OHIO BOARD OF REGENTS

Resource Analysis Documents for
April 18, 1997 Meeting

Description of the Allocation Strategies Considered by the Resource Analysis Simulation Subcommittee

Component Parts of the Cost Allocation

 For the sake of simplicity, we partitioned the cost allocation process into four separate parts:

1. Allocation of each instructor’s salary to the course sections taught by the instructor,
2. Allocation of the remainder of the Schedule 5 costs to the Subject Codes and Levels of course sections supported by the department,
3. Allocation of the Schedule 4A costs, except POM, to the Subject Codes and Levels of course sections offered by the campus,
4. Allocation of the Schedule 4A POM costs to the Subject Codes and Levels of course sections offered by the campus.

In all four cases, the costs are aggregated by the Subject Codes and Levels of Instruction of the course sections. Remember, these Subject Codes represent the subject matter of the courses, not their source of funding.

 Cost Allocation Strategies Simulated

 Allocation of the Instructor’s Salary

For the basic allocation strategy, we compared allocating by Student Credit Hour (SCR) with allocating by Course Credit Hour (CCR). As expected, we saw that using SCRs tends to drive the costs to the lower levels, compared to CCRs. Opinions of the subcommittee members as to which should be used to allocate instructor’s salary were quite divided. CCRs were seen as the classic measure of instructional work load, used to drive salary for part-time instructors and determine load for full-time instructors. SCRs were also considered to be an adequate measure reflecting differences in time spent grading papers and one-on-one interaction with students. Most importantly, we recognized that the quality of the course section data is better in SCRs than in CCRs, and this difference is due to the practice of assigning more than one identification to the same course section.

We paid special attention to Individual Studies course sections. We felt that in these course sections, CCRs are overstated and tend to dominate the allocation. To control this, we tried two special strategies for Individual Studies course sections. In the first, for any instructor who taught any Individual Studies course sections, we allocated that instructor’s salary evenly among all of the course sections taught by the instructor. In the second, we simply allocated no part of the instructor’s salary to these course sections. We noticed that from the simulation data, nearly half of the instructors teach at least one Individual Studies course section in the course of a year.

We considered devaluing the CCRs for Individual Studies on the basis of average class size. For the simulation data from the participants, we had average class sizes of 23, 17, 19, and 16. So, considering CCRs for Individual Studies as 1/20 of other CCRs may be appropriate.

We paid special attention to course sections offered for zero credit hours. These occurred very infrequently in the simulation data, but we suspect that they occur more frequently at other colleges and universities. A strategy we tried was for any instructor who taught any zero credit hour course sections, we allocated that instructor’s salary evenly among all of the course sections taught by the instructor.

We noticed that SCRs have a distinct advantage over CCRs in the quality of the data. Many course sections have more that one identifier. When allocating by CCRs, these are counted more than once. Since the students enroll in only one of the course section identifications, the same problem does not effect SCRs. We resolved this in the simulation by providing special data to adjust the CCRs in this case. However, both the simulation participants and the Faculty-Staff Data Pilot project expressed concern about the cost of collecting this data.

A big problem in using SCRs occurs in the extremes. At one extreme, zero credit hour course sections generate zero SCRs, not withstanding class size. At the other, large lecture classes generate so many SCRs that they can dominate an instructor’s entire annual salary. We found a compromise in allocating by SCRs, based on ranges of SCRs.

We considered allocating each instructor’s term salary to the course sections taught in each term, but determined that reporting term salaries would be too costly in terms of data collection.

We considered allocating each instructor’s departmental salary to the course sections taught by that instructor in that department, rather than campus-wide. We determined that many instructors teach outside of their department and prorating their salary to all such departments is costly in terms of data collection.

We considered weighting credit hours, based on the assumption that high-level credit hours constitute a heavier load for instructors than low-level credit hours. We actually tried weighting graduate CCRs by a factor of 2 and found that it had little effect on the results. We could find no objective basis for supporting weights.

We considered capping the amount of salary allocated to a course section, or really a credit hour. We know there are some high-paid instructors who teach very few courses. In the end, we decided to deal with this case by encouraging institutions to restrict the reporting of instructional salary to the part of the salary relating to actual teaching.

Allocation of the Rest of Schedule 5

We assumed that the instructor’s salaries will be carried in Schedule 5 as well as in the instructor’s salary file. Therefore, in the allocation process, we subtract the instructor’s salary that is allocated above from Schedule 5. This is why we will need a link in the instructor’s file to the Schedule 5 that pays the instructor and why the Instructor’s salary needs to be consistent with the amount carried in Schedule 5.

Since we are aggregating costs in RA by Subject Code rather than Program Code, we needed to decide what to use as the departmental key to Schedule 5. We considered the possibility of identifying Schedule 5s with a high-level Subject Code, e.g., 32xxxx for the English Department, but we soon found out that the Subject Codes of the courses supported by departments frequently do not conform to a single area in the hierarchy of Subject Codes. Therefore, we decided to not have any statewide standard for funding departments. Campuses should use their own organizational structures and their own department codes. Course sections will be explicitly linked to the department supporting them, and instructors will be explicitly linked to the departments paying them.

For allocating a department’s overhead costs, i.e., the amount left after subtracting the instructor’s salaries to the course sections supported by the department, we simulated using CCRs, SCRs, and Instructor’s Salary. As expected, SCRs drive the costs to the lower levels compared to the other two allocators.

Here again, we recognized that CCR data has quality problems due to the practice of using multiple identifications with the same course section. We also recognized that departmental overhead costs are closely related to service to the students taking the courses supported by the department. This makes allocating by SCRs more reasonable.

Allocating Schedule 4A except for POM (Plant Operation and Maintenance)

No changes to this part of the allocation were recommended by the RA consultation, and none were simulated.

Allocating POM

We were motivated to try replacing the time-space data in the existing cost allocation strategy with SCRs for two reasons. Firstly, enrollment data is less costly to submit than CLUR, and secondly, we suspected that the replacement would have little impact on the cost allocation. We did this simulation using data from FY 1995 and found that it did, in fact, have little impact. Earlier we had simulated FY 1993 data and it, also, had little impact as was reported in the Resource Analysis Consolidated Report of August 18, 1995.

 With this change, the POM allocation strategy is:

1. Use NASF (Net Assignable Square Feet) to allocate each campus level POM expense to six identified campus POM expense categories: Student Services, Institutional Support, Library, Classrooms, Labs, Other Instructional Rooms.
2. Use weighted student enrollments to further allocate Library to Graduate and Undergraduate at the campus level.
3. Use SCRs to allocate Student Services, Institutional Support, and Undergraduate Library to Subject and Level.
4. Use instructor’s salary data to allocate Graduate Library to Subject and Level.
5. Use SCRs instead of time-space to allocate Classroom POM to Subject and Level.
6. Use NASF to allocate Lab and Other Instructional Rooms POM from the campus level to Subject Field. Note that this means that we will continue to need some means of mapping Rooms to Subject field.
7. Use SCRs instead of time-space to allocate Lab POM from Subject field to Level.
8. Use instructor’s salary to allocate Other Instructional Rooms POM from Subject field to Level.

Return to Meeting Notes, April 18, 1997

Return to HEI Front Door

http://regents.ohio.gov/hei/RA/notes/allocationstrategies41897.html
Last updated August 19, 1997