CRA-DMP Evaluation Report #1

The Mentors' Point of View: Issues Involved in Mentoring in the DMP

IV. The Role of the Mentor in the DMP

In this section we will discuss the roles that the mentors indicated they performed while participating in the DMP. In our interviews, we asked the mentors about their own definitions of "mentoring." Many responded by describing "mentoring" as a process of assessing the individual needs of the mentee and then attempting to provide for those needs. As a result, we came to understand that the role of the mentor was multifaceted and dependent upon the individual student. The following quote demonstrates the multifaceted nature of the mentoring role.

I: Can you define what you think mentoring is?

R: Gee, that's a hard question. I guess it's a combination to me of being sort of like a mom and a friend and trying to share with the students the sort of good and bad experiences I've had and to guide them to try not to repeat some of the mistakes I've made. An equal part of it is really, I think, opening up to them areas that they haven't studied in computer science, how to work on their own, how to work with groups, really trying to give them a sense of confidence in many cases, and also to let them know that there are a lot of people out there that can help them when they decide they need some help, whether it be for a job or getting more information about a subject.

As discussed in the student section, the undergraduates entered into the DMP with little understanding of graduate school, the nature of research or how to conduct it, the academic community in CS&E and the life of a female faculty in CS&E. Our interviews with the mentors indicated that the main roles they discussed paralleled the students' needs: being a role model, providing professional guidance, and guiding students in their research project.

IV.A. Being a role model: Providing an example of a successful woman in CS&E

Most mentors expressed that part of their role in the DMP was to provide a model of a "real life" example of a CS&E female faculty member. Many commented that this modelling took two different forms: students could observe the mentor in her daily activities, and the mentor could explicitly discuss and provide guidance about the types of experiences she encountered as a female in CS&E. Many mentors mentioned that, although a discussion of their experiences was beneficial to the student, the largest impact on the student occurred by observing the mentor on a day-to-day basis.

I: Did you talk to [your DMP student] a lot about being a female in computer science, and all of that?

R: Not really. Not so much about females. But she did observe, because I also talked to her and my family also interacted with her. Not on a daily basis, but we went out to dinner and so forth. But she also met my family, so she was opened up to see that there are other aspects of a female professional -- there's the family life too. And there's constant juggling between the two. So she can observe this through me, but we didn't really sit down and talk at great depth.

IV.B. Guiding students in their professional development

Virtually all mentors stated that an important aspect of their role was to guide their student in her professional development. This included providing guidance for the student in her career decisions and also introducing the student to the culture of an academic environment in CS&E.

IV.B.1. Guiding the student in her career choices

Most mentors commented that guiding their student in her career choices was a central part of their role. For many mentors, guidance involved introducing the student to new opportunities and encouraging her to take advantage of them. Many mentors commented that students were not always aware of their options or, if they were aware of them, did not have the confidence to take advantage of them.

I think once you have a student who has a goal that there is a great deal that you can do in terms of just getting them information and making them feel okay about themselves. Some of the students will cut off their options simply because they don't think that it's them. And mentoring can help get people over that kind of response to the options that they are presented with and can just present them with options. And I guess I see that's the primary role of mentoring at this point. How many options you have to present and whether we have to dig for them or just point them in the right direction. That depends on the student.

Well, I think mentoring starts with encouragement, sort of recognizing that the student, male or female, has good qualities that you can enhance - and unfortunately not all student have that - and then start by encouraging them and finding out about their likes, their interests, just kind of their personality. ... So I think mentoring begins by like a one-on-one attention, and then from there, not all students are going to want to do graduate school. Some want to do jobs. ... I think it's just kind of giving them attention and giving them help where they need it. That's my definition.

I think the biggest problem is that these women, either they think that they're not good enough to apply, or they think the opportunity of going somewhere over the summer is not going to help them, or they're too shy to do something like that.

