Relation Between Title Attributes And Article Citations In The Field Of Applied Linguistics

The title plays an import role in a research article.An informative and attractive title not only shows the subject of a study at first glimpse,but also catches readers' attention(Yitzhak,1994).Since the article title is attached much importance,researchers have been exploring what attributes make titles stand out and cited more in different academic fields(e.g.Buter & Raan,2011;Nair & Gibbert,2016;Gnewuch & Wohlrabe,2017).The study aimed to identify whether the four attributes,e.g.,length of title,non-alphanumeric characters,title structure and title information,would affect citation counts in the field of applied linguistics.A total number of 500 titles were collected from five top journals in applied linguistics.They were Applied Linguistics,Journal of Second Language Writing,Language Learning,Modern Language Journal and TESOL Quarterly.These journals are of high impact factors,and can be regarded as representative journals in applied linguistics field.Besides,there was a ten-year range of the collected titles between 2002 and 2011.Each journal was randomly selected 10 article titles each year.The average citation counts per year for each title was downloaded from Web of Science,in which citation reports for articles were contained.We conducted regression analysis to the previously mentioned four attributes of title characteristics,non-alphanumeric characters,title structure and title information.The results showed that only titles containing dataset information had a negative impact on citation counts in applied linguistics.The remaining attributes,which were length of title,compound titles,titles with colon or hyphen,title structures(including short: long,long: short and balanced),did not have any impact on citation counts.A regression equation was obtained in the present study,which was Citation counts = 3.511 +(-1.516)* dataset +(-1.018)* two types or more types of information.The equation indicated that a topic only title would obtain a citation count of 3.511.If this is a dataset type of title,the citation count would decrease by 1.156.If this is a title that contains two or more types of information,the citation count would decrease by 1.018.The results also showed that titles containing information of dataset took up 81.82% in the two-types-or-more titles,which indicated that dataset type of titles play a vital role in citation counts in the two-types-or-more titles.Therefore,titles that contained information of dataset had significant impact on citation counts.Since the observation is of field variation,the study suggests that research be conducted within one filed instead of mixed fields.The present study presents a guidance for organization of titles in applied linguistics.Titles with information of dataset is more likely to be less cited.If the articles in applied linguistics intend to obtain higher citation,it is better for researchers to avoid titles with dataset information.Besides,authors are advised to keep title informative,accurate,concise and attractive.The present study also contributes to pedagogical practices of computer-assisted title classification.According to the correlations between title attributes and citation counts,the machine can automatically classify titles into two categories of “well done” and “improvement needed”.This assists teachers to judge the quality of titles written by the students.