Bibliometric Analysis of the Articles from the School of Dental Medicine, University of Zagreb, Indexed in Web of Science Database (1976-2005)

Vesna Borić


Central Dental Library, School of Dental Medicine, University of Zagreb


Address for correspondence:

Vesna Borić

University of Zagreb

School of Dental Medicine

Central Dental Library

Gundulićeva 5

HR-10000 Zagreb, Croatia


Received: August 1, 2006

Accepted: September 1, 2006

Available online: September 22, 2006

Acta Stomatol Croat. 2006;40(3):218-35.


Original scientific article


This article represents bibliometric analysis of published articles with the address of the School of Dental Medicine, University of Zagreb, indexed in Web of Science database in the period 1976-2005. The sample consisted of 428 articles that were analyzed according to the authors’ address, productivity markers (authors, institutions, year of publication, article type, language, journal) and citation. The analysis of the structure showed that the University of Zagreb affiliation was mentioned in 87.7% of the addresses, and School of Dental Medicine affiliation in 98.8% of the addresses, in 13 different ways. Most publications (82.5%) are scientific articles, 15.2% are congress abstracts, while reviews, short communications, letters, and book reviews and personal articles are present in less that 1%. The articles were published in 88 different journals, mostly in Coll Antropol (39.9%). According to the JCR database, the journals are from 31 different scientific areas. Although most of the publications (24) are from the field of Dentistry, Oral Surgery & Medicine, the greatest number of articles is published in two journals from the field of Anthropology. There are 1866 authors of 428 articles, mostly with four authors (23.8%); greatest number of authors being eleven. Mean number of authors per article is 4.4. Collaboration inside the School yielded 45.1% of the articles; slightly more in collaboration with the authors form other Croatian institution (47.4%), and 6.1% in collaboration with foreign authors. The analysis showed that 41.1% of the articles has no citation; the most cited article has 54 citations, mean citation being 2.68 per article. Out of the ten most cited articles, two are published in collaboration with foreign authors, one is a result of collaboration inside the School, and six are result of collaboration with the scientists from the Institute Ruđer Bošković.


Key Words: bibliometrics; authorship; productivity; School of Dental Medicine University of Zagreb



The research represents an important activity of every scientific and teaching institution, and its results are mostly published in scientific journals. That is one of the modes in which a scientist and the institution in which she/he works presents its local work and takes part in the global exchange of the knowledge. Once published, an article becomes accessible to public and as such can be valued and compared to the results of other researchers. The objective criteria of scientific contribution have not been established, but mostly formal markers are used. In measuring of the scientific work one finds more often bibliometric and scientometric methods of productivity analysis and citation (1). Such research is usually performed on the journal, scientific area, institution, and university or state basis. Although the results of these methods, when used on small samples – singular authors, for example – are not always statistically appropriate, there is a certain number of articles that analyze groups of Croatian scientists on the school, institute and state levels.

Scientist productivity is better known in the bibliometric research in Croatia. Analyses of scientific productivity of the School of Medicine (2), Institute for medical research and occupational medicine (3) have been published, and Klaić published an analysis of scientific productivity of Croatian institution from 1980 until 1995 that used journal quality, based on mean area impact factors, as a parameter (4). Somewhat less frequent is the research that, apart from productivity, analyses citations, for example, scientific productivity and citation of the chemists from the Institute “Ruđer Bošković” (IRB) (6), productivity and citation of Croatian anthropologists (7). There is a small number of very significant research that give an insight into world productivity of a certain group of scientists, like the recently published analysis of global dental production (8). The aim of this study was to represent bibliometric analysis of the articles, published by the scientists from the School of Dental Medicine University of Zagreb (School of Dental Medicine), indexed in ISI Web of Science database (WoS) and published with the address of the School of Dental Medicine. WoS is a bibliographic citation database of the Thomson Scientific (Philadelphia, USA), formerly known as Thomson Institute for Scientific Information (ISI) (WoS includes Science Citation Index Expanded (SCI-Expanded), Social Science Citation Index (SSCI) and Arts & Humanities Citation Index (A&HCI) databases).




