Students should be able to:
Students entering the PhD program must complete a Master's degree in Computer Science or a closely related field as partial fulfillment of the requirements for the PhD (those students entering the PhD program with an MS in Computer Science or a closely related field meet this requirement and should continue with the remaining PhD requirements). Under direction of the student's advisor, and with the approval of the student's PhD dissertation committee, the Master's thesis defense can be used to fulfill the paper requirement of the qualifying process.
A graduate student should submit a study list in their first semester of study and must submit a study list no later than the second week of the second semester after admittance. Program of Study Form.
The table below reflects those hours that can be used to fulfill PhD course requirements. Specific courses depend on approval of the student's committee.
*Note that if an MS has been completed and if the 30 MS hours have been accepted as transfer credit, then only 18 hours of additional coursework remain.
Justification for PhD Course Requirements:
NOTE: If you were accepted on or before Winter 2016, you must complete the Breadth Requirement. Please see the MS handbook for details. As indicated above, you must take CS 611, which satisfies one of breadth groups.
PhD students are expected to be resident for the full duration of their PhD program.
Performing research of sufficient quality to merit a doctoral degree takes considerable time, effort, and personal investment. Residency is intended to help students make consistent progress in such research by keeping students (a) engaged with their faculty advisers, (b) progressing on solving doctoral-level research problems, and (c) contributing to the lab and department community of scholars. Simply put, students who are resident are much more likely to produce high quality research and graduate in a timely manner, while contributing to the mission of the BYU CS department.
Resident means being present during most business hours, and does not mean working 40 hours/week (usually it is more) or registering for some number of credit hours.
Petitions for exception to the residency requirement must be made in advance to the student's doctoral committee and must demonstrate how the absence from campus is (a) required for health or unusual personal reason or (b) will help the student make progress toward their doctoral degree. Other reasons for exception and exceptions lasting longer than 6 months will rarely be granted. Once the doctoral committee has approved the petition, it should be submitted to the graduate coordinator for final approval.
The University requires PhD students to register for at least two consecutive 6-hour semesters on the BYU Campus.
A PhD student must teach at least one Computer Science course. Make requests for teaching by indicating your desires for teaching, both the semester and course, at the time you submit your report for progress reviews. These requests should be made well in advance because the department plans teaching assignments in January for the following academic year. Teaching requests for a given academic year should be submitted during Fall semester of the prior academic year, which is eighteen months in advance for those planning to teach in Spring or Summer term.
This gives the student an opportunity to improve teaching and communication skills. In addition, this requirement provides the student an opportunity to see how he/she likes teaching and the possibility of an academic career.
The graduate faculty of the department evaluate the progress of every PhD student two times per year during January and September. The date, time and place will be announced several weeks in advance. All faculty members are invited to attend, and those with PhD students are required to attend. The progress of a student can be declared as satisfactory, warning, marginal, or unsatisfactory. After each review, a letter will be sent to each PhD student. The letter to those students in warning, marginal or unsatisfactory status will indicate what possible steps the department will take as a result. Examples of possible department actions include, but are not limited to, requiring the student to write a letter of explanation to the department explaining warning, marginal or unsatisfactory performance, suspending financial support, or immediately terminating the student's PhD program.
Our objective is to make sure that students are on track, to provide encouragement when it might be needed, to make sure no one "falls between the cracks," and to guide anyone who is not likely to succeed to something better.
In preparation for the progress review, PhD students create or update their online Graduate Profile. The profile provides an indication of each student's progress in the program goals for accomplishing significant milestones. The Graduate Academic Advisor will remind students at least 2 weeks before the scheduled date of the Progress Review.
Students may be dropped at any time as determined by their Committee, the Graduate Committee and the Department Chair.
With the help of the student's advisor, the student selects faculty members to serve as the 2nd, 3rd and 4th committee members.
Following the PhD Qualifying Process, PhD students meet with their PhD committee annually to review progress and receive input. The student is responsible for organizing this meeting.
Regular contact between the PhD student and their graduate committee helps the faculty contribute effectively to the education and professional preparation of the student. Ideally, the student should meet in person with their full committee, but four out of five members is acceptable. Committee members may participate remotely. During these meetings, the student presents a 15-minute accounting of their progress. The graduate committee tracks the time since the last meeting as part of the PhD review. The dissertation proposal may be used to satisfy this requirement during the year it is held.
