Page 112 - DLIS006_INFORMATION SOURCES AND SERVICES

Unit 5: Alerting and Bibliographic Services

Notes
A number of groups interested in meta-analysis also have a particular interest in highly
targeted and systematic literature searching. While the meta-analysis methodology is
used broadly, it is most codified and understood in medical research. The Cochrane
Collaboration established in 1993 has specifically worked to develop and standardise the
methodology applied to medically related research. This is an international not-for-profit
organisation providing up-to-date information about the effects of health care. The group
has a strong guiding role in the methodology development and increasingly features in
the literature on it. It also takes a leading role in the production of systematic reviews
across medical research updating these reviews as new primary research is completed.
Many of the principals and practices detailed in this medically focused literature can and
have been adapted to other areas of research. Reacting to this need a sibling organisation
to the Cochrane group known as the Campbell Collaboration was formed in 1999
(http://www.campbellcollaboration.org/). This group focuses on the systematic review
of social, behavioural and educational research and is part of the wider American Institutes
for Research organisation that focuses on behavioural and social science research
(http://www.air.org/). Another major group with a focus on systematic review and social
research is the UK based Economic & Social research council (ESRC).
While the Campbell and ESRC groups have both identified and generated literature on
systematic searching strategies and methods there is still much development work being
carried out. One issue noted within this literature is the additional challenges faced by
systematic reviewers working outside of the medical field (Jenkins, 2004). Medical
researchers can rely heavily on the Medline database and a highly standardised
terminology usage. The social or business researcher often has a wide range of possible
databases to examine and a much looser terminology usage within the literature. Other
features of the range of databases within these disciplines include poor indexing, lack of
consistent abstract structure, lack of standardisation of terms used, and a variety in design
of database structures with subsequent varying search fields and layouts.
Commonly described Electronic Search Strategies

The typical search description found in recent meta-analysis studies that were examined
as possible instructional examples for the author’s study consisted of the following:
1. Definition of research problem and types of studies to be sought.
2. Scope of search (within a discipline, topic, language, region etc.) and specific variables
being examined.
3. The time period of target studies included in the search ( i.e. relevant studies
completed in a defined period, generally set from a time close to the undertaking of
the systematic study looking back typically 5 to 10 years depending on the topic).
4. Databases used in the search.
5. Key words or phrases used for the search.

6. Assessment criteria for including or culling the usual high number of studies that
this technique typically produces (although this is often incomplete or not described).
7. Follow-up searching using authors’ names taken from relevant studies found in the
above stages.
Additional methods to ensure that a search has been comprehensive include:
1. Use of manual searching – Examination of reference lists from relevant studies
obtained in searches, qualitative studies, review articles or relevant journal or book
searches. Manual examination of relevant journal indexes.
Contd....

LOVELY PROFESSIONAL UNIVERSITY 107