【申请经验】给希望攻读计算机科学研究生的外国学生的建议（翻译）

原文链接：https://users.soe.ucsc.edu/~ejw/advice/。虽然是UCSC并且年代久远，但大部分信息应该还是普遍适用的。其中加粗的部分是重点。

 

Advice for Foreign Students Wishing to Pursue Graduate Study in Computer Science at UCSC

Jim Whitehead
Last updated: April 18, 2012

I frequently receive email from non-US citizens who wish to pursue graduate study in Computer Science at the University of California, Santa Cruz. Instead of replying individually to these messages, I have created this page to collect my personal advice on this matter. These are my own personal views, and certainly should not be viewed as official policy of UC Santa Cruz. That said, I have tried to be as accurate as possible.

My goal in writing this is to help you, the foreign graduate student applicant, better understand the process of applying to graduate school at UCSC, and to be able to provide the best possible representation of your capabilities. The Department of Computer Science is actively interested in enrolling foreign graduate students, since they bring unique skills, and diverse viewpoints to our program. We derive considerable strength from the many active foreign graduate students currently studying and performing research at UCSC.

Relevant Links

Some Facts of Life

It's helpful to have some background on the financial implications of a foreign student being admitted to UCSC.

Since UC Santa Cruz is supported by the State of California, students who are California residents have significantly lower tuition and fees than non-residents. Students who are US citizens can become California residents after a year. Foreign students who do not hold a US green card cannot become California residents. It is very difficult to get a US green card.

The net result is that foreign graduate students cost about 1.5 to 2 times as much as California resident students. From a Professor's perspective, this means that nearly two California resident students can be supported for the same cost as a foreign student. Supporting a foreign graduate student for an entire 4-6 year PhD program requires raising approximately USD $120k-$220k in grant money, depending on many factors, including whether the student receives summer support (increase), how quickly they advance to candidacy (decrease), and whether they already have a Master's degree (increase), and how quickly they finish. Admitting a graduate student is a significant financial commitment.

English Expression and Students who are Non-Native English Speakers

One common issue with foreign students is English is not their native language. As a consequence, they frequently are not as skilled at expressing complex technical ideas in Engligh, in both written and verbal form. This is a very complex activity even for native English speakers, and is doubly so for people whose native language is not English. Research in Computer Science involves investigation into complex technical issues, and then writing up these results in the form of workshop, conference, and journal papers. Writing a dissertation (typically 5,000 - 60,000 words) is a requirement for receiving the PhD degree. Writing articles and giving presentations at technical conferences are critical elements in developing a reputation in a research community. Written and verbal communication are central activities in Computer Science research. If you don't like to write and give presentations in front of groups, you should think seriously about whether you want to go into research.

From a research advisor's perspective, having non-native speakers as students frequently means that any written document by such students requires an additional 1-2 reviews just to get the English expression correct. Even after putting this additional effort into the document, it is generally of lower quality than a document produced by a native speaker. This is due to issues such as less interesting choice of words, less varied sentence structure, and poor control over the rhetorical structure of the paper. Finally, most professors do not enjoy acting as a copyeditor; they would prefer to focus on core research ideas, rather than English expression.

Who Makes Admission Decisions for Graduate Students?

In the Department of Computer Science, all admissions materials are digitally scanned, and made accessible via a Web-based system that allows the faculty in Computer Science (as well as Computer Engineering, Bioinformatics, and Electrical Engineering) to see and review all files. Each file is reviewed by multiple faculty members. Generally students are admitted when they are either visibly outstanding and exceptional students, or when an individual faculty member makes a personal commitment to accepting the student, and providing financial support. Administrative staff typically have a advisory role in admissions decisions, usually by informing faculty that there is a student who might be a good fit for their program.

The most important point is individual faculty make graduate admissions decisions.

Some Advice

Since foreign graduate students cost twice as much, and frequently produce research papers of lower quality than native students, it is critical that applications for graduate study address these two issues. From a research advisor's perspective, a foreign student needs to be outstanding and exceptional to justify the extra financial and time cost (all those extra paper reviews take time) of admitting a foreign student.

There are several things you can do to improve your chances of gaining admission. Contacting a faculty member is critical, since they make decisions on who is admitted. However, when contacting a faculty member, there are several key points to remember:

1. English expression must be perfect. Consider the following excerpt from an email I received from a prospective student:

I did my undergraduate studies at {XXX} University, one of China's the most famous universities with a history 100 years, and kept on ranking in the upper 10 percent among 92 students in the department.

Just in this one sentence there are at least four grammar and expression errors. This clearly says to me that the student does not have a firm command of English expression, and their papers will definitely need multiple reviews. As a Professor I want to bring in students who will make less work for me, not more.

 

2. Show a clear and sincere interest in the faculty's research area. Faculty have very specialized research interests, and are looking for students with demonstrated capacity to perform excellent research in their specialty. If you approach a faculty member and say that you want to perform research in an area outside their specialty, they will show little interest. Additionally, faculty often indicate research interests that are very broad. I personally state that I have a research interest in "Software Engineering." However, this is a very broad field. My more specific interest is in software evolution and software bug prediction. Hence, if you send me email stating you're interested in Software Engineering, that is much less interesting to me than an email stating you're interested in software bug prediction or software evolution.

