retrieved (i.e., what queries are possible). Database design starts before
data collection begins. Without an adequately designed database, the data
that have been acquired at a large cost may not be usable to effectively
address the problem. Forms should be an integral part of the design of the
database and should accommodate transcription or import into the database (Conquest et al. 1993). With a good design, field sampling can often
be improved with preformatted data sheets or integrated digital data collectors. Field crews can be a source of considerable error if data forms are
not designed to limit errors (Conquest et al. 1993). Database design has to
consider the final uses to efficiently serve the data to users.
10.5.2 Data Input
During data input, it is imperative to check and cross-check the data as they
go into the database and again after data entry is complete to ensure
the accuracy of the data. Handheld computers, personal digital assistants
(PDAs), and forms that can be optically read on a computer are becoming
common for data input. Regardless of the type of input, quality assurance/quality control (QA/QC) measures are required to ensure data
accuracy.
10.5.3 Data Standards
The development of standards sometimes requires more effort than data
collection and storage. There are many efforts from federal and state government agencies along with private industry to promulgate standards for
data collection, measurement, QA/QC, archiving, and metadata.
The use of standards is important in developing databases that may be
shared outside the realm of the initial data developer. While individual
researchers have intimate knowledge of how, what, when, where, and why
the data were collected and developed, other users may not be able to use
the data unless adequate standards are used and the data are well
documented.
10.5.4 Hardware and Software Concerns
Data storage technology seems to be growing as fast or faster than
Moore’s law (that computer power doubles every 18 months). Computers
on the individual desktop now have more power and storage capacity than
whole data centers of the recent past. Data centers have progressed to the
point that the storage and manipulation of terabytes of data is currently
feasible. The integration of data servers, operating systems, and server software now requires full-time support to keep the data available. The
complexity of maintaining a software-based server requires a significant
investment.
200
David Hohler et al.
data collection begins. Without an adequately designed database, the data
that have been acquired at a large cost may not be usable to effectively
address the problem. Forms should be an integral part of the design of the
database and should accommodate transcription or import into the database (Conquest et al. 1993). With a good design, field sampling can often
be improved with preformatted data sheets or integrated digital data collectors. Field crews can be a source of considerable error if data forms are
not designed to limit errors (Conquest et al. 1993). Database design has to
consider the final uses to efficiently serve the data to users.
10.5.2 Data Input
During data input, it is imperative to check and cross-check the data as they
go into the database and again after data entry is complete to ensure
the accuracy of the data. Handheld computers, personal digital assistants
(PDAs), and forms that can be optically read on a computer are becoming
common for data input. Regardless of the type of input, quality assurance/quality control (QA/QC) measures are required to ensure data
accuracy.
10.5.3 Data Standards
The development of standards sometimes requires more effort than data
collection and storage. There are many efforts from federal and state government agencies along with private industry to promulgate standards for
data collection, measurement, QA/QC, archiving, and metadata.
The use of standards is important in developing databases that may be
shared outside the realm of the initial data developer. While individual
researchers have intimate knowledge of how, what, when, where, and why
the data were collected and developed, other users may not be able to use
the data unless adequate standards are used and the data are well
documented.
10.5.4 Hardware and Software Concerns
Data storage technology seems to be growing as fast or faster than
Moore’s law (that computer power doubles every 18 months). Computers
on the individual desktop now have more power and storage capacity than
whole data centers of the recent past. Data centers have progressed to the
point that the storage and manipulation of terabytes of data is currently
feasible. The integration of data servers, operating systems, and server software now requires full-time support to keep the data available. The
complexity of maintaining a software-based server requires a significant
investment.
200
David Hohler et al.