Quantitative Aspects of Journal of Quantitative Linguistics

Abstract

As one of the most authoritative journals in the field of quantitative linguistics, the Journal of Quantitative Linguistics has just passed its 20th birthday. In this paper, we collect information from the 20 volumes of this journal published so far (1994–2013) to build a corpus, then extract different categories of information using corpus tools. By quantitative analysis of the corpus data, complemented with information extracted from abstract and citation index databases such as Scopus and Web of Knowledge as important references, we aim to explore the objects, aims and methodologies of quantitative linguistics, its focuses and shifts in different periods, representative achievements in the past 20 years, and possible directions for the future. During the process, the role of JQL both in the development of quantitative linguistics and the wider scientific community are located and evaluated with quantitative-based evidence. The findings are that JQL contributes greatly to the development of quantitative linguistics and is in many aspects representative of changes in this field, keeping close pace with trends in the discipline and the ever-evolving characteristics of quantitative linguistics over the years.

Acknowledgements

This work is supported by the National Social Science Foundation of China (11 & ZD 188).

Notes

²² The h-index for the 1391 documents is 56 calculated by Scopus. We just list the most cited 10 documents and their source.

²⁵ This list is organized with information retrieved from WOK on 24th December 2013, which may vary as time goes on.

¹ From http://www.iqla.org/journal/index.html.

² When constructing the corpus, we met the following three obstacles: first, mixed types of articles in the Column “Miscellany”. This column existed from volume (issue 1) in 1994 to volume 5 (issue 3) in 1998, then sparsely appeared in volume 13 (the double issue 2–3) in 2006 and volume 17 (issue 4) in 2010, in which information about the editorial board, newly published books, newsletters of conferences, reports of projects and discussions on particular academic issues were provided; second, some findings related to Projects were not correctly grouped into the column “Project Report”. This column was previously included in Miscellany, and became independent from volume 3 (issue 3) in 1996, but functions little; third, 12 research articles do not have the component of “abstract”. Our solutions are that articles with academic inclinations in both “Miscellany” and “Project Report” are included in our corpus as research articles, as they too counted within the number of documents in Scopus and Web of Knowledge. For those articles without abstracts, we either choose the first or last paragraph, or summarize the main idea to extract abstracts.

³ Because many QL terms are composed of several words, the analysis of adjacent words will help us understand the specific research questions.

⁴ Scopus does not provide citation information for JQL before 1996, but we still regard it as important references for the following three reasons: first, the missing information from 1994 to 1995, we assume, may not influence the panoramic descriptions of the QL development too much; second, information since 1996 can still be employed for valid corroboration with our analysis in the latter four periods (from 1998 to 2013 in our classification); last but not least, information retrieved from Scopus can consolidate the scientific evaluation of the role of JQL in the wider academic community, because important indices, like the most cited articles, their citation frequency, as well as analysis of source journals and their subject areas, can only be reliably obtained from a large-scale database like Scopus.

⁵ Book reviews, totalling 29, are only qualitatively described because of their small portion of the total.

⁸ It means that word length is an important parameter in the theory of synergetic linguistics.

⁹ The empirical corroboration of the Göttingen project’s theoretical hypothesis has been documented in the book Häufigkeitsverteilungen in Texten (Frequency Distributions in Texts) (Best, 2001a).

¹⁰ The theoretical elaboration of the Göttingen project may also refer to other important literature, “The theory of word length distribution: Some results and generalisations” by Wimmer and Altmann (1996) in Glottometrika.

¹¹ Related studies of the Graz Project on various Slavic texts are compiled in the book Contributions to the Science of Text and Language. Word length Studies and Related Issues (Grzybek, 2007)

¹² This is the pioneering study adopting the Treebank approach to investigate valency features in JQL.

¹³ Studies related to Chinese in the previous periods are: length of Chinese words in relation to their other systemic features (Breiter, 1994); design and implementation of a Chinese financial invoice recognition system (Ming et al., 2002); an approach to off-line handwritten Chinese character recognition based on hierarchical radical decomposition (Shi et al., 2003); the networks of syllables and characters in Chinese (Peng et al., 2008), and probability distribution of dependencies based on a Chinese dependency Treebank (Liu, 2009).

¹⁴ Our analysis is based on the home institution of authors, which may not represent their countries of origin.

¹⁵ The number of research articles from China has included four from Hong Kong and two from Taiwan.

¹⁶ Altmann’s cooperation work of the foreword on the special Issue: Festschrift in honour of Professor Juhan Tuldava (1997) and his reviews on Statistics for Corpus Linguistics (1999), Dynamique du vocabulaire des Fleurs du mal (2000), and Frequency and the Emergence of Linguistic Structure (2004) have not been included, which would be 18 publications if added.

¹⁷ This is a means of expressing attitude in Systemic Functional Grammar.

¹⁹ This figure is retrieved on 24th December 2013, which may vary as time goes.

²⁰ For the sake of illustration, we just snapshot the top three journals.

²¹ From http://www.scopus.com/term/analyzer, we just list the four top most important subject areas.

²⁶ From http://wcs.webofknowledge.com/RA/analyze.do. retrieved on 24th December 2013.

²⁷ From http://wcs.webofknowledge.com/RA/analyze.do, retrieved on 24th December 2013.

³⁰ The latest research article without the component of “abstract” appears in the double issue (2–3) of volume 14 in 2007 – “Becoming Jack London”, which is hardly understandable.

⁶ The number of publications in our analysis in the Second and Fifth Period correspond with that of Scopus, but not the Third and Fourth Period. This is due to the criteria of including documents by Scopus. Scopus includes two pieces of information (on editorial and symposium respectively) in the years 2002–2005, which are excluded from our corpus; but it excludes four research articles in issue 4 of volume 16 in 2009, which we include in ours, thus resulting in the divergence in the years 2006–2009.

⁷ In Figure 2 and the following figures, both singular and plural forms of words and adjacent words are included.

¹⁸ Fig. 12. is taken from Scopus, which is automatically analyzed and visually displayed by Scopus

²³ This figure was retrieved on 24th December 2013.

²⁴ One affiliation – Department of Linguistics and English Language was originally among the list from WOK, which we decided to delete because its documents are the same as those of Brigham Young University.

²⁸ From http://apps.webofknowledge.com/CitationReport, retrieved on Dec, 24, 2013.

²⁹ In Figure 22, the number of documents from China includes 2 documents of Taiwan.