Impact Factor Corrected By Document Types And Its Empitical Studies

Impact factor(IF),one of the most authoritative scientometric indicators for evaluating international status and academic impact of journals,is defined as the average number of citations to each journal in a current year to ‘citable items' published in that journal during the two preceding years.Source items in journals are divided into citable items and non-citable items,Thompson Reuters counts citations to all types of documents,whereas the denominator includes only citable documents.Thus,the incongruence between numerator and denominator is not only a a major defect but also a major factor causing manipulation of IF.To correcr this flaw,with the support of the National Social Science Found of China “multi-dimensional correction of flaws of impact factor and its interdisciplinary evaluation in academic journals”(grant number:15BTQ061),we analyze the quantity and citation characteristics of non-citable items in Science Citation Index(SCI)database,and make the corresponding adjustment to the formula of IF,finally we present 5corrective IFs: IFTotal/Total,IFTotal/AREL,IFAR/AR,IFAREL/AR and IFAREL/AREL.The evaluation of an expert in the field is considered to be the most important criterion for verifying the validity of citation indicators,considering the fact that scholars in a country might be not familiar with some journals from other countries,we just investigated American researchers in the fields of ophthalmology and mathematics to obtain the real impact ranking of American ophthalmologic and mathematical journals in the minds of peer experts as the “gold standard” of journal impact.The correlations between 5corrective IFs and questionnaire score in Ophthalmology and Mathematics are analyzed to verify their journal evaluation effects.Because of the discrepancy of journal evaluation effect of corrective IFs between ophthalmologic journals and mathematic journals,we study the subject distribution features of non-citable items in SCI database.The resultsshow that the number of meeting abstracts included by SCI from 2012 to 2013 is far higher than that of other types,followed by editorial materials,letters,news items and corrections et al.There are little biographical items and book reviews.However,citation per paper on meeting abstracts are lower,and citation per paper on editorial materials and letters are higher.For ophthalmologic journals,the correlation coefficients between the questionnaire score and IFTotal/Total,IFTotal/AREL,IFAR/AR,IFAREL/AR,IFAREL/AREL,IF are 0.698,0.692,0.715,0.713,0.692 and 0.687 respectively,the corresponding correlation coefficients in mathematical journals are 0.510,0.477,0.474,0.480,0.477 and 0.545 respectively.The disciplines published most non-citable items are mainly distributed in oncology,cardiac &cardiovascular systems,clinical neurology and medicine general & internal.Above all,there are great differences in quantity and citation among different types of non-citable items.Individual editorial materials or letters from some journals have higher citations in the effective time window for IF and make greater contributions to IF.For Ophthalmology and Mathematics,it is scientific and reasonable for 5 corrective IFs in journal evaluation,but the differences of corrective effect between different subjects are obvious.For Ophthalmology,the journal evaluation effects of 5 corrective IFs are superior than traditional IF,the corrective effect of IFAR/AR is best,IFAREL/AR is better than IFTotal/Total,followed by IFTotal/AREL and IFAREL/AREL.For Mathematics,the journal evaluation effects of traditional IF is superior than 5 corrective IFs,the corrective effect of IFTotal/Total is best,IFAREL/AR is better than IFTotal/AREL and IFAREL/AREL,the corrective effect of IFAR/AR is worst.Hence,not all subjects need to correct journal IF.It is valuable to correct the IF of journals which published more editorial materials and letters;however,it may not make much sense to correct the IF of journals in other disciplines.Among five corrective IFs,we recommend using IFAREL/AR for journal evaluation.