Computer Science - policies

Academic Honesty Policy

rebelsky — Mon, 10 Aug 2015 12:58:25 +0000

Grinnell's Department of Computer Science has high expectations of Grinnell computer science majors and students in Grinnell computer science courses. Like all departments at Grinnell, we expect students to follow standards of academic honesty and to abide by the course rules and guidelines as stated in the course syllabus. However, experience suggests that students do not fully comprehend issues of academic honesty as they pertain to computer science. In this document, we explain our perspective and provide some examples.

Academically honest behavior is core to the academic mission. Members of the Grinnell community traffic in ideas, and it is important that these ideas receive proper credit. Members of the Grinnell community also care about the accuracy and reliability of ideas and data, and therefore object to actions that undermine such accuracy and reliability.

The Department of Computer Science also views academic honesty as an issue of professionalism -- we do not want to graduate computer scientists who put the public or their employers at risk by copying code without permission, by using code that they do not understand, by sharing information inappropriately, by pretending that their code works correctly when it does not, or by other violations. The professions that most our majors (and many other students in computer science courses) enter are full of occasions for unprofessional behavior that have grave consequences -- temptations that are often backed up with monetary incentives or significant organizational pressures. As a result, we work very hard to persuade our students of the importance of such academic virtues as love of truth for its own sake, accurate allocation of credit for new ideas, thoughtful consideration of the lives and interests of other people who may be affected by our decisions, candor, integrity, and humility.

What norms should we all observe in order to acquire and sustain these traits of character?

When we copy or adapt any part of a published work, or create a derivative work from it, or quote from it, we should formally cite that work, identifying the authors and specifying the facts of publication completely enough to allow any reader to confirm that our copy is accurate and that our adaptation or derivation is legitimate. Responsible scholars cite one another, even when the original authors publish their work under a license that specifically permits copying, both to ensure the correct allocation of credit and to enable readers to trace for themselves the development of the ideas expressed.
For the same reasons, when we copy or adapt unpublished work, in part or in its entirety, we should obtain the permission of the author(s), credit them explicitly, and acknowledge their priority.
For similar reasons, when we receive help from other individuals, we credit them explicitly.
It would be both arrogant and unfair to distort, sabotage, or interfere with the course work of others, or to prevent them from submitting it. We should refrain from seeking unfair advantages of this sort.
We should not lie to gain some advantage in a course, nor misrepresent our own work or the work of others, nor deceive one another about our work in more indirect ways. We should instead seek and respect the truth, even when doing so does not appear to be to our immediate advantage.
In some course assignments, we are required to make records that contain evidence related to their work, such as logs, interaction transcripts, and test runs. The value of such records lies in the fact that the evidence they contain might either support or refute a controversial claim, asserting (for instance) that a calculation or a program is correct or that all of the steps in an experimental protocol were carried out. Honesty therefore requires us to create such documents so that they record actual events and observations, without distortion, and to protect them against subsequent modification by anyone.
Section IV.A.4 of the Faculty Handbook specifies that the grades faculty members report "shall have been determined, in the final analysis, on the basis of the faculty member's own professional evaluation of each individual student's work." In order to reach such evaluations, instructors frequently require students to submit assignments and examinations as individuals, even in cases where students might be able to obtain better results through collaboration or by copying or adapting the work of others. Honesty and fairness require us to observe the restrictions that the instructors impose.
For the same reasons, instructors may limit or prohibit students from receiving outside assistance on certain assignments and examinations, whether from faculty members, class members, tutors, peers, stackoverflow.com, or whatever. If such help is prohibited, honesty and fairness require us not to seek it and not to offer it. If it is permitted, we have an obligation to acknowledge our receipt of it accurately, explicitly, and fully.
For the same reasons, even when allowing some kinds of outside assistance, instructors may limit or prohibit students from receiving other kinds of outside assistance. For example, some instructors might permit discussion of high-level design issues but disallow help with the expression of designs in program code. Honesty and fairness require us to understand what kinds of help are allowed, and to neither seek nor offer prohibited help.
We should not distribute solutions to assigned problems or answers to examination questions without the explicit permission of the instructor who assigned them. To do so would be dishonest and unfair to the students who are sincerely trying to achieve their best individual work.

