FAQ

 Update Committee Member Info

 Submit Minutes
Electronic Resources Subcommittee
Sociological Abstracts Vendor Comparisons
I initially began the comparison using the same criteria I used for Proquest/OVID but since SilverPlatter is accessed via a LAN and CSA is accessed via the web, some of the criteria would illustrate meaningless differences. What I will concentrate on is searching, display/export and help.
Criteria for comparing Sociological Abstracts on SilverPlatter vs. CSA
Criteria SilverPlatter CSA
1. Searching Capabilities:
1.1 Search fields Free Text searches in the following fields:
  • (ti,au,in,da,as,jn,pb,nt,ab,dem,des,ip,sh)

Users can also limit a search statement to any of 27 possible indexes.

AB= Abstract

AF= Author Affiliation

AN= Accession Number
AU= Author
CA= Corporate Author

CF= Conference

CL= Classification

DE= Descriptors

ED= Editor

IB= ISBN

ID= Identifiers

IS= ISSN

LA= Language

NT= Notes

NU=Other Numbers

[publisher number in Findex]

OT= Original Title

PB= Publisher

PT= Publication Type

PY= Publication Year

SF= Subfile

SL= Summary Language

SO= Source

TI= Title
1.2 Search modes There is only one search screen. Users can opt to use all or none of the additional features. The searching capabilities are as advanced as the searcher. Basic and Advanced

  • Basic: Author, Title &Any field


  • Advanced:
    The advanced search box allows you to specify the fields in which words appear using two-letter Field Codes.

Not clear or explained anywhere on the page. User MUST go to the help in order to figure out how to use the boxes.

1.3 Truncators * - asterisk can be used to substitute for one or many characters. Can be embedded or hanging. Cannot be the first character.

? Is forgiven and actually works.
* - asterisk can be used to substitute for one or many characters. Can be embedded or hanging. Cannot be the first character.

Penalized for using ?

1.4 Search screen layout
  • Nicely laid out in the default and can be customized to satisfy the users. The initial screen can be All search history, all record-display or a screen split between the two.
  • Additionally, the search history window automatically breaks phrases up and searches on the individual words before combining. This is a boon for an end-user since the individual words can then be recombined in a new search statement without having to type in the word again.
  • The advanced/basic search boxes are on the page that lists all 60 databases. So clicking on the back button to reword the search statement requires that the user must wait for the entire list of databases to come up as well, about 25 seconds.
  • Worse yet, the search boxes are below that long list of databases. So the user has to scroll down to fix a typo or add an *. Very clunky.
1.5 Search save/rerun capability Yes. No.
1.6 Limiting before initial search Yes by using the "in" statement to find a word "in" a particular index. Yes. By advanced search and limiting by fields.
1.7 Limiting after initial search Yes. Each step of the search is assigned a number that can be used to re-run the search in new ways. No. The users must restructure the search statement and resubmit it.
1.8 Ability to undo limit. Yes. Just double-click on the previous search statement. No.
1.9 Lateral/Sideways searching Yes. The searcher can double-click on a word phrase or subject heading and click on the add to search button. No.
1.10 Easy reversal of actions Yes. Results are displayed in a history window. To undo, the user need only click on the previous search statement. No.
1.11 Browsable Indexes Yes. Too many to list. (All 27 searchable fields also have browsable indexes.) No.
2. Help/Guidance:
2.1 All messages are clear and concise Yes.
2.2 Context sensitive help No. Not when user has not done anything that results in an error message. Standard windows help guides. Helpful nonetheless but only if the user is comfortable using that type of help. "Your request is not complete and therefore cannot retrieve results.

Use your browser's BACK button to return to your search page and complete your search request. You must: Choose at least one database to search by

clicking inside a checkbox that precedes the database title. Enter at least one search term in one of the search field boxes."

This was not helpful since I had already done both things.

2.3 Error messages The error messages are good and have a help button that will bring up context sensitive help. All error message say the same thing. Insufficient information is given to actually help the end-user correct the initial search statement.
2.4 After Error/Help Message, returned to location in system Yes. No.
2.5 Location in system is always known Yes. Yes.
2.6 Appropriate graphics, icons, etc. Yes. Yes.
3. Display/Delivery:
3.1 Screen Design uncluttered. Yes. And changeable if the user decided that it is! No. Results screen is ok. The search screen is awful since the user has to scroll past 60 database names before they arrive at the search box.
3.2 Display format options Brief/All fields button is available on screen. Users can also chose to customize the display fields. Any or all are available as part of the display for the duration of the search session. 1) Full record (all fields, including abstract, where present)

2) Citation format (title, author, and source fields are included)

3) Citation & Abstract format (tags from citation plus the abstract field)

3.3 Display fields Any or all of the 27 fields can be selected by the end-user for display. Also has some packaged options like, Citn or fields with hits only, etc. Any or all.
3.4. Export options Any or all fields can be downloaded, formatted ascii. ASCII with an option to select your platform.
3.5 Records Marking Yes. Done oddly. The user has a click box. After selecting records to display the user scrolls down to the bottom of the screen. The first box available for use at this point is View Marked Records. It would seem logical that after putting check boxes in next to the record I should be able to then view them. Unfortunately, after clicking on View marked records I received an error screen that told me I hadn't marked anything. Upon returning to the result screen, I discovered that the next box said Capture marked records. The capture screen has option for display, download and format. I am still unsure what the View Marked Records button was for and why it was placed before the capture button.
3.6 Reversed chronological results Yes. No. Ranked by relevancy. Not useful if you were not able to run an good search anyway.
3.7 Quality of full-text/image output N/A N/A
3.8 Type of printer required Any. Any
3.9 Fax/email/snail-mail No. Not available on a LAN system Yes. Email is available, not mail or fax.
3.10 Quality of output N/A For screen display the results are in courier and left justified. Very utilitarian.
3.11 Graphics/images N/A N/A
3.12 Smooth transitions down pages and from record to record Yes. No. There is a next group button the user must click in order to move on to the next group of records.
3.13 Linked to holdings No. No.
Comparative Search:

(rational near suicide) not euthanas* not adolescen* not elderly not book-review

Searched 1/74-12/97

18 results.

2 seconds.

Good results.

1963-current

Searched only

"Rational suicide"

couldn't run the full search that I ran in SilverPlatter.

Retrieved 14 records.

Took 2 minutes!

Results unsatisfactory since I had to "manually not-out" elderly euthanasia etc.

Comments:

csa- Being all things to all databases does not work.

csa- I can't even get CSA to run the same search I ran in SilverPlatter.

csa- takes 15 seconds just to get the error message that states the search was incorrect!

sp- Being windows-based on a lan does not allow users to cut and paste from within the interface.

Copyright © 1998 , University of North Florida, May 1998, This study completed by Reference Librarian Caryn Bush

  CSUL HOME | FCLA | WEBMASTER | UPDATE COMMITTEE MEMBER INFO
[ TOP ]