The primary purpose of the PhD qualifying process is to determine if a student has the skills necessary to continue towards a PhD degree. Secondary purposes include a demonstration of (a) breadth of knowledge at the graduate level and (b) ability to think, learn, and work independently and the ability to write and present clearly. Students entering the PhD program with a Master's degree should complete their qualifying process within one year after entering the program. Students entering the PhD program with a Bachelor's degree should complete their qualifying process within one year after completing their Master's degree.
A student may use his/her MS thesis defense to qualify for his/her PhD. If a student chooses to do so, the student will need to make sure that a complete PhD Committee is in place (see above) prior to the thesis defense, and that all members of the PhD Committee are in attendance. The earliest a student can qualify is at the time of their MS thesis defense.
Under the direction of the student's PhD advisor, a student must produce a research paper suitable for submission for publication and defend the contents of the paper in a public presentation.
To be "suitable for submission for publication," the research paper must be a full-length manuscript (not an abstract or extended abstract) and either should have been submitted for publication or should be in a form ready to submit for review to an identified journal or conference.
Turning an MS thesis into a paper under the direction of the student's PhD advisor is a typical way to satisfy this requirement. In this case, the thesis defense can also serve as the public presentation for the PhD qualifying process, but only if the student has (1) completed the breadth courses, (2) turned the thesis into a paper for submission, and (3) invited the PhD committee to attend the MS thesis defense.
Based on the paper and presentation (as well as other interactions between the student and members of the student's PhD committee), the student receives an assessment of independent ability on each of five criteria: Think/Learn/Work/Write/Present. Each committee member makes an assessment for each of the criteria on a scale of 1 (worst) to 5 (best), or N/A when a committee member has had insufficient evidence to make a determination. The score for each of these criteria and the overall score are averaged, with N/A's omitted, for the final score.
Scale: 1 (poor), 2 (inadequate), 3 (marginal), 4 (adequate), 5 (good).
Further explanation of Think/Learn/Work/Write/Present:
Think: be able to independently reason out, conceive of, and envision key problems and ideas.
Learn: be able to independently ascertain, discover, and understand significant thoughts explained in the literature and in presentations.
Work: demonstrate persistence in producing results, both independently and under direction.
Write: have a good command of the English language; clearly organize and articulate thoughts and ideas; show reasoning and analysis skills in the flow of discourse.
Present: be able to clearly organize and explain ideas, motivate the work in a convincing way, explain its significance, and respond articulately and knowledgeably to questions.
There are three cases.
A student who fails once may pass and be advanced to PhD candidacy by rectifying problems before the end of the 24th month after enrollment in the PhD program. A student who fails to rectify problems by the end of the 24th month will be dropped from the PhD program.
To rectify problems with breadth courses, for example, a student may take another course or two and use a different set of four courses to satisfy the breadth requirement. To rectify problems with producing a suitable research paper, for example, a student may prepare another paper for publication.
When a student has completed the breadth coursework and has produced the research paper, the student schedules a time for the public presentation of the paper with their committee. This is done according to the instructions on the PhD Qualifying Process Form by talking with all five committee members to find a day and time that they are all available and having them sign the qualifying process form, schedule a room for the agreed upon date and time, and finally turning in the completed form to the Graduate Academic Advisor, which should be completed at least one week prior to the scheduled date. (You’re not officially scheduled until this form has been submitted.)
Remember that the presentation should not be scheduled sooner than one week after all committee members have received a copy of the research paper. All members of the student's PhD committee must be present. The student should schedule one hour for the meeting: 25 minutes for the public presentation, 5 minutes for public questions, and up to 30 minutes for the non-public assessment.
The dissertation proposal comprises a document and an oral presentation. The proposal presents a problem, discusses closely related work and the broader research area, and describes a research agenda to solve the problem and evaluate the solution. The proposal must explain the importance of the research problem and the research agenda to those who are not experts in the field.
The document must be at most 15 single-spaced pages, exclusive of the title page, abstract and references. The text must be single-spaced with 11 pt. Times Roman or LaTeX default (Computer Modern) font and have one-inch margins in a single-column format. A very useful LaTeX template can be found at our Template Download page.
