Citation patterns in educational science theses at the University of Cordoba.
Maz-Machado, Alexander ; Torralbo-Rodriguez, Manuel ; Gutierrez-Arenas, Ma Pilar 等
Introduction
The origins of the Faculty of Education Sciences Library can be
traced back to the nineteenth century. This was a time of cultural and
scientific renaissance, marked by the emergence of such centres as the
Escuela Normal de Maestros (Teacher Training College) in 1841; this
period also saw the full development of the University of Cordoba, and
the initial shaping of its library.
The library's archives and historical holdings therefore house
all the documentation generated from the founding of the Teacher
Training College to the present day, conserving them in excellent order:
the collection is virtually complete from 1850, and some records date
back as far as the eighteenth century and early nineteenth centuries.
The books that have come down to us clearly testify to the enormous
interest shown by the teaching staff in their Library, particularly at
the time of the First Spanish Republic - in the late nineteenth century
- and later during the Second Spanish Republic. During these periods,
the Teacher Training College played a major role in the city's
cultural, social and educational life, and indeed became the true
driving force of local culture (Rodriguez and Morillo-Velarde, 1994).
In 1995, the old Cordoba Teacher Training College became the
Faculty of Education Sciences; since then, it has trained future infant
and primary teachers, as well as producing graduates in educational
psychology. The faculty's teaching staff were organised into 18
departments, and three doctoral programmes were launched. The library
was continually expanded to support this new academic initiative, and
its holdings now comprise 121,200 modern books and 6,000 old volumes
dating from the eighteenth century to the mid-twentieth century, as well
as some 2,000 journal titles, more than 6,000 special media items (DVDs,
CDs, maps etc.) and a history archive holding almost a million documents
generated since 1850.
The Faculty library is one of eight 'service points'
which together make up the University of Cordoba Library. Like the other
University libraries, it uses the "Millennium" Integrated
Library Management System (Spanish initials: SIGB) comprising
circulation, cataloguing and serially-arranged modules, as well as a
large number of information and electronic resources accessible through
the website.
The faculty library manages journal subscriptions and digital
access independently through the EBSCO information agency. In addition,
a number of electronic journals are included in packages purchased by
the Andalusian University Libraries Consortium (Spanish initials: CBUA)
and others found in Open Access (OA), which are centrally managed by the
Electronic Resources and Projects Service and are accessed from the
university library via the appropriate portal.
In the fifteen years since then, the doctoral programmes have led
to the production of numerous doctoral theses, which have served to
consolidate research in the field of education at the University of
Cordoba. Analysis of the citations in these theses gives a clear idea of
the references used by researchers in a whole range of scientific
fields, belonging to various departments and faculties. This information
is useful in enabling librarians to manage their journal subscriptions
and bibliographical holdings. PhD students also benefit because these
data provide a starting point for their bibliographical surveys of
specific topics and subject areas (Chaparro-Martinez & Marzal, 2008;
Pillai & Dileep, 2010).
Review of related literature
A number of studies have shown that citation analysis is a powerful
tool in identifying the publications that lend theoretical and empirical
support to scientific progress (Garfield, 1978). The social sciences,
and specifically Education, are no exception to this.
Recent research has highlighted the importance of identifying
resource requirements for doctoral theses in university library
collections (Edwards, 1999; Haycock, 2004; Omoba & Ayodeli, 2010).
With regard to theses in the field of education, certain findings of
interest have been published: for example, among theses on Educational
Psychology at the University of Texas, 56% of citations relate to
journals and 37% to books (Fuchs, Thompsen, Bias & Davis, 2006);
Haycock (2004) reported that 44% of citations related to journals in the
field of education sciences. In Nigeria, Okiy (2003) found that 60% of
citations refer to books and 24% to journals; similar results are
reported by Iya (1996) in an examination of education theses at the
University of Maiduguri in Nigeria. No studies have yet been conducted
in Spain on education theses produced at one specific university,
although there has been research addressing the country as a whole, and
focussing on particular areas of knowledge. In Spanish theses on the
teaching of mathematics, for example, the average number of citations
referring to books is 115, while the average referring to journals is 73
(Vallejo, Fernandez, & Torralbo, 2006).
Given that one of the major tasks of any university library is to
evaluate its own impact in terms of usage and loans to users (Iivonen,
Nygren, Valtari & Heikkila, 2009), this study sought to analyse
citations in doctoral theses in the field of education at a single
Spanish institution, the University of Cordoba, with the aim of
establishing which documentary sources are commonly used and of
ascertaining the degree of correspondence between the journals cited and
the journals held and managed at the university's Education Faculty
library.
