In his book on “Documentation”, Bradford derived the law of scattering, based on algebric explanation with the supposition
that n1=n2=n. n1 and n2 are computed based on average no. of articles per journals in the first three zones. An analysis of a small sample of 12
data sets, using t-test suggests that it is unlikely that n1=n2. Further an attempt has been made to identify a suitable model to explain the law of scattering; among the various models
tried, log-normal fits much better than many models including the log-linear model.
Bibliometric studies are mostly empirical nature and they are mostly centred arround presentation of facts and data. There are very few studies which are centred arround theoretical foundation. The facts are gathered either through surveys or from published bibliographies, indexes, data bases. Based on these facts, empirical models and principles are being developed. The normative principles and standards have to evolve from the logical analyses of the empirical models. The stage is set to integrate empirical models of bibliometrics into standards. Future, bibliometrics studies have to address this issue and reach the stage of normative principles.
In this paper, growth models are classified and characterised using two types of growth rates: from time t to t+1 and from time t to 2t. They are interesting in themselves but can also be used for a quick prediction of the type of growth model that is valid in a particular case. These ideas are applied on 20 data sets collected byWolfram, Chu andLu. We determine (using the above classification as well as via nonlinear regression techniques) that the power model (with exponent>1) is the best growth model for Sci-Tech online databases, but that Gompertz-S-shaped distribution is the best for social sciences and humanities online databases.
An attempt has been made to trace and compare the trends in growth of Food Science and Technology (FST) literature (periodical
articles, patents, standards, theses and dissertations) produced by CFTRI scientists, by food scientists in India and by food
scientsts of the world, covering a period between 1950 and 1990; to identify the best fitting growth models for actual and
cumulative growth of data through regression analysis; and αt and α2t analysis; and to compute and compare the growth rates of FST documnets.
The cumulative distribution of the age of the most-recent-reference distribution is the “dual” variant of the first-citation
distribution. The latter has been modelled in previous publications of different authors but the former one has not. This
paper studies a model of this cumulative most-recent-reference distribution which is different from the first-citation distribution.
This model is checked on JASIS and JACS data, with success. The model involves the determination of 3 parameters and is a transformation of the lognormal distribution.
However we also show that the first-citation model (involving only 2 parameters and which is easier to handle), developed
in an earlier paper, gives enough freedom to give close fits to the most-recent-reference data as well.
In recent years, several projects were sponsored by NISSAT of the Goverment of India to map Indian Science. As a part of it,
a database (COMPENDEX) in engineering field was analysed. It has been found that engineers in India publish their articles
mostly in journals; almost all of them publish in English language. They publish in a selected few journals. Only a few of
the institutions are concentrated in engineering research. It has been observed that research output in applied physics, light
& optics, bioengineering and information science are increasing both at the world and India level. In the area of energy technology
metallurgical engineering and food technology, research output is decreasing at both levels.
Authors:Yuxian Liu, I. Ravichandra Rao, and Ronald Rousseau
Two types of series of h-indices for journals published in the field of Horticulture during the period 1998–2007 are calculated.
Type I h-indices are based on yearly data, while type II h-indices use cumulative data. These h-indices are also considered
in a form normalised with respect to the number of published articles.
It is observed that type I h-indices, normalised or not, decrease linearly over a period of ten years. The type II series,
however, is not linear in nature: it exhibits partly a concave shape. This proves that the journals (in Horticulture) do not
exhibit a linear increase in h-index as argued by Hirsch in the case of life-time achievements of scientists.
In the second part of the paper, an attempt is made to study the relative visibility of a journal and its change over time,
based on h-indices of journals. It is shown that:
the h-index over the complete period 1998–2007 of the journal Theoretical & Applied Genetics (h = 62) is much higher than that of all other journals in the field
the relation between the number of publications and the type II h-index for the whole period is not an exact power law (as
it would have to be if the Egghe-Rousseau model were applicable)
in order to study the dynamic aspects of journal visibility, a field-relative normalised h-ratio is defined to monitor systematic
changes in the field of Horticulture. Except for two journals, the Pearson correlation coefficient for yearly values of this
field-relative normalised h-ratio indicates that there is no systematic change of the performance of the journals with respect
to the field as a whole.
Authors:Leo Egghe, I. K. Ravichandra Rao, and Bibhuti Bhusan Sahoo
In a recent paper [H. F. MOED, E. GARFIELD: In basic science the percentage of “authoritative”references decreases as bibliographies
become shorter. Scientometrics 60 (3) (2004) 295-303] the authors show, experimentally, the validity of the statement in the
title of their paper. In this paper we give a general informetric proof of it, under certain natural conditions. The proof
is given both in the discrete and the continuous setting. An easy corollary of this result is that the fraction of non-authoritative
references increases as bibliographies become shorter. This finding is supported by a set of data of the journal Information
Processing and Management (2002 + 2003) with respect to the fraction of conference proceedings articles in reference lists.