The presentation will occur during a 2 hour meeting block. The public presentation and questioning will last for a maximum of 1.5 hours. Following the public presentation, faculty will have 30 minutes to conduct private questioning of the student and to arrive at a decision. Questions by the committee must test the student's knowledge in both the general research area as well as the details of the proposed research.
The following are best practices guidelines for the dissertation proposal. Their purpose is to provide useful, detailed guidance to students and advisors on creating a successful proposal, while allowing flexibility to accommodate individual needs and preferences.
The dissertation proposal should include a discussion of both the broad research area and the narrow research problem. To illustrate this, consider the picture shown at right. Your field is very broadly defined (e.g., graphics, machine learning, security, networking). The area in which you work is somewhat narrower, and considers a set of problems within your field (e.g., Decentralized Authentication). Finally, your dissertation topic is very narrowly focused on a particular problem within your area (e.g., Convenient Decentralized Authentication using Passwords). The dissertation proposal should cover the 20 most important research papers in your area when discussing the area background. It should then separately cover the most directly related research papers specific to the problem you are trying to solve. In most cases there will be only a small overlap between these two sets of related work.
The audience for the proposal should typically be the fourth and fifth committee members. Someone who reads the area background of your dissertation proposal should be able to understand at a high level the entire research area, in preparation for understanding the problem background and research agenda. The proposal should have a teaching style, using figures and wording that explain the area to a non-expert. The proposal should also provide a synthesis of the research; it should not just restate the abstract of each paper you cover.
In addition to discussing the broad research area, the dissertation proposal should answer the following questions as they relate to your proposed dissertation problem:
What problem do you want to solve?
Who cares about this problem and why?
What have others done to solve this problem and why is this inadequate?
What is your proposed solution to this problem?
How can you demonstrate that this is a good solution?
These questions serve as a useful model for real research papers and grant proposals. Ideally, portions of your proposal can be incorporated into future papers and proposals. Although reviewers for conference papers and proposal panels often comprise experts from your research area (the blue area in the graphic), members of your dissertation committee will be outside your research area. One of your goals in writing your proposal is to bridge that chasm as needed in order to make your proposal accessible to your committee.
The following is an example outline of a dissertation proposal. Students should work with their advisor to determine the specific section structure and length of their proposal. For example, the broader research area discussion could be part of the Introduction, placed in the Related Work section, or even included as an Appendix.
Abstract: 1 to 2 paragraphs summarizing the proposal.
Introduction: 1-2 pages answering questions 1 and 2 above.
Related Work: 1-2 pages answering question 3 above.
Thesis statement: 1-2 sentences stating what is to be demonstrated in your dissertation.
Project Description: 6-8 pages answering question 4 above.
Validation: 1-2 pages answering question 5 above.
Dissertation Schedule: about 1/2 page specifying dates for completion of major milestones, including potential papers and their submission dates.
Appendix: Research Area Overview: about 3 pages. (Omit if the research area overview in included elsewhere, and add about 3 pages to the length of the section in which the overview is included---e.g., in the Introduction or Related Work.)
Bibliography: references for all work cited.
The following discussion provides additional detail on the content of the dissertation proposal.
Abstract – This section provides a short summary of the proposal. As part of the dissertation scheduling process, the title and abstract need to be emailed to the Graduate Academic Advisor.
Introduction - This section provides the reader with enough information to understand and appreciate the thesis statement. This includes giving the motivation for the research, defining terms and formulating the problem. Often, subsections labeled “Background” and “Motivation” will be included in this section. This section typically provides answers to the questions “What problem do you want to solve?” and “Who cares about this problem and why?” The Background subsection could contain the area overview discussing the broader research area.
Related Work – This section contains a survey of the literature directly related to the problem you are trying to solve (i.e., thesis statement) and should demonstrate to your readers that you understand the context of your work. This is a place for you to position your contribution to your specific research area relative to other work that has been done, and to state how your work builds on the previous work. In most cases there will be only a small overlap between the papers discussed in this section and the papers related to the broader research area. This section answers the question “What have others done to solve the problem and why is this inadequate?” The area overview could be included in the Related Work section, but, in this case, add subsections (e.g., “Research-Related Work” and “Research Area Overview”) to distinguish the type of related work.
