EE Quals at Stanford

June 21, 2007 at 1:06 pm Leave a comment

From the Stanford Daily

High-stakes EE exams elicit stress, pressure
Fewer than 60 percent pass annual exams

February 1, 2007

By Christopher P. Anderson

This is the first of a two-part series about the high-stakes doctoral qualifying exam in the electrical engineering department. Part Two will run in Friday’s Daily.

It happens every year: Stressed-out students with nervous looks on their faces roam the halls of Packard and the engineering quad. The slow start to the New Year exhibited by electrical engineering graduate students is not just a holiday hangover. For many EE students, regular life shuts down until the doctoral qualifying exams (quals) — held in the second or third week of winter quarter — are over.

The most imposing of several hurdles on the way to a Ph.D. degree, the Stanford EE department’s controversial qualifying exams are known as one of the most intimidating selection processes in the country. Two factors contribute to the stress and pressure of the week. The first is the sheer numbers: Of the over 150 students who take the quals each year, no more than 60 percent will pass. The second is that most students who fail their two attempts will have to leave the program without getting doctoral degrees.

The EE qualifying exams are held in the span of one week in January. Every candidate has a 10-minute session with each of 10 faculty members. Each professor questions the student on a topic in the field, then scores the student between one and 10. The scores are added up and all students are ranked in order. The faculty meet to decide where the passing line will be, and all students above that score pass.

If and when they pass, students can file as Ph.D. candidates. The benefits don’t stop there — students without research advisors anecdotally report that professors are more willing to discuss research with them after they have passed.

The student-selected “examining committee” must comprise professors from at least three of eight subject areas, ranging from computer hardware and software to low-level device physics.

Examiners vary widely in their styles and manners. Some are known to ask the same type of question each year. Many ask basic knowledge questions, but some ask brain-teasers or open-ended design questions. Some professors interact with the student, ask for clarification and give hints. Others take a less hands-on approach and read their email silently during the exam. One professor brought a student to the parking lot and conducted the examination while he re-parked his car.

Though a few daring students choose to take the quals “cold turkey,” most begin studying in the fall. As the date nears, students have been known to de-emphasize coursework and research before catching up immediately after the exams are over.

The varying “styles of the examiners” force students to prepare for everything in their chosen subject areas. Second-year doctoral candidate Tom O’Sullivan formed a study group, meeting with three friends from the department on a regular basis in the fall and in January. The group practiced the 10-minute format with past years’ questions which are available online.

Third-year Ph.D. candidate Kim Shultz tried both preparation techniques. According to Shultz, she put in “about four hours” of studying for the 2005 exam; she “subsequently” finished in the bottom quintile. In 2006, with a more focused attitude, she finished in the top 10 — a score high enough to merit a one-year graduate scholarship while she decided on a research field. Shultz even saw a silver lining in the normally high-pressure examinations.

“Quals would be fun if [they] weren’t so stressful,” she said.

Since they are designed to facilitate academic research, the quals come under fire for not involving research skills in the examination process. Shultz said, “Quals have nothing to do with anything,” implying that the 10-minute format makes it difficult to show broad academic knowledge and that the style of questioning does not reflect the ability of a student to perform extensive research.

All of this adds up to a day of high-stakes interviews for students, with academic futures on the line. Still, some are not as worried. Second-year doctoral candidate Drew Hall told The Daily, “I didn’t feel very stressed…people hype it up too much.”

*EE exams press students
Students, faculty criticize shortcomings of must-pass test

February 2, 2007

By Christopher P. Anderson

This is the second of a two-part series on the high-stakes doctoral qualifying exams (quals) in the electrical engineering department. Yesterday’s article (“High-stakes EE exams elicit stress, pressure”) discussed the examination process and students’ studying techniques; today’s piece covers the controversy surrounding the tests and their aftermath.

Almost all doctoral programs have qualifying exams, which are as diverse as the fields the tests examine. UC-Berkeley’s EE quals involve a one-hour examination between a student and three professors in the student’s area. MIT’s Ph.D. candidates take final exams in undergraduate courses and present their research to a committee.

