Search

Purpose/Thesis: The article compares several statistical methods, which can be used to forecast the rate of the evolution of scientific disciplines. The data sample comprised the citations of specific scientific publications. The article emphasized the possibility of generalization of the results yielded by the analysis of this random sample. It also highlighted the limitations of each forecasting method, and proposed rough solutions to these problems. Approach/Methods: The authors used a data set comprising almost 25 thousands of citations. They applied several distinct statistical methods inspired by econometric models. These were polynomial regression method, regression method with a correction for the autocorrelation in the residual components, an autoregression method, an autoregression method with the correction for non-stationarity of modelled process and Holt's adaptive model of exponential smoothing. The regression methods were tested for the fulfillment of the Gauss-Markov conditions. Moreover, common accuracy measures, as well as the prognosis errors coefficients were calculated and compared for all the methods applied. Results and conclusions: The analysis showed that the most precise method was the polynomial regression method with a correction for the autocorrelation in residual components. The reliability of the autoregression method is comparable with that of the regression methods. The adaptive exponential smoothing method yielded ambiguous results. This suggests directions for further research. Research limitations: The basic limitation of this study was the range of empirical data available to the authors, which was restricted to a single scientific discipline, and, further, limited to Polish-language texts published in journals (periodicals). Originality/Value: The originality of the study lies in the innovative juxtaposition of the well-known quantitative methods, which have not been used to predict the rate of the development of scientific disciplines before. Secondly, the study shows their potential in this field of inquiry, and makes clear the need for further improvement. The study identifies the most promising methodology, which may open the way for the better understanding of science's internal dynamics. [ABSTRACT FROM AUTHOR]

Purpose/Thesis: The study presents an overview and a comparison of several selected statistical methods basing on citations obtained from publications belonging to the selected disciplines. Furthermore, the study analyzed the requirements for and obstacles to generalizing quantitative results yielded on the basis on the random samples. Approach/Methods: On the basis of the data set comprising of almost 25 thousands of citations, the authors have shown a method of establishing confidence intervals for the citing half-life indexes; then the data was subjected to a statistical study using nonlinear regression method, linearized regression method, and a second-degree polynomial regression. Results and conclusions: The central difficulty of applying these methods was their failure to fulfill some of the Gauss-Markov conditions. It is necessary to correct the models applied, or to make use of statistical techniques of a different kind, which suggests future research directions. Originality/Value: The originality of the presented overview lies in the novel juxtaposition of the quantitative methods, which, although well-known, are not commonly used to forecast the rate of the development of scientific disciplines The study highlighted their potential and expected usefulness in this regard, as well as the need of further improvement or testing other, more sophisticated methods. [ABSTRACT FROM AUTHOR]

Purpose/Thesis: Ihe aim of the paper is to provide an evaluative review of literature on the quantitative methods of bibliometric forecasting and assessing the current state of scientific discipline development. By "evaluative" review it is understood that the author discussed and pinpointed the drawbacks of the literature-proposed ways to assess and predict the development of scientific disciplines. Approach/Methods: Ihe method used by the author is the analysis of the literature concerning the problems under consideration. Ihe author also proposed a typology of methods applied in the bibliometric predictions area and emphasized the flaws of some previously used methods. Moreover, the possibilities of practical application of foresight exercises were shown. Finally, the conclusions were drawn, with regard to the methods oriented towards the assessment of the rate and direction of scientific disciplines development. Results and conclusions: First conclusion was that there was a great diversity within the sphere of quantitative methods concerning the subject of evolution forecasting for scientific disciplines. The methods in question often are applied without referring to each other. In other words, the majority of predictive methods are used in isolation. The flaws, occurring in many methods, usually have a negative influence on the results obtained. Next conclusion was that there was a significant potential in the statistical analysis of the time series. However, it seems that this potential has not been fully exploited yet. Furthermore, there is a considerable degree of interdisciplinarity in the scope of bibliometric predictive methods. Often, they are modified and adopted from other disciplines such as mathematics, computer science, statistics and econometrics. The last conclusion was that the predictive process was based either on purposefully designed citation analysis or solely on the number of disciplinary publications. In the first case the methodology must implicitly accept the theses of so-called normative paradigm, existing within the sociology of scientific knowledge and constituting an epistemological background for the methodology. In the latter case the methodology is engaged in the debate over topical relevance, which is observed within the frames of library and information science. Originality/Value: Ihe originality and value of the paper lies in an effort to demonstrate the scope and a high degree of progress of the studies on the possibilities of predicting and assessing the development of scientific disciplines worldwide. In Polish scientific environment such studies are relatively rare, and it seems that the review of the predictive bibliometric methodology could contribute to the stimulation of interest in the research directed towards the forecasting of future scientific trends. It is also important to stress the disadvantages of selected methods of bibliometric forecasting as in this context they create a substantial amount of space destined for possible improvements and modifications of existing methods. In the field of bibliometrics there is a considerable need for such refinements due to the fact that currently there is no perfect method which could be described as "the last word" on the technique of anticipating future research directions. [ABSTRACT FROM AUTHOR]