When members of the department encounter cases in which students have failed to meet departmental or course standards of academic honesty or appropriate behavior, department members normally follow College guidelines and discuss those cases with appropriate officers of the Committee on Academic Standing (CAS) or the College Hearing Board (CHB). In many instances, CAS or CHB will take on the responsibility of determining an appropriate outcome for the case. In some instances, CAS or CHB will decide that the cases do not fall within their purview. In such instances, the failure to meet standards will be considered a violation of course or department policies and will be subject to penalties similar to those imposed by CAS or CHB.

Penalties imposed by the course instructor or department may include, but are not limited to,

A grade of zero (0) on any relevant work.
A reduction in the final grade in the course by the equivalent of one letter grade.
Failure in the course.
Expulsion from the major.
A recommendation to the College that the student be suspended.
A recommendation to the College that the student be expelled.
Any combination of these penalties.

Questions about this policy or the enforcement of this policy may be addressed to the department chair.

Faculty scholarship expectations

Thu, 27 Aug 2009 16:04:54 +0000

endorsed by the Department of Mathematics and Computer Science on October 11, 2004

Note (July 13, 2006): The following document was developed jointly by the computer science faculty and the mathematics faculty when both were part of a single Department of Mathematics and Computer Science. With the restructuring into two departments on July 1, 2006, the administrative structure has changed, but the principles regarding scholarly expectations continue for both departments.

The Department of Mathematics and Computer Science believes its faculty should be interested and involved scholars. Since the Department celebrates the diversity of scholarship by its faculty, the Department believes it must not be overly prescriptive in stating just what might or might not be adequate and appropriate regarding professional involvement. Rather, the Department identifies three general principles for the scholarly activity of its faculty.

Faculty should be able to provide evidence that they are working in their field(s) as engaged scholars.
The Department endorses a wide range of scholarly activities as being appropriate and worthwhile.
Faculty efforts should include activities of scholarship that are peer reviewed.

Engaged Scholars: In mathematics, statistics, and computer science, important new questions are always arising for scholars to address. Also, important old questions may gain new relevance and be amenable to new insights and methods. Further, engaged faculty provide important role models to students regarding the intellectual excitement and challenge of their fields. We believe that, to be effective over the long term, faculty must continually be engaged in their changing and expanding disciplines.

Scholarly Activities: The Department strongly endorses the breadth of scholarship identified in E. L. Boyer's report [1]. That is, we believe that appropriate scholarly activity includes the scholarship of discovery, the scholarship of integration, the scholarship of application, and the scholarship of teaching.

Peer Review: Faculty should have the products of their scholarship tested and refined through interactions. While some activities might not be peer reviewed, the Department uses peer review as a measure to ensure scholarly endeavors meet high standards of rigor and quality.

Discussion and Examples

Although these three principles provide some guidance regarding appropriate scholarly activity, abstract statements sometimes can be difficult to interpret and apply. Thus, the Department offers several examples to help clarify appropriate endeavors. We emphasize, however, that these examples are not comprehensive or complete. Rather, they are meant only to suggest the type of activities that the Department believes fit well within the principles stated.

Faculty as Engaged Scholars

Scholarly engagement normally involves such qualities as focus, intellectual development, and creativity. A scholar expands her or his background in a subject, develops new perspectives, integrates new ideas with past understandings, organizes, and synthesizes. The form in which this engagement is manifested has several models:

A traditional approach to scholarly engagement involves focus on a research project. Intellectual development comes from learning new developments in the field, and creativity yields insights for expanding the field's body of knowledge.
Scholarly engagement may emphasize integration, organization, and synthesis -- perhaps leading to a textbook, lab manual, survey article or edited anthology. In this context, focus involves putting pieces together; intellectual development includes the restructuring and synthesis of ideas; and creativity yields new perspectives and structures.
Scholarly engagement may involve the creation of software, multimedia, or other materials to support research or in the development of learning materials. Such activity draws heavily on multiple disciplines, focusing on new ideas, viewpoints, techniques, and relationships.
Often a faculty member provides her or his own direction and scholarly focus, but other models are possible and appropriate. Here are two examples.