Stanford’s chemical engineering department utilizes “pre-quals,” with students presenting another scholar’s published research before they present their own findings in the actual qualifying exam. The mechanical engineering department requires that students gain research experience before taking the quals.

The crushing attrition of the EE quals — no more than 60 percent of students will pass in any given year — can be attributed in part to the numbers: There are simply not enough positions in research groups (nor is there enough funding for research) for the 100-plus students each year who seek them, and not all students possess the proficiency to do doctoral-quality research. The department enrolls approximately 200 graduate students each year, and typically passes fewer than 90 through each year’s qualifying exam.

Second-year EE graduate student Drew Hall said that the forced competitiveness of the qualifying exam makes the process imprecise.

“There’s so much pressure to make the questions harder each year, so professors end up with questions that most people can’t even get an answer for,” he said. “And it’s harder to score those kids, which introduces noise into the process.”

Hall cited a bottleneck in the middle of the scoring — a difference of a single point above or below the passing mark can involve a dozen or more students.

“It is only good for picking out the very top and the very bottom,” he added.

Several students said their advisors had privately expressed frustration with the quals, especially regarding the hassle of the process and the need to make questions harder each year. EE Department Chair Bruce Wooley noted that students tend to get too worked up about the exams.

“The important thing is to get an advisor,” he said, citing examples of students who passed but could not get a faculty member to sponsor their work.

Hall and third-year EE doctoral candidate Kim Shultz both feel that more students should be “weeded out” of the Ph.D. track in the admissions office. Hall says that the quals receive unfair criticism in light of the sheer numbers.

“It does what it needs to do, yet everybody blames it,” he said.

However, Wooley says changes in admissions are not feasible and that stricter admissions standards would impede the quality of research.

“Especially with all the international students, we just can’t tell by the transcript who is Ph.D. material,” he said. “Sometimes it’s not the high scorers [in classes] who are successful in research.”

Wooley added that quals were a necessary obstacle that forces students to consider if they really want a Ph.D.

“If we could make [admissions] better, we would,” he said.

A recent policy change may have helped reduce the field and increase the pass rate. Several years ago, the department added a Master’s-only option to their admissions process, accepting a batch of students who are barred from taking the quals. The department had previously admitted all graduate students into the Ph.D. track.

The number of students taking the quals has declined in the last three years, possibly aided by this policy. A departmental administrator declined to release admissions statistics to The Daily.

The Master’s-only option caters to students who do not aspire to earn a doctoral degree. Second-year graduate student Christina Hernandez is anxious to graduate and return to the working world.

“I have no life right now,” she said jokingly. “Six more months and I have a life. I’m looking forward to having nothing to do at 6 p.m.”

There is a final appeals process for students who fail to pass the quals after two years. A faculty committee reviews appeal requests and considers both the student’s academic performance and research portfolio. Students who show potential stand a decent chance at having their results overturned — another reason Wooley and other faculty members emphasize finding a research advisor early in the graduate career.

Upon passing the quals, students can officially declare as Ph.D. candidates, provided they obtain two faculty members to serve as dissertation advisors. Students must then complete a doctoral research project, present and defend their results to a committee and write a thesis, which must be approved by the advisors.

In electrical engineering, the entire process can take anywhere from three to seven years, depending on the project’s complexity and the student’s initiative.

EE students typically hit the night life scene twice — once for finishing the exams and again when they get their results.

Shultz celebrated her high score by attending her favorite bar in Mountain View and sleeping in. Second-year doctoral candidate Tom O’Sullivan went out for dinner with friends, then headed to the Rose and Crown for drinks. O’Sullivan is thrilled the quals are behind him.

“There’s nothing like your first time,” he said.

Original links.


Entry filed under: Grad school.

On Social Engineering in the local Universities Sexual Frustration (Poem)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Trackback this post  |  Subscribe to the comments via RSS Feed

takchek's Photos


Blog Stats

  • 70,693 hits
June 2007
« May   Jul »
_uacct = "UA-1905437-2"; urchinTracker();

%d bloggers like this: