PSY6002
Workshops
in Psychology/PSY6031 Workshops in Economic Psychology
Basic Research Skills: Bibliographic aids
Methods for storing your own bibliographic material
Simple solutions | Commercial
solutions | dbCindex
The problem
Once you have collected a set of abstracts and references, you need to
store them in some way. But it isn't very helpful just to keep them
as a series of downloaded text files or emailed messages. Specifically,
you want to be able to:
-
Separate the download files into their individual references
-
Store these references in some kind of orderly way
-
Add your own notes and comments, which might be quite extensive
-
Retrieve a particular reference some time later, when you may have forgotten
most of the details about it
-
Generate lists of references and abstracts relevant to particular topics
-
Generate bibliographies in publishers' standard formats for papers that
you are writing
But although you do want quite a number of facilities, you don't want to
recreate PsycLit or the ISI databases: you want something that is simple
and portable, and contains the information that you are likely to need.
This session describes a number of different strategies for solving this
problem. Not all the strategies will make all the above facilities
available, and each of them will be better on some points than others.
They also differ in cost - the cost to the user in terms of setting up
and maintaining a system, and the cost of buying the software it requires.
It's important to take the right decision about creating a reference data
base - if you continue in a research career, the database you create now
may well be with you to the end of your working life.
This document outlines the principles of some of the available solutions.
More details about procedures are available
in a separate document.
Simple solutions
You can use standard software to maintain a basic database of reference
material, in at least 4 ways. All these solutions are essentially
free, in that they make use of software you will already have access to.
None of them does everything we identified about as desirable, but they
may come close enough for your purposes.
-
You could just keep a list of references in APA (or some other) format
in a Word (or other word processor) document. The sorting facilities
available in any modern word processor would enable to you keep them in
an orderly way, and the search facilities would enable you to do simple
searches on words in title. Advantage: Very simple and transparent.
Snags: The document will quite soon get very long; and if you try
to keep any other information with the references (e.g. abstracts or your
own notes) the strategy will probably break down.
-
In Word, and probably in other word processors as well, you can use the
"Table of Authorities" procedure to store references as a list of short
descriptions - basically the citation form, such as "Smith (1984)".
The full bibliographic detail can be carried along with these short forms
like a floating footnote, and if you copy the short form and paste it into
something you are writing, the full detail will be carried along in a concealed
form, and a bibliography can be constructed automatically. Advantage:
Your reference list will not look so long. Snags: as in the previous
case, difficult to keep ancillary information with the references.
-
Word processors normally contain a simple database function, called MailMerge.
This enables you to construct a Master Document (often referred to as a
Form Letter or Catalogue), which consists of some basic text plus named
blanks. You can then construct a Data Source, which in the case of
Word 97 consists of a table, whose headings correspond to the names you
used for the blanks in the master document. Once you have done that,
you can Merge the master document with the data source to produce a word
processor document in which the blanks have been filled in with the corresponding
data from the data source. It is easy to use this to generate bibliographies.
Advantage: You can use different master documents to generate bibliographies
according to different conventions, so you aren't wedded to APA style;
Mailmerge format is very standard so it is likely that you would be able
to convert your data to more advanced systems you might use in the future
- specifically, it would be easier to convert to using a fully-featured
Data base manager, the next option discussed below. Snags: Understanding
how the MailMerge system works will take a little time (though it is a
generally useful skil); Entering the data will be a bit slower than with
the simpler systems.
-
Office suites usually include, along with a word processor and a spreadsheet,
a Data Base Manager (DBM). In the case of Microsoft Office, it is
called Access. You can use Access (or another such system, such as
Lotus Approach) to construct a data base of bibliographic information.
As in the Mailmerge approach, the "fields" of the data base would be things
like the authors of the paper, the year of publication, etc. And
because modern office suites are quite tightly integrated, you can use
the data in an Access file to generate bibliographies etc as Word documents
in exactly the same way as you would a Mailmerge data source. Specifically,
to quote from the Help files for Microsoft Access (with additions in
italics):
-
If you are using Microsoft Word version 7.0/95 or later, you can use the
Microsoft Word Mail Merge Wizard to create a mail merge document in Word
using a link to Microsoft Access data. Once the link is established, you
can open your document in Microsoft Word at any time to print a new batch
of form letters or labels, or a new reference list, using
the current data in Microsoft Access.
-
For any version of Microsoft Word, you can export Microsoft Access data
to a mail merge data source file that can be used with the mail merge feature
of Word.
-
You can save the output of a datasheet, form, or report as a Rich Text
Format (.rtf) file. A Rich Text Format file preserves formatting, such
as fonts and styles, and can be opened with Microsoft Word and other Windows
word-processing or desktop-publishing programs.
