Editorial: Citation analysis – focus on leading Australian nurse authors

Indices such as the h-index are being used increasingly in nursing (Thompson & Watson 2010, Hunt & Cleary 2011). Publication analyses of Canadian (Hack et al. 2010) and UK nurses (Thompson & Watson 2010) have previously been undertaken, and we now present an Australian-based nursing citation analysis. Hack et al. (2010) observed that nurses with an h-index of 10–14 indicated an excellent publication record, so we sought to identify nurses working in Australia with an h-index of 10 or more. To capture a list of nursing academics in Australia, we initially searched Scopus using the following strategy (16 September 2010). The terms nurs* and Australia (affil) were used as a starting point, and articles published prior to 1996 were excluded. From the resulting list, all subject areas appearing on the Scopus pull-down menu except for nursing (3206), health professions (239) and multidisciplinary (13) were excluded. From this list of articles, the top authors ranked by publication number were searched individually to capture a complete listing of articles for each author. We were confident that this strategy would provide a list of nurses with the most publications. In Scopus, author’s names with different initials can appear more than once on a list. The resulting list of the top 100 names included authors with 109 articles to 10 articles attributed to each name. It would be highly unlikely for an author to have a high h-index with 10 or fewer publications; therefore, names with fewer than 10 articles were ignored. Duplicate names, recently retired academics, or those working outside the field of nursing (e.g. co-authors working in non-nursing areas) were excluded from the analysis. Next, an extensive Scopus search using the author’s last name and first initial was completed for each of the remaining authors. We identified nurses with an h-index of 10 or more using this method. A list of all publications listed on Scopus attributed to each author was sent to them via email for verification (October 2010). All but five nurses verified the publications sent to them and some provided additional articles that did not appear on the lists. A final update was completed on 22 November 2010 for each of the 24 nurses using Scopus to calculate their h-index (selecting all publications and then selecting ‘view citation overview’) and total number of citations (Table 1). Next, self-citations were excluded from the overview and the h-index was re-calculated minus self-citations. This list was then printed to calculate the c-index for 2008 and 2009 for all articles that received two or more citations. The c-index is a convenient measure of current impact proposed by Taber (2005); it is the number of articles cited more than once by other research groups in the most recent calendar year. The number of publications may not be complete but for this analysis only the articles with the highest numbers of citations – called the h-core – are important for determining a person’s h-index (Hirsch 2005). We also listed the top 10 articles that received the most citations for this data set (Table 2). While comparing our list (h-index range 10–26) with the UK (4–22) and Canadian list of nurses (14–26), it is of interest that most of the top 10 papers in the current study appear in either medical journals are literature reviews or appear in highly ranked nursing journals. Five of the articles appearing in highly ranked medical journals were co-authored by the highest cited nurse in Table 1 (Kristjanson). The c-index shows that some authors are currently accruing large number of citations from other researchers, which is recommended as a means to reduce bias favouring old articles over recent ones (Taber 2005). Table 1 shows that large numbers of publications or high numbers of total citations do not always equate into a high h-index. Table 2 shows that some of the papers co-authored by Australian-based nurses accrue citations that are considered high in most fields of research (>100) and titles reflect a wide range of topics. It should also be noted that once a paper enters a person’s h-core, it only adds one to the total, no matter how many times it has been cited. A good example of this can be seen in Table 2, where one highly cited paper (3828) accounted for 82% (3828/4680) of the total citations for P. Davidson. We are aware that some Australian-based nurses with an h-index of 10 or more may not appear in Table 1 and apologise for any oversights. A more complete nation-wide survey of all nurses in Australia similar to the one conducted by Hack et al. (2010) would provide a more comprehensive list, but this would be very time consuming and beyond the scope of this editorial. We also acknowledge the difficultly of capturing a full citation record for an individual, especially those with common names (or name changes), those who publish in several areas, have numerous affiliations and that databases often have errors of attribution of articles to a person, who in fact was not a co-author (Jacso 2008). However, in this study, we do use a straightforward search strategy to capture nursing researchers working in Australia using common Scopus search functions and contacted nurses to verify their publications. We also acknowledge that a person’s h-index is dependent on a number of factors. For example, apparent differences in h-indexes between groups of nurses from the UK (Thompson & Watson 2010), Canada (Hack et al. 2010) and the current study may be due to the different ways nurses were identified and the database used to calculate the h-index. The UK study used only citations appearing in the Web of Science, which tends to be a more conservative estimate of a person’s h-index (Bakkalbasi et al. 2006), and the two other studies used Scopus, but selected articles based on different publication years. The nurses whose output we have drawn on in preparing this editorial are all leaders in their fields. Furthermore, all are well-established researchers and scholars, and all have been writing and publishing for several years. In presenting this information, we honour the work of these nurses and draw attention to their longevity and productivity. When these leaders began publishing, indices such as the ‘h’ or the ‘c’ were not widely known or understood. As publishing becomes more sophisticated, there will continue to be new ways of judging or evaluating performance. It is not our intent either to judge or evaluate the nurses whose work we present, and so, we have made a decision to present the data alphabetically rather than in any form of rank order. Despite the many ways that publication outputs can be measured, calculated and ‘crunched’, it is important not to lose sight of the primary aim of nursing research and the published discourses that arise from our research and scholarship. That is, to improve patient care, enhance the health care experiences of patients and families, and strengthen the skills, sustainability and well-being of the nursing workforce. The imperative of nursing’s scholarly discourses to inform practice and the obligation to evidence-based practice, means that the true influence of many papers will not be wholly captured by numerical measures and indices. Rather, it will be found in the therapeutic encounters between nurses, patients and their families and seen in the new and innovative models of care, and nurses’ role extension.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.