A faculty member with broad experience may be asked by others to consult regularly. In these circumstances, the outsiders provide an on-going series of scholarly problems. Sometimes the faculty member might draw largely upon past experience, with only marginal scholarly engagement. In other cases, however, the consultant may need to learn new areas, integrate ideas, and play a leading role in part of the research. This may lead to considerable intellectual development and creative problem solving.
A faculty member may develop extensive professional contacts over time, and discussions with these colleagues may yield an on-going stream of scholarly activity. The focus of work may depend in part on the interests of the other colleagues, but the collaborative investigations may yield significant scholarly development for all involved, and the interactions may spark considerable creativity.

In any form, scholarly engagement requires active involvement, not just passive observation. Although faculty are encouraged to attend conferences for exposure to new ideas, simple attendance does not guarantee active involvement. To demonstrate full engagement, faculty attendance at conferences should include active participation -- at least periodically.

Whatever the nature of the scholarly activity, the scholarship should show evidence of clear goals, adequate preparation, appropriate methods, significant results, effective presentation, and reflective critique. Altogether, a faculty member should be active within the discipline, using appropriate methodology, creativity, and focused intellect.

The Range of Appropriate Scholarly Activities

With the breadth of scholarship identified by Boyer, this department wishes to be particularly cautious in listing what scholarly activities might be appropriate. We have chosen here simply to list, with annotations, some relevant contributions made in the past by department members. These examples are meant to illustrate, not limit, the range of appropriate activities. The length of the list is not intended to suggest that each individual ought to engage in more than a few such activities, and the ordering of the list is not intended to indicate any inherent preference or value for one type of scholarship over another.

article in a research journal: This represents the most traditional mechanism for organizing and presenting new work in a subject area. Publication may be in paper or electronic journals.
article in a teaching journal: Although some educational articles may be descriptive and informal, opportunities abound for innovative experimentation, careful methodology, and formal analysis.
article in a conference proceedings: In some disciplines, the preferred mechanism for publishing scholarship is through a conference proceedings. This allows for quick dissemination of written results within the focused community of a conference.
conference and colloquium talks: The practice of presenting one's scholarship to an audience allows for a sharing of ideas in an environment of intellectual energy and immediate feedback. Examples span the range from informal, relaxed discussions to reviewed, formal presentations. Invited presentations are particularly noteworthy.
panel talks: When presentations involve several participants, the level of scholarship for the participant may be difficult to judge. The effort of the session organizer, for example, may be much greater than that of the panelists. Involvement with invited sessions may suggest heightened effort and scholarship.
workshops: Many conferences hold extended sessions that provide conference attendees new insights in emerging research and technology. Session leaders must synthesize much new material and present it within a highly-constrained framework.
consultations: Some faculty work regularly with those in other departments. This provides an interdisciplinary perspective on scholarly endeavors.
student-faculty research, perhaps leading to a published paper or presentation: Mentored Advanced Projects (MAPs) and guided independent projects involve mentoring of students while moving research along. Progress often is slow, as faculty expend considerable energy and time in bringing students up to speed.
professional boards: Programs and policies only make sense if they fit with perspectives and understandings in the discipline. Thus, involvement in national committees and policy boards requires participants to draw upon a broad understanding of their disciplines.
posters provide a fine mechanism for getting feedback on interim results. These can be particularly relevant to projects involving student-faculty research.
books: Published books may have a research focus or may synthesize material yielding a textbook or laboratory manual. Either emphasis is appropriate in this department.
software development: The development of large software packages and applications draws upon a deep knowledge of the application field, extensive problem solving, thoughtful design, integration of algorithms and data structures, and the innovative integration of ideas.
development of teaching materials: Considerable scholarly activity can be involved in the development of an extensive package of audio, video, and/or paper-based teaching materials. (We do not regard the routine preparation of class handouts as a significant scholarly activity.)
posing or solving published problems: Much of mathematics, statistics, and computer science relates to the on-going identification and solution of interesting problems. Often this work may occur informally, such as the work of the faculty to find solutions to problems posed in the annual Putnam Mathematical Examination. Sometimes, however, this activity may become rather formal and systematic. For example, the American Mathematical Monthly solicits statements of interesting problems, identifies those who have submitted correct solutions, and publishes the best results.
refereeing papers and proposals: Many journals, conferences, and granting agencies invite faculty to review submitted materials -- either individually or in groups. A serious referee's report requires thoughtful consideration of new material.

