Back to Annual Meeting Page
|
133rd Annual Meeting & Exposition December 10-14, 2005 Philadelphia, PA |
||
Manuella Adrian, MScHyg1, Emma Ergon-Pérez, MA2, Brenda Lerner, PhD2, Michèle M. Jean-Gilles, PhD2, Jessy G. Dévieux, PhD2, and Robert M. Malow, PhD3. (1) (1)Department of Economics, (2)College of Health and Urban Affairs, School of Policy and Management, Universityof Guelph (1), Florida International University(2), 300 Bayview Drive Suite 1507, Sunny Isles Beach, FL 33160, 305-948-8023, madrian@the-beach.net, (2) College of Health and Urban Affairs/ AIDS Prevention Program, Florida International University, 3000 N.E. 151 Street, ACI-260, North Miami, FL 33181, (3) Robert Stempel School of Public Health/ AIDS Prevention Program, Florida International University, 3000 N.E. 151 Street, ACI-260, North Miami, FL 33181
Background: Prudent research management requires the archiving of original source documents (SD), e.g., questionnaires, in case they are needed during, or after, a project for research or audit purposes. Due to space and other resource constraints, paper archives are being replaced by electronic storage archiving such as CD-ROM/DVD of scanned documents, on the assumption that the scanned image is identical to and contains all the information in the original SD.
Methods: We examined to what extent the scanned SD image stored on CD-ROM is valid for research purposes, by analyzing SD information contained on a CD-ROM archive of alcohol or other drug use and risky sex behavior data, collected between 1999 and 2002 in Miami-Dade, for completeness, legibility, and record locator accuracy.
Results: Of 454 commercially scanned initial intake subject records, at least 11.2% were incomplete, having the following errors: Missing or non-existent subject records (3.3%); Missing or mislabeled record locators; Missing pages within a record; Incomplete pages within a record ; Incomplete words within a record or research project code book (e.g., unidentifiable variable names); Pages/data elements scanned out of order.
Conclusion: Because alcohol or other drug use and risky sex behaviors are socially stigmatized or illegal, there are known difficulties in ascertainment and obtaining accurate respondent information; errors may be compounded further due to current record archiving technology.
Learning Objectives:
Keywords: Data Collection, Biostatistics
Presenting author's disclosure statement:
I wish to disclose that I have NO financial interests or other relationship with the manufactures of commercial products, suppliers of commercial services or commercial supporters.
The 133rd Annual Meeting & Exposition (December 10-14, 2005) of APHA