Software is essential to scientific research. However, its academic value has always been underestimated or even neglected, and the contribution of its developers has not been widely recognized. Although some scholars at home and abroad began to realize the importance of software recently, it still calls for a better understanding of the value of software and more attention to the usage and citation of software. Content analysis is employed to analyze the usage and citation of software in the journal articles in the field of Library and Information Science (LIS) in China to reveal the importance and the situation of software citation in domestic LIS research.
Firstly, we select 9 224 articles published between 2007 and 2016 in nine LIS journals indexed in CSSCI, a famous core journal database in humanities and social sciences in China. Secondly, we create content analysis categories and quantization forms for coding. And then, seven trained coders label these 9 224 articles according to the categories and quantization forms. Finally, we analyze the labeled data and draw conclusions. The findings are as follows: 1) The proportion of articles using software is rising yearly between 2007 and 2016, meaning the LIS research has been increasingly dependent on scientific software. Information Science is relatively more dependent on software: the proportion of articles using software is more than 20% in Information Science in 2016, while the proportion of articles using software is nearly 10% in Library Science.2)The uncitedness of software in domestic LIS research is noticeable, and standardization of software citation behavior requires improvement. The average software citation ratio in nine journals is only 0.16 and has no rising tendency between 2007 and 2016. The researchers tend to cite the URL of software with the ratio up to 5292%. Furthermore, non commercial software gains more citation than commercial software and there is a significant difference between the two. 3) The researchers mention the basic description of software randomly. In the 1 901 times of software usage, we find that less than 30% mention version information, only 6.26% and 8.10% mention creator and website information, and more than 60% have no any mentions of the above software information. 4) The usage frequency and types of domestic software are much less than that of foreign software. Among the 118 pieces of software with the usage frequency greater than or equal to two, 95 pieces (81%) are foreign software, while 23 pieces are domestic software (one of which is from Taiwan province). It suggests that more investment in software research and development (R,D) is needed.
This research demonstrates the importance of software based on the detailed and reliable data. It provides decision making support for relevant departments, assisting in incorporation of software into research evaluation, helping to establish a more open and inclusive evaluation system, and then to improve scientific research efficiency as well as to optimize the allocation of resources. It also helps to reveal the extent of dependence on software in domestic LIS research, to have a better understanding of the discipline characteristics of LIS. The research evaluates the software citation practice in domestic LIS scientifically, promoting the standardization of software citation. 2 figs. 7 tabs. 43 refs.