The health care chaplaincy mapping study employed the methodology outlined in the NAHRS Mapping the Literature of Nursing and Allied Health Professions: Project Protocol 11
. The study evaluated the citation patterns from selected journals from the discipline of health care chaplaincy over a three-year period as an indicator of which journals provide the highest number of citations for the discipline.
The protocol calls for source journals to be selected from the discipline through a variety of methods. The authors of the study, two health sciences librarians and two faculty members teaching pastoral care and counseling, made an initial hypothesis that the core journals would be Chaplaincy Today, the Journal of Health Care Chaplaincy, and the Journal of Pastoral Care and Counseling. However, for this study, the authors chose to query a purposive sample of seven nationally prominent faculty members generated using the following inclusion criteria: active involvement in health care chaplaincy as a researcher and/or educator; association with US CPE programs; experience as editors, reviewers, and authors; and more than ten years' experience as a professional chaplain. Although sampling bias may be introduced by use of nonprobability sampling, within the context of this study, purposive sampling is appropriate because it provides representation of those who are most involved as both consumers and producers of published theory and research. The faculty sample averaged more than twenty years of experience and represented different geographic and institutional affiliations in the United States.
Members of the sample were contacted by email by one of the researchers and asked to name the top five journals they regularly read in their role as hospital chaplains. A total of twenty-one different journals was generated from this sample. Through an iterative process, consensus was reached on three source journals—Chaplaincy Today, Journal of Health Care Chaplaincy, and Journal of Pastoral Care and Counseling—which agreed with the authors' original hypothesis.
All articles published between 2008 and 2010 in the three source journals were considered for inclusion in the study. Items that were excluded from the analysis were editorials, historical reprints, reference lists, and books reviews. Each reference was assigned a unique identifier, and the following elements were collected: source journal title, volume and issue, format of cited reference, publication date of cited reference, and article title. The formatting categories assigned to each reference were, in order of importance, journals, books, government documents, Internet, and miscellaneous. Journals were defined as serial publications, either print or electronic, produced on a periodic basis. Books, whether print or electronic, included monographs, textbooks, book chapters, manuals, and encyclopedias. Government documents were documents published by government agencies that did not meet the aforementioned criteria for journals and books, and included publications on the Internet. The Internet category was assigned to websites created and maintained by any nongovernmental organization. Finally, the miscellaneous category was used to categorize any items that did not fall into the previous four categories and could include personal communications, unpublished research, dissertations, and newspaper articles. In the case of references that could be classified as more than one format, the format that occurred highest on the list was chosen to categorize the reference. All references were entered manually into a Microsoft Excel spreadsheet.
The final reference list was sorted to determine the total number of citations from each source journal and proportion of formats present. Because the study used nominal data, a chi-square analysis was conducted to determine whether the use of publication formats varied significantly for the entire sample over time. To determine the journals producing the highest number of citations in health care chaplaincy, Bradford's Law of Scattering was applied. This law predicts that for any discipline or specialty, there will be a small set of journals that produce the highest number of articles in the field. A larger number of journals will produce an area of scatter beyond that set that captures journal titles that are referenced less frequently in the discipline. Ranking the order of journal titles in decreasing order of citation frequency allows the list to be divided into three zones of frequency, where each zone produces approximately one-third of the total citations. Zone 1 will contain a small number of highly productive journals in the discipline, Zone 2 will produce a larger number of moderately productive journals, and Zone 3 will produce an even larger number of journals with low productivity.
Major bibliographic indexing or abstracting databases were consulted for the coverage of cited journal titles in Zones 1 and 2. Because of the inter-professional nature of health care chaplaincy, it was judged best to include databases that covered health, counseling, and religion and/or spirituality. Five databases were selected to meet these criteria: MEDLINE (produced by the National Library of Medicine), CINAHL Plus with Full Text and Academic Search Complete (produced by EBSCO), PsycINFO (produced by the American Psychological Association), and ATLA Religion Database with ATLA Serials (produced by the American Theological Library Association and selected due to its wide coverage of scholarly literature in religion). Indexing coverage of each journal in Zones 1 and 2 was determined for 2010 or the last year of publication available in each database. Each of the selected databases was searched to determine if it contained journals from Zones 1 and 2. If the journal was found in a database, the number of articles indexed from it for 2010 or the last available year of publication was captured. These numbers were compared for each journal across the selected databases. Per the NAHRS protocol, the database indexing the highest number of articles for a journal was assigned a full coverage score of 5. The remaining 4 databases were assigned a coverage score based on the ratio between the number of articles indexed and the maximum number of articles indexed for 2010 or the latest available year of publication. Coverage scores were given according to the following rule: 5
1%–24%, and 0
less than 1%. The scores were compared to determine which database provided the most comprehensive coverage for Zones 1 and 2.
As a secondary analysis, the proportion of self-citations for each source journal, was determined. Self-citation was defined as an occurrence of a journal that cited itself in a list of references, for example, using an article from the Journal of Pastoral Care and Counseling
as a reference for a later article in the same journal. Given that there were few titles devoted exclusively to health care chaplaincy, it was of interest to determine the rate of self-citation by each source journal. The percentage of self-citations found in the source journals during the study period was reported by journal. Self-citation is not uncommon in journals that are prominent in a selected discipline, as authors tend to reference seminal work in their own research 12