V.C. Lansingh and M.J. Carter, Does Open Access in Ophthalmology Affect How Articles are Subsequently Cited in Research? Ophthalmology, June 20, 2009. The article doesn’t yet appear at the journal site, so I’ve linked to the abstract in PubMed. Abstract:
OBJECTIVE: To determine whether the concept of open access affects how articles are cited in the field of ophthalmology.
DESIGN: Type of meta-analysis.
PARTICIPANTS: Examination of 480 articles in ophthalmology in the experimental protocol and 415 articles in the control protocol.
METHODS: Four subject areas were chosen to search the ophthalmology literature in the PubMed database using the terms "cataract," "diabetic retinopathy," "glaucoma," and "refractive errors." Searching started in December of 2003 and worked back in time to the beginning of the year. The number of subsequent citations for equal numbers of both open access (OA) and closed access (CA) (by subscription) articles was quantified using the Scopus database and Google search engine. Number of authors, article type, country/region in which the article was published, language, and funding data were also collected for each article. A control protocol was also carried out to ascertain that the sampling method was not systematically biased by matching 6 ophthalmology journals (3 OA, 3 CA) using their impact factors, and employing the same search methodology to sample OA and CA articles.
MAIN OUTCOME MEASURES: Number of citations.
RESULTS: The total number of citations was significantly higher for open access articles compared to closed access articles for Scopus (mean 15.2 versus 11.5, P < .0005, Mann-Whitney U = 20029, and Google (mean 6.4 versus 4.0, P < .0005, Mann-Whitney U = 21281). However, univariate general linear model (GLM) analysis showed that access was not a significant factor that explained the citation data. Author number, country/region of publication, subject area, language, and funding were the variables that had the most effect and were statistically significant. Control protocol results showed no significant difference between open and closed access articles in regard to number of citations found by Scopus: open access: mean = 17.8; SD (standard deviation) = 23.70; closed access: mean = 19.1; SD = 20.31; Mann-Whitney test, P = 0.730, Mann-Whitney U = 20584.
CONCLUSIONS: Unlike other fields of science, open access thus far has not affected how ophthalmology articles are cited in the literature.