Peer Reviewed Activities

Peer review offers the broadest accepted mechanism for ensuring the approval by an intellectual community of a faculty member's scholarship. While some of a faculty member's scholarly output may not be peer-reviewed, having one's work reviewed by one's peers provides a convincing means of assessing the scholarship's value. Several examples of types of peer review follow.

Many fields within pure mathematics follow a traditional process, in which refereed journals provide the primary forum for scholarly materials. This format is consistent with practice in many other academic disciplines.

Within computer science, peer-reviewed presentations provide the preferred mechanism to disseminate scholarly results. The best conferences, for example, are extremely selective, often accepting only 25%-35% of the papers submitted. Unlike many disciplines, the field of computer science changes so quickly that conferences provide the primary forum for the communication of many results; books and other monographs simply take too long to appear. This is described more fully in ``Best Practices Memo. Evaluating Computer Scientists and Engineers For Promotion and Tenure'' [2], which describes practices for both theorists and experimentalists:

For theorists, ``conference publication is highly regarded in the theoretical community'', although results are often rewritten for other journals (following other disciplines to some extent). The CRA Memo continues, ``For experimentalists conference publication is preferred to journal publication, and the premier conferences are generally more selective than the premier journals [...]. In these and other ways experimental research is at variance with conventional academic publication traditions.''

Within statistics, the review of statistical consulting may occur in several ways. The American Statistical Association explains possible review and assessment with the following statement (see [3]):

Mathematical sciences departments should also recognize the value of statistical consulting as a legitimate and important form of scholarship and professional development. This can involve:

consulting on projects that may lead to joint authorship on peer-reviewed publications
consulting on scholarly projects even if joint authorship is not attained
consulting on student research projects
consulting on commercial projects that may involve proprietary information that precludes peer-reviewed publication

The first of these can be evaluated by the usual peer-reviewed means, although the department should recognize that the journal may be in the applied discipline rather than in statistics. Such consulting would not likely result in sole- or first-authorship but can nevertheless be very valuable. The second and third of these can be assessed through testimony of the scholars and faculty members for whom the statistician performed the consulting. The fourth of these can be evaluated through testimony of the client.

Rather than publishing shorter peer-reviewed pieces, some faculty members will work on large-scale projects which yield books put out by established publishers. Any reputable press will pass a book through an imposing review process before a contract is agreed upon, followed by extensive editorial work. Such practice rises to the level of peer-review.
In several disciplinary areas, granting bodies, such as the National Science Foundation, utilize an extensive and rigorous peer review process, with acceptance rates of approximately 25%.

Practices differ in various sub-disciplines of mathematics, statistics, and computer science regarding preferred venues for the dissemination and publication of scholarly materials. No faculty member, however, is necessarily tied to a specific form of peer review. While peer-reviewed work is certainly expected, an holistic view which incorporates all of one's scholarly activities, including those which overlap with teaching and service, is employed. Consultation with the department chair and the Dean is recommended to address individual questions as to whether the quantity and quality of one's scholarly activity are sufficient for personnel decisions.