Materials and methods

A “General Search” of the Wos database was performed for the period from 1976 to 2005 (total of 30 years). The searched parameters were: dent* and zagreb*, stom* and zagreb* and zahn* and zagreb*. The collected group of 505 articles was stored in an MS Excel 2003 file that was used for analysis and statistical analysis of the data. In the WoS database all the authors’ addresses are given in one field, so it is possible to find one of the given parameters in the address of one institution, and the other in the address of a different institution. Therefore a preliminary analysis was performed and 77 articles were excluded, since the searched parameters did not include School of Dental Medicine. The remaining 428 articles represented a sample for further analysis. At least one author had stated School of Dental Medicine as her/his main affiliation. The correctness of the obtained sample is proportionate to the correctness of the address given by the authors. Analysis did not include articles from the School of Dental Medicine scientists with addresses outside Croatia, as well as articles with different affiliations, which can happen due to various contracts. The collected data were analyzed by means of descriptive statistics. The analysis included authors’ addresses, productivity parameters (authors, institutions, year of publication, article type, language, journal) and citation.





Addresses’ analysis

The basic problem in identifiying a scientific institution is its address, depending on the language in which the article is written, as well as the structure of the address itself. In the WoS database the addresses of all co-authors are listed in one field and separated by semicolon. In 428 analyzed articles there were 900 addresses of institutions. A detailed check of every address has shown that in 511 it was irrevocably confirmed that one of the authors comes from the School of Dentistry (9), and in 389 remaining addresses the authors were from different institutions.


Address structure

It was expected that the address would first state the university, than the institution, department, and then city and state. University of Zagreb affiliation was mentioned in 448 addresses (87.7%). One address (0.2%) stated the non-existent Univ Dent Med, while 62 addresses (12.1%) did not mention university affiliation at all.

The analysis showed that 505 addresses (98.8%) stated the School of Dental Medicine, and the remaining 6 (1.2%) did not mention School of Dental Medicine at the first place, but it mentions a university or department(s). There is a total of 13 ways of stating the name of te School of Dental Medicine (Table 1). The most used term is Sch Dent Med in 293(57.3%), which is also used by the School in its official publications. Six addresses did not mention School of Dental Medicine, and the most unusual form of the name is Fac Dent, Sch Dent.

Due to partial employment in different institutions, 14 addresses stated two institutions, and one address stated three institutions (Table 2). It has been noted that the same authors sometimes use one, and sometimes the other institution.


Departments’ analysis

The analysis according to the departments was difficult to perform, since the official names of the departments have changed during the years; there were inconsistencies in name stating as well. Since the department analysis would not have influenced the results on the School of Dental Medicine level, only two examples of department names are mentioned (Table 3), which point out the problems in this analysis.


Articles’ analysis

The number of articles according to year of publication and type is depicted in Table 4. Out of a total of 428 articles, 353 (82.5%) are scientific (A), one (0.2%) is a review ®, 65 (15.2%) are congress abstracts (M), three (0.7%) are short communications (N), four (0.9%) are letters (L), one (0.2%) is a book review and one (0.2%) is a personal article (I).


Journals’ analysis

428 articles were published in 88 journals, 425 in English, two in Croatian and one in German. One article was published in 52 (59.1%) journals, their single input amounting to 0.2%. The remaining 36 journals (40.9%) published 376 articles (range 2 to 171 articles); most articles were published in Coll Antropol (Table 5). Six Croatian journals published 200 articles (46.7%), and 82 foreign journals published 228 (53.3%) articles.

(Journals with only one article published: Acta Anat., Acta Paediatr., Acta Pharm. Jugosl., Acta Zool. Acad. Sci. Hung., Am. J. Hum. Genet., Anticancer Res., Arch. Med. Res., Arch. Oral Biol., Arch. Toxicol., Biomed. Pharmacother., Biometals, Brit. J. Oral Surg., Br. J. Plast. Surg., Clin. Genet., Clin. Investig., Deut. Zahnarztl. Z., Endod. Dent. Traumatol., Eur. J. Histochem., Eur. J. Oral Sci., Eur. J. Orthodont., Eur. J. Pediatr., Experientia, Folia Microbiol., Head Neck-J. Sci. Spec. Head Neck, Homo, Int. Dent. J., Int. J. Mol. Med., Iugosl. Physiol. Pharm. Acta, J. Chem. Inf. Comput. Sci., J. Chromatogr. A, J. Chromatogr. B, J. Clin. Periodontol., J. Dermatol., J. Mater. Sci.-Mater. Med., J. Mol. Med., J. Periodont. Res., J. Sep. Sci., J. Eur. Acad. Dermatol. Venereol., J. Trace Elem. Med. Biol., Life Sci., Mater. Lett., Nucl. Instrum. Methods Phys. Res. Sect. B-Beam Interact. Mater. Atoms, Oper. Dent., Opt. Commun., Oral Dis., Oral Oncol., Photomed. Laser Surg., Polym. Int., Radiologe, Vet. Med., Vet. Arh. and Wien. Klin. Wochen.)