Purpose/Thesis: The aim of this paper is to: describe the basics of statistical time series analysis, investigate the possibility of using this approach in short-term forecasting of the development of scientific disciplines based on data showing the citedness of research papers, assess the uncertainty of the forecasts (feature available with this method) and read the results of the research in terms of science studies. Approach/Methods: The analytical method of time series decomposition was used to isolate so-called development trends based on the observations of the dynamics of mass phenomena. In this case, the mass phenomenon was the citedness of scientific literature in the field of Polish earth science. The method consists of the formulation of an equation which quantitatively describes the course of the phenomenon and distinguishes the main trend from the accidental variability (fluctuations). The equation serves as a keystone in the process of predicting future values of the time series. Results and conclusions: The results confirm the possibility of an effective application of the approach in short-term forecasting of the scientific discipline development rate. In the case of the research in question, the forecasts proved acceptable as regards two-year horizon of the forecast. However, it is assumed that building the forecast in the form of confidence intervals or slightly modifying the approach could contribute to the lengthening of the horizon. The results were read as a simplified immediacy index, available in particular for comparative applications. Originality/Value: The authors were unable to find any examples of research which was conducted in a manner presented in this paper. Time series analysis was earlier applied to so-called clusters only, which are a result of dividing disciplinary publications into narrow groups. The groups are usually considered to be representative for certain scientific subdisciplines. Clustering is always done on the basis of one out of many available similarity criteria, in particular the one called a co-word similarity. Furthermore, there is a wide variety of clustering algorithms, which typically yield considerably differing results. As a result, it seems that the approach based on direct citedness of scientific publications helps to eliminate the redundant vagueness and the inevitable lack of universality which appear when one analyzes the results of an algorithmic clustering. [ABSTRACT FROM AUTHOR]

Purpose/Thesis: The aim of the article is to provide an evaluative review of a literature on the quantitative methods of bibliometric forecasting and assessing the current state of scientific disciplines development. By "evaluative" review it is understood that the author discussed and pinpointed the drawbacks of the literature-proposed ways to assess and predict the development of scientific disciplines. Approach/Methods: The method used by the author is the analysis of the literature concerning the problems under consideration. The author also proposed a typology of methods applied in the bibliometric predictions area and emphasized the flaws of some previously used methods. Moreover, the possibilities of practical applications of foresight exercises were shown. Finally, the conclusions were drown with regard to the methods oriented towards the assessment of the rate and direction of scientific disciplines development. Results and conclusions: Taking into consideration the fact that this article is first part of a larger whole comprising the analysis of the literature in question, the conclusions drawn within the first part have only a tentative character. It was emphasized that the methods proposed in the trailblazing publications are still vital and fulfill a leading role in bibliometrics. Second, it was indicated that the drawbacks accompanying the majority of mentioned methods are an integral part of the evolution of the research mainstream which is being discussed. 'Third, it is necessary to note that there is a strong correlation between bibliometric predictive methods and an algorithmic means of clustering groups of publications into topic consistent subgroups. These subgroups are usually considered to represent disciplinary subspecialties or, in other words, disciplinary problem areas. Originality/Value: The originality and value of the article lie in an effort to demonstrate the scope and a high degree of progress of the studies on the possibilities of prediction and assessment of the development of scientific disciplines worldwide. In the Polish scientific environment such studies are relatively rare and it seems that the review of the predictive bibliometric methodology could contribute to the stimulation of interest in the research directed towards the forecasting of future scientific trends. It is also important to stress the disadvantages of particular methods of bibliometric forecasting as in this context they create a substantial amount of space destined for possible improvements and modifications of existing methods. In the field of bibliometrics there is a need for such refinements due to the fact that currently there is no ideal method which could be described as "the last word" on the technique of anticipating future research directions. [ABSTRACT FROM AUTHOR]