Select Bibliography

E. L. Boyer, Scholarship Reconsidered: Priorities of the Professoriate, a report for the Carnegie Foundation for the Advancement of Teaching, 1990.
D. Patterson, L. Snyder, and J. Ullman, "Best Practices Memo. Evaluating Computer Scientists and Engineers For Promotion and Tenure", Computing Research Association, 1999.
American Statistical Association endorsement of the Mathematical Association of America "Guidelines for Programs and Departments in Undergraduate Mathematical Sciences." (undated web page)

Graduation with honors

stone — Mon, 03 Aug 2009 19:16:08 +0000

According to the current Grinnell College catalog, Each department recommends for graduation with honors those senior majors who have clearly distinguished themselves within their major field of study, subject to the approval of the Dean of the Faculty. The catalog imposes the following condition: In order to qualify for recommendation, a student must, after seven semesters of college work, have achieved at least a 3.5 grade-point average in the major field and a cumulative grade-point average of 3.4. The required seven semesters need not all be completed at Grinnell ...; however, only credits completed at Grinnell and Grinnell-in-London will be used in determining grade-point eligibility.

The catalog also specifies that these are minimum standards and should not be regarded as the only criteria. For majors in computer science, the Department of Computer Science considers the following guidelines in addition to the college's grade-point requirements:

To be considered for honors in computer science, graduating seniors must not only meet the College’s general requirements for honors but also demonstrate exceptional commitment to the discipline and its values, as evidenced by significant engagement in the department and excellence in computing-related work, both in the classroom and beyond.

In the past, we used a much more detailed list of criteria. We have decided that such a detailed list is not appropriate. We keep it here for historical purposes. It should not be interpreted as a guideline for what needs to be done. Talk to a faculty member in the department for more details.

A. Core courses of study

Completion of Computer Science 211 or Physics 220, and
Completion of Computer Science 213

B. Additional course work that is not used to fulfill another requirement

Completion of a 200- or 300-level course in computer science, or
Completion of a statistics course at the 200-level or higher (MAT 209, 309, 335, or 336), or
Completion of MAT 220 (Differential Equations), MAT 306 (Mathematical Modeling), MAT 314 (Topics in Applied Mathematics), or MAT 321 (Foundations of Abstract Algebra), or
Completion of Physics 220 (Electronics), or
Completion of PSY 222 (Industrial Psychology) or PSY 260 (Cognitive Psychology).

C. Participation in local activities related to computer science, judged to be excellent by department faculty; such activities might include

Giving talks at Extras sessions, or
Doing independent projects (totaling four credits or more) in computer science, or
Carrying out research under the direction of a member of the department, or
Developing a successful software package with positive assessment by department faculty.

D. Participation in the study or use of computer science outside of the department, judged to be excellent by department faculty; such activities might include

Achieving a score at or above the 75th percentile on the Graduate Record Examination in Computer Science, or
Receiving an award in the Mathematical Competition in Modeling, or
Achieving a strong performance in the ACM Programming Competition, or
Having a paper accepted by a refereed computer science journal or conference, or
Developing a successful software package with positive assessment by outside referees or evaluators, or
Contributing to an open-source software package or resource, or
Carrying out supervised research elsewhere.

Policies

stone — Mon, 03 Aug 2009 19:04:17 +0000

Learning goals and objectives: a statement of the learning goals and objectives for majors and non-majors within the Department of Computer Science
Faculty Scholarship Expectations: a 2011 working document regarding expectations of computer science faculty for scholarship. An earlier, 2004 document from the combined Department of Mathematics and Computer Science also is available.
Graduation with honors: the guidelines by which the department's faculty decide which graduating seniors merit honors
MathLAN use policies: appropriate use of the department's local-area network
Advice to authors: guidance for users who publish through the department's Web server
Disclaimer: alerting visitors that the department does not censor users' Web publications