Methodology
This study used bibliometric methods commonly applied in this type
of research. Thirty theses were examined, all submitted to the
University of Cordoba Faculty of Education Sciences between 1997 and
2010 and deposited in the university library.
The covers of all theses were photocopied, as were as the pages
containing bibliographical references. All the information obtained was
entered into an ad hoc database with the following fields: a) author b)
title c) date d) department; the following data were collected regarding
citations: e) number of citations per thesis, f) number of citations per
source-type g) frequency of journal titles cited.
All references were listed and grouped by document type: journal
articles, books, proceedings, government documents, online resources,
theses and other resources. These were subsequently ranked by citation
frequency and by source language. The cited journals were classified in
descending order according to the number of times they had been cited.
The country of publication for the most cited journals was then
established in order to determine which region was the most influential.
The findings were collated in tables and diagrams.
Results and discussion
The 30 dissertations yielded a total of 8,771 bibliographical
references, with an average of 292.4 per thesis. Monographs, chapters of
books and published books were grouped generically as citations of
books. Book citations accounted for 56% of the total, while journal
citations accounted for 34%; there was thus a clear preference among
doctoral students towards citing books rather than journals, despite the
fact that the faculty library has access to electronic scientific
journals in addition to subscriptions to printed journals. Other type of
bibliographical source accounted for only 9.6% of total citations (Table
1).
Table 1. Number of citations by source type
Type of citation No. of citations Percentage of citations
Books 4,914 56.03
Journals 3,014 34.36
Proceedings 212 2.42
Online resources 163 1.86
Government documents 141 1.61
Theses 119 1.36
Other resources 208 2.37
Total 8,771 100.00
Most of the cited documents were available in printed form. Online
resources accounted for only 1.86%, a poor level of usage on the part of
doctoral students. However, 65.6% of online citations came from 2009 and
2010, suggesting that this source may be starting to gain momentum in
research programmes.
Overall, so-called 'grey literature', in other words
unpublished or limited circulation documents not issued with an ISBN or
ISSN, such as government documents, online resources, theses and other
resources, accounted for 7.19% of the citations.
Comparison of these findings with other citation studies focusing
on doctoral theses in education from around the world reveals that the
percentage of articles cited lies midway between the figures reported
for the USA and for Nigeria, while the book citation figure is much
closer to that of Nigeria (Table 2). The average number of article
citations per thesis was the highest of all (292.36), almost three times
the average noted in Haycock's (2004) study, which had the second
highest average (105.62).
Table 2. Comparison of findings with other studies
Haycock Beile et Tunon & Fuchs et al.
(2004) al. Brydges (2006)
(2004) (2005)
Country USA USA USA USA
Area Education Education Education Educational
Psychology
Period 2000-2002 2000 * 1997-2002
N[degrees] 43 30 143 52
theses
N[degrees] 4542 1842 10023 1560
citations
% 44 45 67.58 56.41
articles
Books, 56 33 18.51 27.9
book
chapters
Okiy Maz-Machado
(2010) et al. (this
paper)
Country Nigeria Spain
Area Education Education
Period 1992-2002 1997-2010
N[degrees] 70 30
theses
N[degrees] 4012 8771
citations
% 24.5 34.36
articles
Books, 60.3 56.03
book
chapters
* Not available.
Analysis of the language of citations enabled the relative primacy
of each to be determined. Spanish was found to be as the most cited
language, accounting for 59.8% of the total, followed by English at
35.2% (Table 3). The presence of French and Portuguese was minimal,
indicating that literature in these languages plays a very secondary
reference role. Other languages involved in citations tended to be
European or local in scope, e.g. Catalan and Galician.
Table 3. Citations by language
Language N[degrees] of citations %
Spanish 5249 59.8
English 3098 35.2
French 214 2.4
Portuguese 113 1.4
Other 97 1.2
Bradford's Law (Brookes, 2006) claims that a majority of
articles are published in a limited number of journals. That law was
applied to determine the nucleus and successive Bradford zones for the
most cited journals, using a procedure previously tested by various
researchers (Ruiz de Osma, 2006; Gorbea, 1996) to identify the nucleus
and zones. Table 4 was drawn up on the basis of the number of journals
and the number of citations relating to each journal.
