NSF initiates program targeting junior faculty

By Gerald L. Engel

The National Science Foundation initiated the Faculty Early Career Development (CAREER) program in 1995. The purpose of this NSFwide program is to "strongly encourage the early development of academic faculty as both educators and researchers."

The program replaced the NSF Young Investigator program, the Minority Research Initiation program and the Research Initiation Award program of the Computer and Information Science and Engineering Directorate and the Engineering Directorate. The CAREER program is open to "junior faculty who intend to develop academic careers involving both research and education."

The deadline for the CAREER program varies by discipline. This year's CISE deadline is expected to be sometime in the fall. Awards come out of the individual programs within NSF. For more information, contact the officer for the program that best fits the proposal. (See the NSF federal funding chart on Pages 6-7 for help in locating the appropriate CISE program officer.)

Because the program was designed to address the research and education aspects of an investigator's developing career, the program announcement (NSF 94-101 (New)) required a Career Development Plan consisting of a) a research plan, b) an education plan and c) a departmental endorsement. The entire plan was not to exceed 15 single-spaced pages.

Research Plan: The research plan should have no more than 10 single-spaced pages. It should have followed the standard NSF proposal format, and it should have presented:

• The objectives and significance of the proposed research.
• The relation of the research to the current state of knowledge in the field.
• An outline of the general plan of work, including a description of the methods and procedures to be used.
• A summary of prior research accomplishments.
• A brief description of the impact of the proposed research (refer to GPG, Section 11.C.12.d) for proposals submitted to the Directorate for Engineering.

One hundred thirty-six proposals were submitted to the six programs (Theory of Computing; Software Engineering; Numeric, Symbolic and Geometric Computation; Computer Systems; Operating Systems and Systems Software; and Programming Languages and Compilers) of the Division of Computer and Computation Research (CCR). Of these proposals, 81, or about 60%, had research plans that were 10 pages or longer--seven exceeded the page limit. The average length of the plans was 9.32 pages. And the plans, on average, contained 42.46 references to the research literature.

By way of contrast, the education plan was much more specific, perhaps anticipating some significant confusion regarding this aspect of the proposal.

Education Plan: The education plan should not have exceeded five single-spaced pages. It should have addressed the applicant's planned education activities over the proposed award period as appropriate to the development of a full, balanced academic career and described the applicant's education accomplishments. It should have included:

• The objectives and significance of the proposed education activities in relation to the applicant's career goals and the goals of the institution.
• A summary discussion of the education activities the applicant plans to undertake.
• A discussion of the types of teaching activities the applicant plans to undertake during the award period or a statement of equivalent planned educational responsibilities. (Teaching assignments, if known, may have been included.)
• A summary of teaching and other education accomplishments.

In addition, some administrative details (the effective date of the applicant's initial full-time tenure-track or equivalent appointment) and a statement endorsing the proposal by the administrator responsible for the program were to be included in the plan. These items typically took half of a page.

The CCR review panel expressed some significant concerns regarding the nature of the education plan. While agreeing with the concept of conveying the importance of teaching, there was concern that not enough thought had gone into the implementation in this particular program. It was noted, for example, that a new assistant professor has rather limited latitude in program development. It also was noted that although the proposers have records of accomplishment in research that can be evaluated, they have relatively little experience in teaching. Therefore, the metrics to use are far from clear.

Looking more objectively at the education plans, the average plan was about 3.33 pages. About 14% of the plans were five pages. The education plans averaged 1.18 references. And 98 proposals (72%) had no references. Only six of the education plans (4%) included some concept of evaluation.

The typical plan cited some success as a teaching assistant, a desire to introduce a new course in the area of specialty and a desire to involve undergraduates--often from underrepresented populations--in the proposer's research activity. Although all of these are commendable goals, such a description is difficult to classify as a plan, especially when there is no reference to the literature of computer science and engineering education and no concept of evaluation of the results.

If we are to be successful in conveying the view of the importance of teaching in the career development of young faculty, additional work will be required both inside and outside NSF. There is a rich literature in computer science and engineering education, and this needs to be conveyed to our graduate students and recent graduates. There is also a growing literature of effective methods of project evaluation--for example, NSF 93-152: User-Friendly Handbook for Project Evaluation: Science, Mathematics, Engineering and Technology Education. The importance of approaching education with the same sense of inquiry, quality and scholarship with which research is approached must become a priority.

Gerald L. Engel is program director for special projects in NSF's Division of Computer and Computation Research. He is on assignment from the Leonhardt Chair of Computer Science and Engineering at the University of Connecticut at Stamford.