Astra Schedule Help (7.5)

Implementation Considerations

Hide Navigation Pane

Implementation Considerations

Previous topic Next topic  

Implementation Considerations

Previous topic Next topic JavaScript is required for the print function  

Platinum Analytics is primarily a solution that facilitates improvement in the course schedule development and refining process. A successful implementation should, at a minimum, provide decision-support information to academic departments, highlighting a portion of their schedule that might be changed.

Important points to consider when implementing Platinum Analytics:


1.The focus of this form of analysis should be on identification of “change candidates.” These are the courses wherein the analysis suggests needed change in a highly predictive way. Other courses will fall into either a “no change needed” or a “outlier” category. While review of these two categories is important, it should not distract from the “change candidates” since those candidates represent the opportunities for improvement in the analysis term.


2.Typically, only a small group of high-impact changes are urgently needed in any schedule. Changing offerings in 20-50 of 1,000 courses is not uncommon or undesirable. Change should be thought of as continuous improvement – making the roll-forward schedule a little better in high-impact ways each term. That way, progress is made without the unnecessary political fallout that can occur from drastic and immediate change.


3.Courses in the “outlier” category are less predictive for a variety of reasons:


a.Courses might be in a degree rule that is needed by many students but is rarely if ever offered. This group of courses will be eliminated, or at least greatly reduced, with the introduction of Predictive Program Analysis (future).


b.Courses might be discontinued and replaced by other courses. This group should be eliminated, or at least greatly reduced, with the introduction of Course Mapping (future).


c.Sub-sets of source data regarding students, degree rules or tentative schedules might be incomplete or inaccurate. (Examples:  Degree rules for graduate students have not been built; section information is incomplete for the analysis term.)


d.Program Analysis might be misinterpreting a combination of student data and degree rules in a subset of the analysis.


4.Beyond existing development plans to address 3a and 3b, the Ad Astra team will work with your implementation team to reduce the impact of items 2c and 2d. This is also a continuous improvement process wherein the goal should be reducing this pool of ‘outlier’ courses each term. While Ad Astra’s investment in stabilizing and improving the software continues, it is critical to understand that “outliers” – of any type – do not necessarily invalidate the merits of using “change candidates” data for strategic scheduling enhancements.