Table 4. Distribution of scientific journals in descending order
of number of cita (following the Bradford model)
A # B # C Total D E F log
Journals Citations citations Accumulated Accumulated (r)
A*B journals citations
R(r)
1 110 110 1 110 0.0000
1 63 63 2 173 0.3010
1 61 61 3 234 0.4771
1 60 60 4 294 0.6021
1 53 53 5 347 0.6990
1 44 44 6 391 0.7782
1 37 37 7 428 0.8451
2 34 68 9 496 0.9542
3 28 84 12 580 1.0792
1 27 27 13 607 1.1139
1 24 24 14 631 1.1461
1 23 23 15 654 1.1761
2 21 42 17 696 1.2304
1 20 20 18 716 1.2553
2 19 38 20 754 1.3010
2 18 36 22 790 1.3424
1 17 17 23 807 1.3617
2 16 32 25 839 1.3979
2 15 30 27 869 1.4314
5 14 70 32 939 1.5051
3 13 39 35 978 1.5441
3 12 36 38 1014 1.5798
7 11 77 45 1091 1.6532
2 10 20 47 1111 1.6721
5 9 45 52 1156 1.7160
13 8 104 65 1260 1.8129
13 7 91 78 1351 1.8921
17 6 102 95 1453 1.9777
26 5 130 121 1583 2.0828
28 4 112 149 1695 2.1732
69 3 207 218 1902 2.3385
172 2 344 390 2246 2.5911
789 1 789 1179 3035 3.0715
A # G H = E - G I % Zone
Journals Accumulated Residues Residues
citations R(r)-Rc H/E
calculated (r)
Rc(r)
1 -1173.1545 1283.1545 1166.50% 0
1 -769.8648 942.8648 545.01% 0
1 -533.9555 767.9555 328.19% 0
1 -366.5752 660.5752 224.69% 0
1 -236.7449 583.7449 168.23% 0
1 -130.6658 521.6658 133.42% 0
1 -40.9773 468.9773 109.57% 0
2 105.2435 390.7565 78.78% 0
3 272.6239 307.3761 53.00% 0
1 319.1946 287.8054 47.41% 0
1 362.3124 268.6876 42.58% 0
1 402.4541 251.5459 38.46% 0
2 475.2771 220.7229 31.71% 0
1 508.5332 207.4668 28.98% 0
2 569.8345 184.1655 24.43% 0
2 625.2882 164.7118 20.85% 0
1 651.1513 155.8487 19.31% 0
2 699.6648 139.3352 16.61% 0
2 744.4426 124.5574 14.33% 0
5 843.2939 95.7061 10.19% 0
3 895.4324 82.5676 8.44% 0
3 943.2805 70.7195 6.97% 0
7 1041.6532 49.3468 4.52% 1
2 1066.9539 44.0461 3.96% 1
5 1125.7740 30.2260 2.61% 1
13 1255.6043 4.3957 0.35% 1
13 1361.6833 -10.6833 -0.79% 1
17 1476.4005 -23.4005 -1.61% 1
26 1617.1516 -34.1516 -2.16% 1
28 1738.2617 -43.2617 -2.55% 1
69 1959.6741 -57.6741 -3.03% 1
172 2298.0930 -52.0930 -2.32% 2
789 2941.7505 93.2495 3.07% 2
Columns A and B are input fields to calculate Bradford. The
remaining columns are calculated as set forth:
A * B = Total number of citations.
R = accumulated number of journals.
R (r) = accumulated number of citations.
Log (r) = the decimal logarithm of r.
Rc (r) = accumulated number of citations, calculated as set by
Bradford model: Y=a+bX, where Y = Rc (r), "a" is the intercept,
"b" is the slope of the line; "X" is the decimal logarithm of the
accumulated number of journals: Log (r).
H = E-G = remainder, difference between actual and calculated
number of accumulated citations.
H / E = Percentage of the remainder compared to the actual number.
For graphical representation, the decimal logarithm of the number
of journals was plotted on the X axis and the accumulated number of
citations on the Y axis. In terms of Bradford's Law, the graph
displayed a curved area corresponding to the nucleus and a linear area
corresponding to the successive Bradford zones (Figure 1).
The Bradford distribution shows an initial zone or nucleus followed
by a straight section. Sometimes a further zone may be detected beyond
this straight line, where the number of articles increases more slowly
(Gross, 1967). By clustering the data to the right of the nucleus into
groups with numbers of citations similar to the total found in the
nucleus, Bradford zones can be identified.