The analysis of the scientific fields of the journal where the articles were published (Table 6) used the list created by Popescu (10). One journal title had one discipline attached, mostly in concordance with the WoS disciplines. For the purpose of this research, two disciplines from social sciences were added (Anthropology and Information science & library science) that are not included in the database. Analyzed articles are cathegorized in 31 scientific fields. Table 6 gives data on fields, number of articles, number and percentage of articles in a given field.


Authors’ analysis

Total of 1866 authors cooperated on 428 published articles, and the total number was achieved by adding the number of authors per article. The analysis has confirmed that most of the articles had four authors (102; 23.8%), followed by five authors (86; 20.1%), three authors (68; 15.9%), two authors (43; 10.1%), seven authors (19; 4.5%), eight authors (9; 2.1%), one author (8; 1.9%), nine and ten authors (4; 0.9%), and eleven authors (1; 0.2%) (Fig. 1). The mean number of authors was 4.4.

The analysis of the author number per article by decades revealed following: in the first decade (1976-1985) the mean number was 2.5, in second decade (1986-1995) it was 3.8, and in the third decade (1996-2005) it was 4.6.



All 428 articles were authored by at least one author from the School of Dental Medicine, as well as 8 single-author articles. The analysis of each single article revealed often coauthorship inside the School of Dental Medicine (185 articles, 43.2%). 203 articles were published in cooperation with scientists from other Croatian institutions (47.4%). International cooperation was not great (26 articles, 6.1%), and the number of articles coauthored by scientists from the School of Dental Medicine, other Croatian institutions and foreign scientists is very small (6 articles, 1.4%). Since 185 articles were published in coauthorship of the scientists from the School of Dental Medicine, the analysis was needed for the remaining 243 articles. The institutions were classified according to the criterion, thereby making the analysis more difficult due to the different data obtained. Schools are scientific and educational institutions, as well as some healthcare institutions were a portion of the clinical teaching is performed, rendering their double-function. Multiple overlapping is completely possible due to the fact that clinicians need to follow new scientific knowledge. For the purposes of this article five groups were established (three great ones): clinical, scientific and educational (schools). There is a total of 15 clinical institutions whose employees have coauthored 96 articles. All authors from healthcare centers and private practices were included in one institution, and Clinical Hospital Center Zagreb included locations of Šalata, Rebro and Petrova. Scientific institutions group included 13 addresses with coauthors on 81 articles. Educational group registered 16 institutions, their employees coauthoring 125 articles (Table 7). Cooperation with foreign authors is not extensive so the results are shown in the text. The total of 37 articles was coauthored by 12 authors from Germany and Japan (6 each), 5 from the Netherlands, four from United Kingdom, Slovenia and Bosnia Hercegovina, two from Austria and USA and one each from Switzerland, Serbia and Macedonia.

All institutions that were not presented graphically were allocated to group 3. These are HAZU (unified institution of a general scientific profile), Museum of archaeology and Institute for renovation (humanistic sciences). The scientists from these instutions coauthored six articles.

Coauthor analysis was performed for 243 articles, and sum of groups is 345. The difference is due to the fact that on a certain number of articles there was cooperation between more than two institutions. The shares of different groups are depicted in Fig. 2.

If the medical institutions from other groups (Institute for medical research and occupational medicine, Institute of immunology, Veterinary institute, Institute for transfusional medicine and Schools of Medicine in Zagreb, Osijek and Rijeka) are combined with the clinical institutions from the first group, the result is coauthorship of 188 (54.5%) of articles, and with the remaining institutions 157 (45.5%).


