Scientometric Analysis of Literature on Public Health Using Scopus Database

Abstract

Objectives Measuring global Public Health (PH) research output in a scientometric study is important because it reveals many facets of research including: priority areas, thriving fields, future growth of research output, values of research works, age of literature used and the information needs of researchers, scientists and subject experts, importance of different types of publication, the shape of development of a discipline at different times, regions etc. This study of the literature published on Public Health both quantitatively and qualitatively is the first of its kind. The main objective of the research is to assess the growth and development of public health literature globally and specifically within Bangladesh during the years 2000 to 2015 and review several components of scientometric study including the patterns of growth of literature, authorship, author collaboration and productivity, citation analysis and bibliometric laws. Methodology The current study is an exploratory research in nature reviewing secondary literature extracted from a bibliographic database and also analytical in nature with the application of suitable statistical and scientometric tools to strengthen the empirical validity. Various scientometric indicators including qualitative and quantitative indicators and a number of bibliometric laws were used at different levels of aggregation including micro, meso, and macro level. Scopus database was used as the source database to quantify and qualify data from various points of view during the period of 2000-2015. Various statistical tools including arithmetic mean, percentage, cumulative percentage, time series analysis, simple linear regression, correlation coefficient analysis, 'f-test', 't-test', ANOVA etc. were used and various scientometric indices were used for the study. The current study also used software including MS-Excel, SPSS, Harzing's Publish or Perish (PoP) to analyze and test the data. Results The year 2003 was the most productive year and the 2015 was the least productive year in Annual Growth Ratio (AGR) of public health published literature across the globe during the period under study. The average Relative Growth Rate (RGR) globally was 0.23 and at this rate the literature of public health doubles every 4.16 years. The average Rate of Growth (RoG) for Bangladeshi production of literature was 1.20 with an average RGR value of 0.27. At the same rate of relative growth, the literature for Bangladesh doubles every 3 years. Globally, the average CC (Collaborative Coefficient) value for global authors was 0.37 demonstrating that there was no significant magnitude of collaboration among worldwide authors; however 92% of total publications output from Bangladesh were collaborated authors, the mean value of CC being 0.62. The average Productivity Per Author (PPA) for global authors is 0.49 which means public health authors produced less than half a publication each year during the study period and this rate for Bangladeshi author is less than world's average (0.31}. If we consider the Activity Index (Al) from 2009 to 2015 the research efforts of Bangladeshi authors were higher than world's research efforts (137.32}. The Activity Index (Al) was maximum in 2015 (490.28) and the lowest in 2001 (12.06}.The citation per publication for global authors was 14.21, which was 27.37 for Bangladeshi authors. Mr. McKee, having publication on public health (292} was the most prolific global author on public health, who had also highest h-index score (75) among the global authors. From the Bangladesh authorship, M. Yunus ranked highest number for publications on public health (21} and h-index score was highest for Mr. Ahsan (62). It was found that research productivity of public health by Bangladeshi author conformed to Lotka's inverse law. On the other hand, Bradford's law of scattering fitted to the data of public health journal globally. It was also observed that Zipfs Law approximated the relationship between rank and frequency of keywords of public health. The results of all hypotheses were tested at 0.01 level of significance (p<0.01). It is interesting to note that all null hypothesis was rejected, which compelled to accept concerned alternative hypothesis to follow throughout. There was always a trade-off relation between RGR and Dt (a) values. It was observed that inequality remains in the values of different measurements of collaboration (Cl, DC and CC). The present research determined the level of collaboration by CC value which was never been revealed previously. For assessing an author's multiple cited rates by single publication ACPCP (Average Citation Per Cited Paper) was proposed. Concluding Remarks The assessment of public health literature with the help of scientometric indicators and bibliometric laws could be very useful to the researchers, scientists, library and information professionals, policy makers, and government agency relating to the concerned fields.