growing collaboration (see Figure 3).

International collaboration is even less favourised in the scientometrics community. There is a modest 7% of Scientometrics papers having more than one country in the authors’ affiliation section in the by-line of the publication. Nevertheless, the tendency is unambiguous: the fraction of internationally co-authored papers more than doubled from the first to the second 25 volumes.

The 25200 references of the papers form, as it were, the intellectual “hinterland” of research reported in Scientometrics. Clearly, they constitute a vast treasury of information about the history, sociology, epistemology of the field and its journal, and several attempts were made to make the most of this information (Schubert and Maczelka, 1993; Wouters and Leydesdorff, 1994; Schoepflin and Glänzel, 2001).

De Solla Price (1970) introduced an index, later named after him, with the aim of distinguishing between “harder” and “softer” sciences. The Price Index is defined as the percentage share of references to items not older than five years at the time of publishing the citing paper. Typical “soft science” journals (German Review, American Literature, Studies in English Literature, Isis, in Price’s original study) have an index value less than 10%, while, e.g., some research front physics journals may reach 80%.

The Price Index of the journal Scientometrics is around 45%, i.e., it occupies a medium position on the hardness scale.

Among the highly cited sources there is a clear dominance of SCI/SSCI-covered titles: 19 of the 25 titles belongs to this category. References to the journal itself (journal self-references) constitute 13.6% of all references – it is a typical value for a consolidated primary journal.

The journal Scientometrics itself is categorised into two subfields: Information Science & Library Science and Computer Science, Interdisciplinary Applications.

It can be seen that Information Science & Library Science would remain the main source of information for the journal even if self-references were disregarded, while Computer Science, Interdisciplinary Applications would disappear from the chart without them.

Among science journals, those from broader-scope subfields (general physics, general chemistry, general medicine, multidisciplinary sciences) contribute the most to the reference base of Scientometrics. The presence of Analytical Chemistry among the top cited fields may be connected with the fact that this is the original and main fireld of the Editor-in-Chief of the journal.

The journal self-citation rate of 47.3% is rather high, particularly if compared with the self-reference rate of 13.6%. It indicated that the “outside world” pays less attention to the journal than vice versa.