Citation analysis

Table 8 shows the numbers of articles and citations. Out of 428 articles there is 176 (41.1%) without citations, and 252 (58.9%) have been cited at least once. The most cited article was cited 54 times, and the mean citation number is 2.68 citations per article.

Table 9 shows the ten most cited articles. One of the articles was coauthored only by the scientists from the School of Dental Medicine, and two in cooperation with the scientists from the USA. All other articles were written in cooperation with domestic scientists/institutions, the greatest number in cooperation with IRB (6). All ten articles are original scientific articles, published between the years 1981 and 2002.




Table 1 School of Dental Medicine name forms

Table 2 Multiinstitutional addresses

Table 3 Deparments` name forms

Table 4 Number of papers according to the year and article type

Table 5 Distribution of papers according to journals

Table 6 Productivity of the School of Dental Medicine according to the scientific discipline of journals

Figure 1 The distribution of the number of authors per article

Table 7 Co-authors per institutions

Figure 2 Distribution of co-authors by institution type

Table 8 Number of papers and citations

Table 9 Ten most cited articles

Discussion and conclusions

Bibliometric research should represent an exact fundament for the comparison of certain scientific communities, but it is often that, in an attempt to compare the results, problems arise due to the fact of the different variations of databases are being used. If the WoS database includes 5900 journals, in the same year the SCI (printed version) includes around 3500 journals, and the SCI (CD version) includes 3700 journals (1), the relevant comparison would be only the one using the same database. Klaić (4) analyzed three types of articles (article, note and letter) from SCI (CD version), SSCI and A&HCI (printed version) and found 58 articles from the School of Dental Medicine. WoS database search with same parameters and same year span (1976-1983) resulted with 64 articles, and when all types of articles are included, there is a total of 75 papers from the School of Dental Medicine. The differences can arise due to the focus of the investigation to some segments of the results. Multiple narrowing of the search results can be seen in the work of Gil-Montoya et al. (8), who chose the SCI segment of WoS. After the search-by-state of origin, the results were narrowed down to only article and review. For citation analysis only the articles published in the journals categorized in Dentistry, Oral Surgery & Medicine inside JCR database were used. This has been narrowed down to the first ten journals in every analyzed year, the results for Croatia being 43 articles. In this analysis the whole WoS database was used (SCI, SSCI, and A&HCI), and all types of articles were used. The search resulted in 190 articles, and when narrowed to Dentistry, Oral Surgery & Medicine, the search in the JCR database gives 136 articles. These examples clearly show that the differences in bibliometrical studies can vary depending on the database selection, time span, methods and narrowing of parameters. Thus, only results obtained by same criteria/conditions can be compared, so standardization of main bibliometrical parameters is needed.

WoS is a general gobal database that enables complex bibliometrical research, but it also has its limitations. For example, the state of origin in the address is standardized, and since Croatia was recognized and standardized in 1993, and is an independent state since 15th January 1992, most articles is registered under “Yugoslavia”. Other elements of the address are not standardized and the words are abbreviated in a unique way, provided the whole words are provided in the address. If an author uses abbreviations, they will not be corrected.

A database search gave 900 addresses, and 511 of them can irrevocably be confirmed to represent the School of Dental Medicine. The analysis showed that many articles do not have the principle of hierarchy in the address structure (university, institution, department). Some addresses are incomplete, lacking one or two elements of the structure. Great problem in identification of the articles was the name of the institution, since there were 13 different variations of the name. If we want to increase the recognizability in global scientific literature, we must pay more attention to the formal identification of the authors, institutions, university and state.