-
You can save the output of a datasheet, form, or report as a Rich Text
Format (.rtf) file and automatically load the file into Microsoft Word.
Advantages: You can do more with data in a data base manager than you can
in a mailmerge data source. Snags: Although Access is a great improvement
on earlier DBM systems, it is still a new (though potentially very useful)
skill you will have to learn; Cheaper issues of Office suites may not include
the DBM; Running the DBM at the same time as a word processor and other
software may overwhelm the memory resources of a smaller/older computer
and cause tiresome crashes.
Commercial solutions
The alternative is to use specialised software solutions, specifically
designed to store bibliographic information and make it available in the
form you need. A large number are available, and the University of
Bristol has published a valuable comparison
of the main ones. Among the best known are:
-
Papyrus, supplied by Research
Software DesignTM. Only available for DOS on
the PC, but cheaper than the others, and well recommended. A Macintosh
version is available.
-
Endnote, supplied by Niles Software inc.
A fully-featured Windows programs, with Macintosh versions also available.
-
Research Information Systems supply
two different systems, Reference Manager, only available for PCs
(but fully featured for Windows), and ProCite, which is available for both
Windows and Macintosh. It's not clear whether they are currently
converging into one system, or diverging, though both seem to offer rather
similar facilities.
All these systems seem likely to offer an excellent service, at a price,
and in particular all can be integrated smoothly with the bibliographic
source databases such as PsycLit and the ISI databases. Papyrus is
currently relatively cheap, and IT services are recommending it, but students
have found it difficult to work with in the past. The other two are
more expensive. Neither the university nor the School has a site
licence for any of them at present. All three offer free demonstration
downloads, so you can try them out and see how they compare if you are
considering buying one for yourself (or from a research project grant).
Members of the School have expertise with Papyrus and probably with the
other systems as well.
A local compromise: dbCindex
In the absence of a site licence to one of the commercial systems, the
School offers its members, visitors and students the chance to use a locally
produced program called dbCindex, which offers rather more facilities than
the simple solutions listed above, but unlike the commercial solutions
is available without charge. It also offers the unique feature of allowing
you to build a semantic network of keywords to index your references.
It is prototype software, and written by an amateur, so you use it at your
own risk, and it does not have anything like the flexibility of the commercial
systems. It is available for PCs running Windows 95/98 only.
Essentially it allows you to:
-
Maintain a standard database file of references, and a second file of topic
names.
-
Navigate easily within each of these lists
-
Add any amount of comment about a reference
-
Make links both from the references to the topics, and between topics
-
Read in reference lists from PsycLit or BIDS output, or from bibliographies
typed in APA format
-
Generate APA formatted bibliographies for any topic
If you want to use dbCindex on a computer on the university campus on which
it is not yet installed, you will need to do the following:
-
Find the folder courses\bibliographics on psycho
-
Copy the folder dbCindex (and, therefore, all its contents) from this folder
into a suitable place on your hard disk (for example, the c:\temp folder)
-
Open the file readme.txt in this folder, and note its contents
-
Use Windows Explorer to open the dbCindex fodler you have copied to your
hard disk
-
Double click on the Setup.exe entry, in order to install dbCindex on your
computer.
-
When you have installed dbCindex, use Windows Explorer to find the file
dbcindex.exe, and drag its icon onto the Windows desktop for future use.
-
Before you use dbCindex for the first time, create a new folder to
contain your data base of bibliographic information.
To use dbCindex on your home machine, you can either copy the files in
the dbCindex folder to floppy disks (3 disks will be needed), or download
the program from the web.
One advantage of dbCindex is that it keeps the biblographic information
in a standard database file, which could be imported into Access (and probably
into any of the commercial sources) if you subsequently decided to change
your strategy; and it does its subject indexing through standard relational
database techniques, again allowing a route for subsequent upgrades.
A second advantage is that because it has been written locally, it may
be possible for new facilities to be introduced if local experience suggests
they are needed - though no promises are being made about this. Obviously
I would welcome a report of any bugs you find: there are almost certain
to be some, and there are certainly some limitations that may be unexpected.
Although dbCindex is supplied free, it remains copyright. Anyone
working in the department is welcome to use it, and to take it with them
when they leave. But you are asked not to copy it to third parties,
and you are not allowed to repackage or sell it.
Stephen Lea
University of Exeter
School of Psychology
Washington Singer Laboratories
Exeter EX4 4QG
United Kingdom
Tel +44 1392 264626
Fax +44 1392 264623
Send questions and comments to the
School administrator or to the
author of this page
Goto home page for: University of
Exeter | School of Psychology
| Staff | Students
|
Research |
Teaching
| Miscellaneous


(access count since 24th November 1998).
Document revised 2nd August 1999