Table 5. Percentages of the number of journals and citations in
each Bradford zone
Zone N[degrees] N[degrees] % Journals % Citations
of Journals of Citations
Nucleus 0 38 1014 3.22% 33.41%
1 180 888 15.27% 29.26%
2 961 1133 81.51% 37.33%
Total 1179 3035 100.00% 100.00%
Table 6. Bradford nucleus of cited journals
Journal N[degrees] Country of
citation
Ensenanza de las Ciencias 110 Spain
International Journal of 63 United
Science Education Kingdom
Cuadernos de Pedagogia 61 Spain
Revista de Escuelas Normales 53 USA
Bordon 44 Spain
Journal of Personality and 37 Spain
Social Psychology
Infancia y Aprendizaje 34 USA
Science Education 34 Spain
European Journal of Science 28 USA
Education
Journal of Research in Science 28 USA
Teaching
Revista de Educacion 28 USA
Cultura y Educacion 27 Spain
Revista de Investigacion 23 USA
Educativa: RIE
Investigacion en la Escuela 21 Spain
Psicothema 21 Spain
Revista de Psicologia del 20 Spain
Deporte
Revista de Pedagogia 19 Spain
Developmental Psychology 18 USA
Journal of Applied Behavior 17 Spain
Analysis
Revista de Psicologia Social 16 Spain
Aula de Innovacion Educativa 14 Spain
Boletin de la Institucion Libre 14 Spain
de Ensenanza
Cuadernos de Literatura 14 Spain
Infantil y Juvenil
Educational Research 12 USA
Revista Interuniversitaria de 12 Spain
Formacion del Profesorado
Apuntes de Psicologia 11 Spain
Journal of Teacher Education 11 USA
Science 11 USA
Sex Roles 11 USA
Alambique 9 Spain
Revista Espahola de Pedagogia 9 Spain
International Journal for 8 Netherlands
Educational and Vocational
Guidance
Journal of Child Psychology and 8 United
Psychiatry Kingdom
Revista de Ciencias de la 8 Spain
Educacion
Revista de Logopedia, Foniatria 8 Spain
y Audiologia
Cuadernos de Psicologia del 7 Spain
Deporte
Harvard Educational Review 7 USA
Journal of Educational 7 USA
Psychology
A total of 1179 journals were cited in the doctoral theses
surveyed. The data show that the 38 journals contained in the nucleus
accounted for 33.41% of all journal citations. Spanish journals
accounted for 57.89 % of the nucleus and American journals for 34.21%,
suggesting that doctoral theses at the University of Cordoba rely
heavily on Spanish research. The most cited publication was the Spanish
journal Ensenanza de las ciencias.
The Faculty of Education library stocks and manages 185 of the
journals cited, which accounted for only 16.24% of the total, i.e.
doctoral students have library access to only 16.24% of the journals
cited in their. Of these 185, 122 are Spanish journals, meaning that
home-grown publications account for 65.9% of the journals accessed
through the library, while foreign publications account for only 34.1%.
Conclusions
The literature cited in education-related theses at the University
of Cordoba is written largely in Spanish. Most derives from books, and
to a lesser extent from journals. The use of online resources is
negligible. The percentages of article and book citations in education
theses at the University of Cordoba do not match the figures obtained
from research carried out in the USA.
It is clear that the journal collections held and managed by the
library are not cited by doctoral students. If journals report the
findings of the most recent research and these are not being cited by
the doctoral students it implies that the collections chosen do not meet
the major needs of the students of the certain faculty This ought to
induce a process of reflection on the advisability of persevering with
these journals and instead trying to incorporate the journals that are
predominantly cited in the theses.
This type of study is particularly valuable in times of crisis,
such as the present, when economic resources start to become scarce and
need to be optimised. This study in particular will enable library
managers to make decisions on the basis of objective findings that will
undoubtedly lead to benefits for both the library and its users.
References
Adeola, F., & Ayodeji, B. (2010). Evaluation of references in
dissertations and theses against the holdings in a university library.
Library Philosophy and Practice, (February). Available:
http://unllib.unl.edu/LPP/omoba-fabunmi.htm.
Beile, P. M., Boote, D. N., & Killingsworth, E. K. (2004). A
microscope or a mirror?: A question of study validity regarding the use
of dissertation citation analysis for evaluating research collections.
The Journal of Academic Librarianship, 30(5), 347-353.
Brookes, B. C. (2006). Bradford's law and the bibliography of
science. Nature, 22, 953-956.