The analysis of the number of articles per year and article type (Table 4) has shown an increase in the number of articles since 1996. Mean number of articles per year in 1976-1995 period was 4.1, and between 1996 and 2005 it was 35. It can be supposed that this was caused by the change in the regulations. Until the end of 1993 the Law on regulated education (Official Gazzette 11/91) (11) that stated that a candidate for a scientific vocation needed “published scientific articles that confirms her/him as an active scientific worker with adequate scientific results”. The council of the institution  “obligatory, every five years gives an assessment, whether the active scientific worker performs the educational work successfully”. In October 1993 the Law on high education institutions (NN 96/93) was accepted (11). This law prescribes that a candidate for a higher vocation or re-election must “have new articles, published after the election to the previous vocation”. Such condition is motivating and obligatory for all active participants, and might have resulted in an increase of productivity. Subsequently, in 1997 the Ministry of science and technology of the Republic of Croatia has issued the “Minimal conditions for election in the scientific vocations” (Official Gazzette 38/97) (11) that, apart from the number of published articles, includes the article evaluation. There are two basic criteria for evaluation: an article published in a journal with international peer-review and indexation of the journal, ie. Article in the Current Contents and/or other bibliographical databases. Since this regulation moves the importance from the quantity to quality, once can expect a further increase in the number of articles published in high-quality international journals.

The analysis of the journals showed that 428 articles were published in 88 journals, most of them in Coll Antropol (171 articles, 39.9%), Table 5. Possible reason is a relatively small number of Croatian journals in the ISI databases. Since the indexation undoubtly increases the visibility of the articles, and since the authors correspond more easily with domestic editorial offices, they decide more frequently to send their manuscripts to domestic journals indexed in global secondary databases. The analyzed articles were published in journals classified in JCR database in 31 scientific fields (Table 6). This database can include a journal, due to multidisciplinary investigations, in a number of fields (for example. Dent Mater in Dentistry, Oral Surgery & Medicine and Materials Science, Biomaterials), so for a more detailed analysis one would have to compute the shares of the field in the journal. Still, even such detailed analysis would not give a more precise picture, since the journal’s inclusion in the field does not imply the inclusion of all articles. Therefore, Popescu’s scientific fields were used (10). One journal title was connected with only one field, and field titles are mostly in accordance with those stated in WoS. Two social science fields were added for the purposes of this analysis (Anthropology and Information science & library science). Although the greatest number of journals (24) is from the Dentistry, Oral Surgery & Medicine field, most articles were published in two journals from the field of Anthropology. Although it was supposed that dentists mostly publish in dental journals, the analysis showed that most of the articles (172, 40.2%) were published in two anthropological journals (Collegium Antropologicum (171 articles) and Homo (1 article)). Such results are not surprising, since anthropological research is traditional in Croatia since the end of the 19th century. The School of Dental Medicine is one of the few schools in the world that has these investigations are organized in the Department of Dental Anthropology. Many multidisciplinary investigation were published were published in collaboration with the scientists from the Institute of Anthropology.

There were 1866 authors of 428 articles. The analysis (Fig. 1) showed that most of the articles had four authors (102, 23.8%), and one article had eleven authors. Mean number of authors per article is 4.4, little more than Croatian mean for all sciences (3.82) (7).

The analysis of the number of authors showed that in the first decad the mean number was 2.5, in the second it was 3.8 and in the third 4.6. Such an increase of the coauthors of the School of Dental Medicine is in accordance with the trend in SCI database, where a mean number of coauthors on “Source Items” was 2.56 in 1980, and 3.55 in 1994.

The analysis of each article showed frequent coauthorship inside the School of Dental Medicine (193 articles, 45.1%), and somewhat more often with authors from other Croatian institutions (203 articles, 47.4%). Formal division of the institution in five groups (clinical, scientific, educational, foreign and others) gave the basis for coauthorship analysis. Many institutions have many activities (clinical, scientific, educational and others), so it is sometimes difficult to establish which is the primary activity, and the data differs in different sources. Finally, it is possible to find coauthors from different groups, so these results only give an overview of the cooperation. There were authors from 15 clinical institutions on 96 articles, from 13 scientific institutions on 81 articles, from 16 educational institutions on 125 articles, as well as from 12 states on 37 articles and from 3 other institutions on 6 articles. The coauthorship of articles from medical institutions in total amounts to 188 articles (54.5%), and with other institutions 157 (45.5%) articles.

Citation analysis has confirmed that 176 (41.1%) articles do not have any citations, while 252 (58.9%) articles are cited at least once (Table 8). The most cited article has 54 citations, and mean number of citation is 2.68. It must be stated that the citation analysis was based on the data from “General Search”, that does not include all cited articles that have an error in the entry so do not carry an adequate “article-to-article” link. Often mistakes are incorrect first name/last name/initials entries, incorrect journal titles, year, volume, number or page numbers. In order to include all citations, every article has to be searched for by a number of parameters. One of the examples is shown in Table 9. “General Search” in April 2006 registered 46 citations, and additional search by “Cited author” in July 2006 and the establishment of “article-to-article” link with the correct last name and initials entry showed 51 citations (Table 10). The difference in the number of citations occurred by the registration of a new citation in WoS database in May 2006.