The following mentor commented that one of her DMP students was struggling with deciding between graduate school and teaching high school. The mentor then gave the student examples from both careers so that she could make an informed decision.

I: What has been your experience in mentoring undergraduates?

R: Okay. I would say quite variable. It depends on the individual person. People are interested in different things, they need different things. For example, for the Distributive Mentoring Program, [my student] was at a period where she was very uncertain as to what she wanted to do when she graduated from college, whether she wanted to go to graduate school or whether she wanted to go into the teaching profession, all high school. What she was interested in was seeing what this research thing is all about and just having an opportunity to think about which way she wanted to go. So I introduced here to a former colleague of mine who had taught in high school as well as taught in college and had done a small amount of research. 'Cause she has a background different from mine it would be useful for [this student] to talk to her as well as to me. (pause) She also wanted a fair amount of direction. She wanted to work on one of my projects, so we sort of pinned down a topic and she then was able to sort of carry the ball from there. A lot of what she did was based on her own initiative, her own ideas, and she did an excellent job on that. I think she had the opportunity to see, "This is what someone doing research does," and had an opportunity to do it herself.

Thus, most mentors considered it part of their role to identify the student's interests and provide career guidance based upon those interests. Many mentors provided guidance by discussing with students their options and providing encouragement, as well as providing examples of successful people in their career.

IV.B.2. Introducing students to the culture of an academic career in CS&E

Almost all mentors considered an important component of their role to be introducing students to the various aspects of an academic career in CS&E. This included teaching the students about the entire research process as well as introducing the student to the professional culture of the academic field of CS&E.

IV.B.2.a. Introducing students to the research process

Many mentors realized that their students had little or no conception of what research was about and, as part of their role, discussed with their student the nature of the research process. In the following two interview excerpts, the mentors discussed their role in teaching their students about the research process.

I look at [mentoring] as trying to teach somebody more of a step-by-step process. You're not really trying to teach them the technical content of anything. You're really trying to teach them a process of doing science. I look at my advisor and I say, "Okay. Somehow or another she taught me how to write, somehow or another she taught me how to make presentations. And what does it take to be successful in this kind of a career." And I think a lot of it is just by example, they watch what you're doing.

I: How would you define mentoring?

R: Well, I guess it's sort of a guiding in not just the technical area but the culture surrounding it. So you want to show them how to do research, the technical details involved. You want to show them the context for that research. You want to teach them the etiquette of that field like about giving credit to other people for their work. In certain areas when you write a paper names are alphabetical, other areas it's like who's the most senior. And you sort of make them aware of these things and what it is in different areas. What conferences cover the area that you're working in. What kind of job would you get in this area. So just the whole context for research and the research area.

As these mentors explained, this immersion involved introducing students to the day-to-day process of "doing" research, providing them with the context of how their area of interest fits into the larger body of current research, introducing students to other people who can assist them, and giving them names of journals to which they can submit their research. In other words, the mentors felt it was an important part of their role to introduce the student to the entire spectrum of issues and processes involved in academic research. The following quotes addressed each of these aspects.

Just to kind of show them the excitement of research, that you find in the research. They can see, when they do something how you get all excited about it and they can see what kind of questions you ask, what you'd want to do next and I think that they can start to get a feeling if this is something that they would want to do. Just you know the excitement of research, the methodology, the tediousness of it sometimes.

[My second DMP student] doesn't know a lot about the context of the area so I spent more time like giving her things to read, like biographical things about some of the researchers we were studying. And I really felt that she sort of had no idea of the culture, no idea what a researcher really does. I mean even if she could do research, she didn't really know what the life was like or how that was different from industry. So I was more doing things like giving her other things to read, trying to describe some things to her about what I liked about research.