Chaparro-Martinez, E. I., & Marzal, M. A. (2008). Analysis of
information use in agricultural science PhD theses at Central University
of Venezuela (1986-2002). Library Review, 57(2), 123-137.
Edwards, S. (1999). Citation analysis as collection development
tool: a bibliometric study of polymer science theses and dissertations,
Serials Review, 25, 11-21.
Fuchs, B. E., Thomsen, C. M., Bias, R. G., & Davis, D. G.
(2006). Behavioral Citation Analysis: Toward Collection Enhancement for
Users. College & Research Libraries, 67(4), 304-324.
Gao, S. J., Yu, W. Z., & Luo, F. P. (2009). Citation analysis
of PhD theses at Wuhan University, China. Library Collections,
Acquisitions, & Technical Services, 33, 8-16.
Garfield, E. (1978). Citation studies as a tool in journal
evaluation. Science, 178, 471.
Gorbea, S. (1996). El modelo matematico de Bradford: su aplicacion
a las revistas latinoamericanas de las ciencias bibliotecologicas y de
la informacion. Mexico: UNAM.
Groos, O. V. (1967). Bradford's law and the Keenan-Atherton
data. American Documentation, 18, 46.
Haycock, L. A. (2004). Citation analysis of education dissertations
for collection development. Library Resources & Technical Services,
48(2), 102-106.
Iivonen, M., Nygren, U., Valtari, A., & Heikkila, T. (2009).
Citation analysis of dissertations in the field of economics and
administration. Library Management, 30(3), 185-203.
Iya, A. I. (1996). A citation study of Education Dissertations at
the University of Maiduguri, Nigeria. African Journal of Library,
Archives and Information Science, 6(2), 129-132.
Okiy, R. B. (2010). A citation analysis of education dissertations
at the Delta State University, Abraka, Nigeria. Collection Building,
22(4), 158-161.
Rodriguez, D., Morillo-Velarde, J. (1994). Las bibliotecas
universitarias andaluzas: apuntes para una historia [Cordoba]. En
Rodriguez-Pantoja, M. (dir.) Las universidades andaluzas y el libro (p.
9-27). Sevilla: Junta de Andalucia.
Ruiz de Osma, E. (2006). Aplicacion del modelo Bradford en la
produccion cientifica del area biomedica de la Universidad de Granada
(1988-1996). Encontros Bibli: revista electronica de biblioteconomia e
ciencia da informacao, 2, 1-23.
Pillai, K. G. y Dileep, V. (2010). Scientometric Study of Doctoral
Dissertations in Biochemistry in the University of Kerala, India (2010).
Library Philosophy and Practice (e-journal). Paper 398. Available:
http://digitalcommons.unl.edu/libphilprac/398
Tunon, J. & Brydges, B. (2005). Improving the Quality of
University Libraries through Citation Mining and Analysis Using Two New
Dissertation Bibliometric Assessment Tools. Paper presented at the 71st
IFLA (International Federation of Libraries and Associations)
Conference, Oslo, Norway August 22, 2005.
Vallejo, M., Fernandez, A., & Torralbo, M. (2006). Patrones de
citacion en la investigacion espanola en educacion matematica. Revista
Espanola de Documentacion Cientifica, 29(3), 382-397.
Waugh, C., & Ruppel, M. (2004). Citation analysis of
dissertation, thesis, and research paper references in workforce
education and development. The Journal of Academic Librarianship, 30(4),
276-284.
Alexander Maz-Machado
Universidad de Cordoba - Spain, ma1mamaa@uco.es
Manuel Torralbo-Rodriguez
Universidad de Cordoba - Spain, ma1torom@uco.es
Ma Pilar Gutierrez-Arenas
Universidad de Cordoba - Spain, ue2guarp@uco.es
Francisca Morales Sillero
Universidad de Cordoba - Spain, bg3mosif@uco.es
Maz-Machado, Alexander; Torralbo-Rodriguez, Manuel;
Gutierrez-Arenas, Ma Pilar; and Morales Sillero, Francisca,
"Citation patterns in Educational Science theses at the University
of Cordoba" (2012). Library Philosophy and Practice (e-journal).
Paper 853.
http://digitalcommons.unl.edu/libphilprac/853
Alexander Maz-Machado
Manuel Torralbo-Rodriguez
Ma Pilar Gutierrez-Arenas
Faculty of Education Sciences
University of Cordoba
Cordoba, Spain.
Francisca Morales Sillero
Faculty of Education Sciences Library
University of Cordoba
Cordoba, Spain.