Use of the On Demand Dataset tool to extract information from the warehouse has a number of advantages over other data query tools, such as Hyperion, PSQuery, or MS Access.
Most reports are only required once a term or once a year and a typical end-user doesn't use a query tool often enough to be comfortable creating queries. ODD jobs are easy to request, submit, and schedule and don't generally require any training. The output can be manipulated using MS Excel, Word, and Access, allowing for the users to easily create and modify reports based on this output.
- ease of use by gate-keepers and advanced users
The views and functions used by ODD model our local business practices and ensure that common oversights do not occur. They have been optimized by analysts to eliminate the need for joins and remove the many potential pitfalls associated with one-to-many relationships and inner/outer joins. The existence of predefined populations and output lists makes it easy for the gate-keepers to construct new data sets and create new jobs for end-users.
The field names and values generated by ODD are standarized across all views The criteria used to select populations are also standardized. Since ODD views are based on the database history tables, the output generated by different ODD jobs should always match each other and data presented in web reports. If not, the discrepancies can be reseached by an analyst. This is much more difficult when a variety of people produce reports using a variety of query tools.
A common reporting requirement is to generate a list of records that meet a certain criteria who were not included the last time the list was generated. This capability is built into ODD, allowing users to select either all or incremental results from any dataset without having to remember or know when the job was last run. Figuring out when a student became a member of a particular population can be extremely complex and is often outside the capabilities of other query tools.
- increased flexibility in population selection
Because ODD populations are based on functions, they have the ability to use selection criteria that are difficult to express in other query tools. For example, "the prospects who have visited campus" or "students who have ever been on probation" or "students who mentioned an ethnic background".
The views and functions used by ODD have been created by analysts and optimized for performance. Nested queries and outer joins created often have huge performance penalties that can be addressed better in the more controlled ODD environment.
The gatekeepers can determine which users are allowed to receive which data. After the gatekeeper has set up a dataset, only approved end-user can generate that data. These makes it easier to ensure compliance with FERPA guidelines while still providing data to a variety of end-users.
A history of ODD data sets is maintained in the database, allowing analysts to research the criteria used to generate any data set that has been previously generated.
Very few data sets are requested only once--most will be desired again at some point in the future. Unfortunately, the future is often a year or two later and users have forgotten the original criteria used to generate the earlier data set. Since ODD can store a dataset definition and always stores a history in a central location, it is always be possible to recreate a data set at a later date.
Comments (0)
You don't have permission to comment on this page.