A 30% error was earlier stressed by a detailed analysis of “invisible” citations in SCI database (12). Finding “invisible” citations significantly influences the analysis of citation distribution, especially in articles without citations and those with small number of citations.

Table 11 represents the distribution of cited articles from the School of Dental Medicine, IRB and SCI (13). IRB was chosed due to the fact that there is published data on its citations (6). Although these are two different databases (Wos and SCI), and different time periods, the distribution of citations is relatively similar. The share of one-citation articles for the School of Dental Medicine and IRB (without error correction) was similar, but the IRB share after the correction was decreased, and the share of multiple-citation articles is growing.

Many single-citation articles in SCI database are the result of an erroneous citation of some of the parameter that define the “Source Item”, and of the citation of certain pages of the monographic publications, where every page is considered to be a separate entity. The correction of citations and unification of book citation would decrease the share of single-citation articles in SCI (58.2%).

Therefore, the results of 176 articles (41%) without citations, as well as the fact that 81 articles from the School of Dental Medicine have only one citation should be corrected by a systematic search that defines “article-to-article” link. Special attention should be given to the specific Croatian system of person identification (mostly women) with two surnames that are too often cited in the wrong way.

Although there are different bibliometrical analyses published, their results vary depending on the database selection, time period searched and type of articles selected. The results of the productivity and citation of the School of Dental Medicine could be compared only with the results of other dental schools in the world, as well as with other Croatian institutions, that were performed according to the same criteria.



Table 10 «Cited author» search

Table 11 Articles distribution according to the number of citations


1. Jokić M. Bibliometrijski aspekti vrednovanja znanstvenog rada. Zagreb: Sveučilišna knjižara; 2005.

2. Petrak J, Božikov J. Journal publications from Zagreb University Medical school in 1995-1999. Croat Med J. 2003;44(6):681-9.  :. ABSTRACT :.

3. Radošević-Vidaček B, Simeon-Rudolf V. Bibliographic output of the Institute for medical research and occupational health in Zagreb between 1994 and 1998. Arh Hig Rada Toksikol. 1999;50(4):423-35.

4. Klaić B. Pokušaj vrjednovanja u znanosti u nas. Rugjer. 1996;1(5):3-11.

5. Klaić B. Scientometrijska analiza znanstvene i odgojne djelatnosti kemičara zaposlenih u Institutu “Ruđer Bošković” u razdoblju od godine 1961. do 1983. Kem Ind. 1986;35(3):141-9.

6. Andreis M. Scientometrijska analiza Instituta “Ruđer Bošković” (1975.-1995.). Rugjer. 1997;2(8):17-30.

7. Bencetić Klaić Z, Klaić B. Scientometric analysis of anthropology in the Republic of Croatia for the period of 1980-1996. Coll Antropol. 1997;21(1):301-18.  :. ABSTRACT :.

8. Gil-Montoya JA, Navarrete-Cortes J, Pulgar R, Santa S, Moya-Anegon F. World dental research production: an ISI database approach (1999-2003). Eur J Oral Sci. 2006;114(2):102-8.  :. ABSTRACT :.

9. Stomatološki fakultet Sveučilišta u Zagrebu. Ljetopis: 1948 – 1993: Stomatološki fakultet Sveučilišta u Zagrebu. Zagreb: Stomatološki fakultet Sveučilišta u Zagrebu; 1994.

10. Popescu II. Scientific journal ranking by average impact factors [homepage on the Internet]. 2002 December [cited 2006 July 25]; about 20 screens. Available from:

11. Narodne novine [homepage on the Internet]. Zagreb: Narodne novine, d.d.; [cited 2006 Jul 28]. Available from:

12. Andreis M. Metodološki problemi u scientometrijskoj analizi. Rugjer. 1997;2(7):3-9.

13. Garfield E. Current comments. Curr Contents. 1984;27(23):3.