[My second DMP student] was sort of in a different personal situation because she had already made a firm decision that she definitely wanted to go to graduate school. A very independent person, somewhat of a loner in terms of work style, and what she needed was not really someone to work closely with, but someone to provide a supportive environment and a lot of contacts, like, "You really should talk to the lab people about designing your experiment because they have expertise that neither of us has." So I was able to provide feedback on what she was doing and provide all sorts of appropriate contacts here, suggestions for, "Here is a conference where it might be appropriate for you to submit a paper based on your work." Basically I think there I was providing contacts and information and sort of pointing her in the right direction more than working closely and bouncing ideas back and forth a whole lot.

IV.B.2.b. Introducing students to the academic environment

In addition to giving students an idea of the research process, almost all mentors we interviewed said that they spent time orienting students to the professional culture of academic CS&E. Some mentors took their students to conferences and remarked that this was a way for the students to interact on a professional and social level with other CS&E professionals.

The first [DMP student] I was able to take her to a conference and so she really got to experience a lot of the things first hand and things that come up at the conference that we could talk about. I could point out who some of the famous people were. I could tell her what kind of talks to go to. I could tell her, "See how this person gives a talk." And things like that. To tell her how to participate in discussion...You don't want to just talk about the latest movie or something.

I: They were engaging in dialogue with these people?

R: Yeah, they were engaging in dialogue, and even just socializing. Everybody was just assuming that they were a grad student.

One mentor commented that taking her DMP student to a conference had a significant influence on the student. She felt that since most of her DMP student's friends planned to work at industry jobs after graduation, this experience at a conference would provide her with insight about another type of option she may not have been considering.

But I also sensed for her that she was not sure whether she wanted to go to grad school. Or, I think she knew she did but just because of her particular situation I felt like if she went to a conference she would love it and that it would really sort of confirm that she wanted to continue along that path, the research path.

IV.C. Setting up a framework for the student's research

In this section we will discuss the role of the mentor in working with the student on her research project. This role included preparing for her student's arrival, arranging meetings with the student to discuss her project, and guiding her student in her research.

IV.C.1. Preparing for the student's arrival

Many mentors prepared for their student's arrival by setting up accounts for her on the local system, getting an office for her, and helping her to find housing. The following is one mentor's experience.

I: Did they ask you to arrange the housing or was that something you just considered as part of your role?

R: Well, it kind of felt like I had to do that the first year. I guess I felt like I had to do everything. I took them around campus and got them their ID's and everything. I mean I spent probably two days just getting everything set up for them 'cause I felt like I should do that. When I look back on it I probably could have left them and they would of done fine.

I: Why did you feel that you had to do everything?

R: I guess I felt like I was their host here and I was the only person they knew and that I should at least get them set up. ... I kind of learned a lot from last summer in terms of what to expect and what I would get and what they would get and what to you know, how to set things up and all.

A few mentors suggested that the DMP provide a list of instructions of what should be done to prepare for a student's arrival.

I guess it will be helpful for next time we get instruction list so we can check what we can do before we start it so we know. Because everybody's so busy, and we may just look over something.

IV.C.2. Arranging meetings with the student

Our interviews and survey results indicated that most of the mentors met with their student at least once a week and many mentors met with their student more than once a week. In this section we will discuss the frequency and format of the meetings.

IV.C.2.a. Some mentors met with their students on a casual, informal basis

All of the mentors we interviewed placed their student in an office in their building so the mentors could easily stop by the student's office. Many mentors commented that they did not have a set meeting time with their student, but made sure that they were in contact with their student at least a few times a week.

I: How often a week did you meet the students?

R: Sort of as needed. I did not set up formal meetings with any of the students. In other words, "Let's meet once a week or once a day." They all had sort of offices in the same building, fairly close, so they could come by my office if they wanted to see me. I would go by their office to sort of see how they were doing, and then we would sometimes set up more formal meetings, say, "Let's sit down for a couple of hours and discuss this." So a lot of the interaction was unscheduled and informal, and that was facilitated because they had offices in the building. In fact, it was sort of in graduate student offices.