Thesis statement – A clear and concise statement of what is to be demonstrated or developed in your dissertation work. A good thesis statement makes a specific claim that your readers care about. Ideally, your introduction will give your readers the background they need to understand your thesis statement and to conclude that it matters.
The following are examples of good thesis statements from proposals in the BYU Computer Science department:
Project Description – This section describes your preliminary ideas on how you might solve the problem and the anticipated contribution your research will make to the field of Computer Science. This section should convince your committee that you are qualified to pursue the research and have high potential to eventually be able to objectively and convincingly defend your thesis statement. This section answers the question “What is your proposed solution to this problem?”
Validation – This section describes the methods you will use to validate your proposed solution. This section answers the question “How can you demonstrate that this is a good solution?”
Dissertation Schedule – This section contains a proposed schedule for the completion of your dissertation work. The schedule should include deadlines for submission of the dissertation to your advisor, submission of the dissertation to your committee members and the dissertation defense. In addition, this section contains the titles of three or more papers that will be published from this work, along with potential submission dates and venues. You may also include other appropriate research milestones.
The Department requires that you allow at least three weeks between the time you schedule the defense and the time you actually defend the dissertation. In order to schedule your final dissertation defense, your first three committee members must have read and approved the dissertation. In order to allow sufficient time, you should plan on approximately eight to ten weeks between the time you first give your completed dissertation to your advisor and the time you defend. This time is an approximate time; work with your advisor and committee to determine the actual amount of time that will be required.
Bibliography - This section contains references for cited work. References should be complete and written in a uniform style, consistent with your particular sub-area of computer science. Current journals in the student’s area can be consulted to determine appropriate reference styles.
The presentation should address both the broad research area and the narrow research problem. Since both parts have equal importance, a suggested agenda is as follows. The student first gives a 25-minute presentation on the broad research area, with 20 minutes for questions. The student then gives a 25-minute presentation on the narrow research problem, with 20 minutes for questions.
The student and advisor are responsible to establish a clear agenda with the committee at the beginning of the presentation in terms of the anticipated length of the presentation and any preferences for when questions are handled. The student and advisor are responsible to manage the time effectively so that both parts are sufficiently addressed.
The primary purpose of the dissertation proposal is to determine whether a student has established an appropriate research agenda for PhD research. Secondary purposes include showing that the student can think/learn/work independently as a researcher and can present persuasively.
The presentation of the proposed dissertation research should augment the written proposal. The presentation should be polished and practiced. The purpose is to convince the committee that the research topic is important, that the research methods are sound, and that the proposed research can be successfully completed. In short, the proposal should convince the committee that the dissertation, when complete, will make a substantial contribution to the field of Computer Science. The presentation is public, but the committee meets privately before finalizing the results.
The result is pass/fail and will be given to the student immediately following the committee consultation. As discussed, judgment will be based on the perceived suitability of the dissertation and preparation of the student. If passed, the student should proceed with the dissertation. University policy states that if two committee members vote to recess or fail, then the examination is recessed or failed. If failed, the student will be asked to overcome the perceived deficiencies and reschedule the proposal. Failure, repeated failures, and delayed rescheduling are all subject to discussion and action during subsequent PhD progress reviews.
Since two committee members are sufficient to fail a student, the student should be sure to teach the material in a style that is approachable to all committee members. It may be helpful to receive feedback on the proposal from the fourth and fifth committee members before scheduling the presentation.
A passed dissertation proposal is not a contract with the student that if the stated work is accomplished as proposed, final passing is guaranteed. Because of the longer term and dynamic nature of dissertation research, final passing of the dissertation rests with the committee at the dissertation defense.
At Least 1 Week Prior to the Proposal:
Dissertation format: University guidelines specify the format of the title page, abstract page, and acknowledgments page. With consultation of the dissertation chair, the rest of the thesis should be written in a style and form consistent with the Computer Science literature for your research area.
BYU requires a department to certify the consistency of format, captions, and references for all dissertations. Format and captions pose few, if any, difficulties, but citations are often inconsistent. (Typically, students copy citations from various places without reworking them into a consistent format.) To ensure consistency--and to avoid having to rework citations at department sign-off time-- follow the instructions in either (1) or (2) below.