Of course, just listing a few buzzwords doesn't impress much either. Some emails I receive appear to be form letters, with a fill-in space for research interest buzzwords. These are pretty easy to spot, and don't demonstrate any sincere interest in performing research in that field. If you want to rise above the crowd, you will identify potentially interesting faculty, then go and read one or two of their recent papers. This will give you a very good idea of their research specialty (you may find you don't like it!) and will allow you to write a customized letter that speaks directly to that faculty's current research interests.

Here's another example email I received:

Checking into the web site, I have found that your research areas cater to my type of research interests.

I interpreted this as follows. First, the student doesn't know or care what research they want to perform, since they didn't state their research interests. This is a red flag, since such a student might be accepted into the program, chew up a year of grant support, and then discover they are really interested in some other research area and switch to another research group. While students often do switch research areas as their interests mature, as a Professor I look to reduce the possibility of this happening, since this is disruptive to a research agenda.

Second, the prospective student obviously didn't do their homework. They didn't bother to read my web site enough to determine my current research interests.

Right now you may be thinking, "Reading all these papers takes too much time, especially for a preliminary email." This is true, it does take more time to really understand the research agenda of a faculty member. However, by applying for graduate school, you are asking a professor to make a large financial and time investment. My attitude as a faculty member is, "Why should I admit a prospective student who is not willing to spend 1-2 hours to determine my current research interests?" On the flip side, you are looking to make a commitment to a particular research area, and this can have significant impact on your career and happiness. Doesn't it make sense to have a good idea of what you're getting into?

3. Use an ASCII-representable name in your email if your native language is ideographic. A lot of spam email has "from" addresses that are not ASCII. Many faculty receive over 100 spam email messages a day. To ensure your message isn't ignored as spam, use an informative Subject line, and an ASCII From address.

4. Check to see how many students the professor currently has. Research groups are hard to scale up. Generally, faculty members do not like to have more than 3-10 students. If a faculty member has a lot of students already, even if you're Albert Einstein they still might not be interested.

Some Advice for the Actual Application

If you have decided to formally apply for graduate student in Computer Science at UCSC, you will be asked to provide a personal statement (an essay), letters of recommendation, GRE and TOEFL test scores, and your undergraduate transcript.

Personal Statement

By now it should go without saying that the English in the personal statement needs to be perfect. Have a native speaker review your essay. Pay money to get such a review, if necessary. I frequently see high GRE verbal and TOEFL scores accompanied by a poorly written personal statement. This says to me that the student takes tests well, but still has problems with written English expression.

In the personal statement, I personally look for evidence that the student has the capacity to perform high quality research. Specifically, I look for evidence of critical thinking, and discussion of prior relevant research or project experience. I also like to see students make a strong case for why they are well suited for graduate study, especially at UCSC.

Many prospective students write about how they have wanted to study Computer Science since they were a small child, and how it's a lifelong dream and ambition. In extreme moderation this is fine (1-2 sentences at maximum). However, it is reasonable to assume that nearly all prospective students willing to spend the time and effort to apply to the Dept. of CS for graduate school have a deep and sincere interest in Computer Science. One deduction you can make is that spending valuable space in your personal statement on this matter does not differentiate you from the other applicants.

I personally would like to hear about the difficult aspects of some project you worked on. Tell me how you displayed excellent problem solving skills, were a leader, or showed initiative. Especially if you have done research before, tell me about that research experience, and what you learned from it. Finally, tell me why UCSC is the best place for you to study. There are many universities in the US; why is UCSC the best match for you?

Don't Send Every Award You've Ever Received

A few years ago, one student sent a copy of their karate black belt certificate as part of the supplemental materials for their application. While this is a commendable achievement, one that implies a significant degree of discipline, it doesn't say very much about the student's ability to perform graduate studies. In fact, it's a negative, since it indicates the student doesn't have the critical analysis skills to determine what information can truly help their application. We threw the karate certificate in the trash. If you send supplemental information, make sure it is directly relevant to the pursuit of graduate study in Computer Science.

Make Sure Your Letters Arrive on Time

The recommendation letters are critically important to your case. Make sure they arrive on time. Letters from academic sources are preferable to those from industry, and a letter from an advisor on a research project is the best, so long as it directly addresses your performance on the research project, and your skills as a researcher.

Make Your Point Fast

In the preliminary review of files, the typical application receives 2-8 minutes of consideration by each faculty member that reviews it. Follow-on reviews are typically more detailed, since there are fewer files at this point. Personal statements are first scanned, then carefully read if they seem interesting. Your personal statement should ensure that the most important points are made in the first 1-2 paragraphs.

You may be thinking, "This isn't very fair. I'm going to spend hours putting together my application, only to have it reviewed very quickly." In defense of the practice, I'll make two observations. First, we receive many hundreds of applications, and must make rapid decisions on them. In order to provide timely response to applicants, each application cannot be reviewed for very long. Second, after reviewing hundreds of applications, faculty get very good at making rapid assessments of academic records. Does this mean we never make mistakes? No. It's an inherently error-prone process, since it involves making highly subjective decisions based on a small number of data points about a person's future academic abilities. Predicting the future is hard.

Hopefully you've found this advice useful. My sincere best wishes to you in your pursuit of graduate study in Computer Science!

posted @ 2021-04-07 22:49  Ryan0v0  阅读(276)  评论(0)  编辑  收藏  举报