R: I met them on an as needed basis always. I always made sure that they were here. So that somebody was looking out to make sure that they weren't in an accident, or whatever. I always met them, made sure to say, "Hello." Did they have any questions? Did they have any problems? That was every day that I was here. But we didn't have a scheduled time per week because usually I'm sure I would see them two or three times a week on an as-needed basis.

These mentors felt that the almost day-to-day interaction precluded the need for weekly meetings because they could provide guidance as it was needed.

IV.C.2.b. Some mentors set up weekly formal meetings

Either in addition to casual meetings or by itself, many mentors set up weekly meetings for their student. Many of these mentors commented that the structure of weekly meetings provided a framework and guidance for the student learning and motivated the students to work on the project.

I: What if you had not met with them but once a month. What would have happened?

R: That would not work, no. They needed a deadline, I think. And then they also needed somebody to say - If they say, "Well, I've pushed this as far as I can." Then I could say, "Well, push a little bit in this direction. Or, well, let's forget about this and concentrate on this stuff." I think it was necessary for them to meet with me so I could help them see where to focus their efforts.

Some mentors commented that weekly meetings helped to ensure that their student was enjoying her project and was also working hard on it.

R: [It is a key element of success of the DMP that] on the part of the students them to be very, very strongly motivated. They were very strongly motivated, and then on my part, I feel like if I agreed to do this, that I want it to be productive for both of us. I guess I also feel like I want to make sure they're not taking advantage of the program.

I: How is that?

R: That they're just not going to take $5000 and sit back and relax for the summer. Not for any good reason, but just it would irk me to be a party to that. I guess on my part, having very clear goals, and making sure that I keep them working hard but on something they're interested in. I guess for me the structure was very important, to sort of check in weekly, to make sure that we're both working really hard, that we're getting somewhere, and we know where we're going.

Evaluator Point of View This section indicates that both students and mentors can have a positive experience with casual or formal meetings. Regardless of the configuration, it is important to note that the mentor needs to clearly define at the beginning of the program the frequency and nature of their interactions.

IV.C.3. Guiding the student in her research

All the mentors guided the student in her research project by providing direction throughout the ten weeks. In this section we will discuss the ways in which the mentors guided their student in her research project.

IV.C.3.a. Providing guidance and direction

Almost all mentors stated that undergraduates did not have enough experience with research to be able to independently choose a research topic and direct their own research. Thus, all mentors provided the direction for their student's research in the form of defining a problem for them to solve, or giving them ideas for a future topic to explore. The following interview excerpt typifies the attitude the mentors had toward undergraduates and research.

I came up with the direction and the project. Well, I think that's asking too much for them to come up with their own direction. I get them journal papers to read, and sort of limit the amount of information they have to learn in order to do a project, otherwise they could just spend their whole time learning. So I'm afraid if I tell them to come up with a research direction, I don't know. That would almost be too challenging, too overwhelming.

While defining the structure of the project and providing overall guidance, most mentors expressed that they encouraged their students "work out the details" of the project. These mentors expressed that it was important for the students to have freedom to explore solutions within the framework of the project in order to get some experience with research.

I think that giving people as much lee-way as possible you know to make decisions about how to accomplish the goals. As long as the goals get accomplished. I think it's a lot more fun if you're able to make more of the decisions yourself. I think it's, provided that they're up to it, probably a lot better than if I try to micromanage how it ends up being accomplished.

Then I said, "These look like things that keep recurring. Which one of these problems do you want to work on?" And these were problems my system can't handle and so she said, "I want to handle this thing." And then I said, "Okay, now how would you handle it?" And then she had to go out and do some research. So she sort of developed it on her own.

Many mentors struggled to find the balance between guiding the student and letting the student struggle on her own. Many emphasized that they wanted the student to be challenged by the project, but not get so frustrated that she would give up.