Use the same style for the same kind of publications throughout--italics, bold, quotes, punctuation, names (full names or initials), numbering, and indentation.
Include all components of a citation--all author or editor names and the title and year; and, as applicable, other components such as journal name, volume, number, pages, month, publisher, city, state or country, and institution for MS theses, PhD dissertations, or technical reports.
Order the bibliography alphabetically by last name of first author.
The actual organization of a traditional dissertation is flexible. The organization is similar to that of a Master's thesis, but the amount of effort is higher throughout. A typical organization would be as follows.
In addition to a traditional dissertation, two other alternatives are also acceptable. One is a paper collection, and the other is a book.
Audience: Researchers well acquainted with the topic.
The dissertation should be written as a collection of papers, each supporting some part of the dissertation work. Each chapter, except the first and last, should be a paper. These papers may be in manuscript form and need not be constrained by page limits. Ideally, some of the papers would have already been published, and others would have been submitted for publication. Chapter 1 should be an introduction to the collection of papers and should provide background to motivate the work. The final chapter should summarize the work, tie the conclusions of the included papers together into a coherent whole and discuss future research issues. Additional work such as user manuals and code documentation may be included in appendices.
Comments: Since each central chapter is a paper, we must allow for redundancies across introductions and other explanatory sections. Also, a student may wish to include a longer manuscript version of a paper published under tight page constraints either in place of the actual published paper or as an additional paper.
The dissertation should be written as a book. It would normally be in manuscript form, but should have all the characteristics of a book.
Audience: CS faculty and CS graduate students. Do not assume that either the faculty or the graduate students are knowledgeable in your specialty research area.
General Comments: The presentation should be similar to one that would be made in a graduate colloquium. Again, the presentation will have ideally been practiced in front of an audience, and feedback for improving the presentation will have been taken into consideration.
The defense is open to the public and consists of a 30 - 40 minute presentation of the candidate's research followed by questions from the audience. At the end of this public part of the defense, everyone except the committee is excused. After the committee makes a preliminary assessment and decides on further questions, if any, the candidate returns to answer the questions of the committee. Afterwards, the candidate is again excused, and the committee votes.
The committee may vote to pass, pass with qualifications, recess, or fail. If two or more examiners vote to recess, the examination is recessed. A second and final examination is rescheduled, but not sooner than one month after the recessed examination. A second examination cannot be recessed. If two or more examiners vote to fail for either the first or second examination, the candidate fails and the graduate degree program of the student is terminated. If fewer than two vote to recess or fail, a majority vote decides between pass and pass with qualifications.
The dissertation defense should be modeled after a graduate colloquium because this can help prepare students that need to give presentation on their dissertation research during the job interviewing process.
In general, no dissertation should be scheduled for defense that is not of sufficient quality to pass the examination. This places the main burden of PhD quality assurance on the graduate advisor and the second and third members of the student's PhD committee.
Approximately 1 Month Prior to the Defense:
No Later Than 3 Weeks Before the Defense (earlier is better):
No Later Than 2 Weeks Prior to the Defense:
Defense day or after:
Applications must be submitted to the Graduate Academic Advisor by the university deadlines for graduate students. Please contact the Graduate Academic Advisor with questions about these deadlines. You can apply online at the Graduate Studies website.
Students cannot apply for PhD graduation unless they have (a) completed all their coursework, (b) successfully proposed their dissertation, (c) satisfied their teaching requirement, and (d) satisfied their residency requirement, and (e) have a GPA of 3.0 or higher, and (f) have a current ecclesiastical endorsement.
Students may not defer graduation. If they did not successfully meet all the graduation requirements for the semester in which they applied to graduate, the student needs to contact the Graduate Academic Advisor to have their name withdrawn from the graduation queue, and they will need to reapply for graduation at a later date. There is no fee for graduate students to apply for graduation.
The University requires all students to register for at least 2 credit hours during the semester in which they complete the submission of their electronic thesis or dissertation (ETD). If students miss the graduation deadlines for any given semester they must register for at least 2 hours or pay the equivalent minimum registration fee and will graduate the following semester.
Computer Science Department
Brigham Young University
3361 TMCB PO Box 26576
Provo, Utah 84602