I would ask her questions [to guide her]. But I also didn't want her to get too frustrated, so I would let her go for a couple of days and then I would start to ask more pointed questions and help guide her. I think if anything I erred in the side of not letting her do enough research because I didn't want her to get frustrated.

IV.C.3.b. The role of the graduate student in assisting the students in their research

The mentors who had graduate students working with the DMP students mentioned that their roles differed from the graduate student. While the graduate student answered the day-to-day technical details of the project, the mentor provided the global picture of where the project was headed and the context of how the project fit with other research either in the department or in the research community at large. The following two excerpts relate to this issue.

R: The graduate student and I kind of designed a project together and then she would come to him for real specifics on- he was the one generating the output that she was using as input. So she would go to him for problems with, "Does this graph look like it's supposed to look?" 'Cause she wasn't familiar enough with what it was supposed to look like. And he would help her out. And then [with] design questions she was usually coming to me with, in terms of, we were kind of designing what things should look like as we went.

I: Oh, designing the graphs?

R: Yeah, designing what the tools should be doing and some of the, what kind of buttons should it have on it and that kind of thing and she'd come to me with those kinds of questions. She met with me about once a week or more, yeah. She probably saw me more like three or four times a week. I think she probably saw him every day.

I: Okay, so how did [the graduate student supervisor's] role differ from what you did that summer?

R: Well, she was there all the time, so she was in the lab with [the DMP student]. She had a twenty hour a week appointment. And she was with [the DMP student] the whole time. She really is the mothering type. She really loved teaching and tutoring and she taught [the DMP student] a lot and I was off doing other things. We always saw them, most likely once a day, if not more, and had several meetings a week, but we weren't there eight hours, four hours a day.

I: Right. So her discussions with [the DMP student] would be more technical?

R: Yeah, more like, I can't get this loop to work. I can't get this program to work, can you find the bug? And they'd work through the bug.

I: Right, and then your role was more over-arching.

R: Yeah, more "This is what we want to do on the global, even more the global approach, trying to work on some of the algorithms with them, but not the nitty-gritty details of the algorithms. It's sort of sketching out the proof, but not filling in the details of the proof.

IV.C.3.c. Defining the nature of interaction in the meetings

The degree of collaboration between the mentors and the students in the meetings varied among mentors. Many mentors did not consider their relationship with their student as collaborative. These mentors said that their meetings with the student were a way to discuss the student' results and provide guidance on her project, but they viewed the student's work as an independent project.

I: You met weekly, so in that [meeting] you were able to supervise where the direction would go?

R: Oh, yeah. We would review what she had accomplished and talked about what the next steps were. Occasionally she wouldn't have gotten accomplished what we thought she could and some times she would do more. It varied.

I: So your role in the research was mainly to guide her? Did you work together on certain things?

R: She had some programming problems at one point and I helped her out with those. I think I also wrote some code for her because she didn't know the language she was using to go into it. But there wasn't any side-by-side, "Let's work on this program together," the way one might do as a class project. It was mostly set up as an independent project.

Other mentors described their relationship as fairly collaborative. The following two mentors stated that in their meetings they treated their students as graduate students.

I: Now you said that there were things that she was supposed to do and then things you were supposed to do. What were those things you were doing? I mean the reason I'm asking that is it sounds like it was quite collaborative then.

R: Very much, very much. So every time she had a new idea I could usually figure out how to manipulate it or transform it to push even a little bit further. Or else I would just maybe have to go back and think about it and see what the implications were and what we should do next or to look up maybe a book somewhere or to contact somebody to see if this was known or something like that. But frequently it was, there were technical things that I had to do.

I: How did you guide them in their research? Did you give them problems, and then they would work on their own and come back, or did you work on problems together?

R: A little bit of both. I would give them projects and ask them to see what they would come up with. Then I would work through them with those projects, you know, do the board or whatever, and then send them off again. That's how I conduct all my research with all my graduate students, so I pretty much treated them like a graduate student.