CRA-DMP Evaluation Report #1

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

V. Strategies in creating a successful DMP experience

In section II, we discussed the roles that the mentor. In this section, we will discuss the strategies that the mentors employed in order to create an atmosphere conducive to their student having a positive experience in the program. It is interesting to note that in this section, each of the strategies helped to create a positive experience for both the student and the mentor.

V.A. Creating a well-defined and "doable" project that challenged the student and was part of the mentor's research program

Most of the mentors we interviewed commented that having a well-defined and doable project that challenged the student and was part of the mentor's research program was essential to having a successful DMP experience for both the student and mentor. In fact, some mentors commented having a well-defined project was so important to the success of the program, that if they didn't have such a project, they would not feel comfortable participating as a mentor in the program.

If I had a project that I felt was the right size for a summer project, absolutely [I would do the DMP again]. I didn't this year because I didn't have a project that I thought could be done in that period of time by an undergraduate. For myself, I wouldn't want to bring in a student onto an ongoing project and have them just hang around on the periphery for eight weeks or whatever it was over the summer. I don't think that they would get much of a feeling of accomplishment from that. So I would prefer to have something self-contained that was theirs before I did it again.

V.A.1. Tailoring the project around the student's background and abilities helped to ensure completion of the project

Many mentors framed the project around their students' background and abilities and attempted to create a project that was challenging enough for the student to be interested, yet not too difficult that she would get overly frustrated.

I think you have to be prepared to think about what you're going to ask the mentee to do and make sure it's within their abilities and not something that's just going to drive them crazy and have them come out with a negative experience. I think, you know, it can be a very, very positive experience if you go into it with the right preparation.

I think the only thing that I was concerned about [was] how to treat an undergraduate student who you are giving a slightly advanced problem such that they don't get intimidated by it. I mean, there is some line about making a problem so difficult that you don't get anything out of it, it was just too hard. And if you make it too easy then it seems like it's just a course project, so I wanted it to be somewhere in between. I wanted the student to be able to contribute to the research as well as get some implementation experience and some real world experience so that they could go and use it in a job or even their graduate studies. I just wanted to sort of make sure that I made the student feel comfortable trying to solve this advanced problem, definitely more advanced than what they had seen in courses, but at the same time, feeling that this was something that they could battle.

Many mentors commented that it was important to have a clear goal and outline for the project since their student did not have experience with an open-ended research process and were likely to get frustrated with too little direction.

I think defining a good project is really key and this is something that I've learned, maybe it seems obvious in retrospect. But in the past two years of supervising students, it depends on the student, but a project that's clearly defined -- where all the pieces that the students will need [are in place] in order to accomplish the project and understand what it is that they're supposed to do. The more that can be in place the much more likely it is that the project will be successful especially when students are less experienced. They do best with a very well defined project.

Many mentors stated that, given the short time frame of the program, it was important to be able to assess their student's background accurately so that the student can work on a project without a large degree of initial preparation time. Some mentors commented that their student was unable to finish her project because they had overestimated her background knowledge and needed to spend time at the beginning of the program providing information. These mentors expressed that they should have contacted their student before she arrived to determine her background in order to frame a project more suited to her needs.

R: It seemed to me that it might have been better if she had seen the projects before hand or [if] we could have talked before to figure out that this is what she wanted. I know that there is no easy way to do this because everybody has limited time, but I think that this might have helped. And also, from her resume, I knew the courses she had taken, but I did not know the skill level. And I think she got A's in everything, but it's hard to gauge. If I had known a little bit more about her before, or if I had talked to her before to get an idea of what was her, I could have asked her, "What was the longest C-program you ever wrote?" And then maybe I could have tailored the assignment such that it would have made her more comfortable. So, we did all of that in the first few weeks that she was here, but it might have helped if we had done it earlier.

I: Because you could have saved time?

R: Because we could have saved time. Because then I would have known exactly where to start off from. I assumed that she didn't have any problems with C, but it was only later on in the project that we realized that she needed to brush up on C a little bit more. And maybe it was a while back since she took the course. And I think I would have encouraged her right from the beginning, or at least sat down with her and discussed it a little more if I would have known about it.

I guess the other thing is that it might have been useful if I had more information about the students before they arrived. And I probably should have called [my DMP student] and talked to her since she wasn't at [at my university] -- You know, called and talked to her quite a bit about what she might work on. So that [then] I could have picked a project in advance and made sure that everything was set up so that she could work on that. As opposed to when she showed up, deciding that she needed to work on something and then we had to go and find the right graduate student who could advise her on that.

I guess my advice would be to maybe even talk to the students before they came. I haven't done that, but I think I would in the future just to make sure that kind of your understanding of what the summer was going to be like meshed with theirs.

V.A.2. The completion of a project that contributes to the mentor's research increased students' confidence and gave them a sense of accomplishment

Most mentors stated that by successfully completing a difficult research project, students felt a sense of accomplishment, developed more confidence about their ability to do research, and also came to realize that they could contribute something of interest to the research community.

Think carefully about the project and make sure that it is something that an undergrad is likely to have the skills for and that it's contained in the sense that at the end of the summer, assuming everything goes reasonably well, the student can look back and see that they've really accomplished something that seems like it was worthwhile. I think it's nice if it can have a place in a bigger research project because then it seems more clear that it's important if it's not just a stand alone thing but it's going to contribute in some other bigger project that's going to continue after you've left.

[For] the project itself, concentrate on something tangible that they can take with them, because I think that was something important to [my DMP student] that she could take this and say, "I wrote this." Or, "I did this, and this was a research project." Rather than [if] it was implementing some little module in a huge program, and she could do that, but I don't think that would be as interesting or as fulfilling.

Pick a project that a student can easily do in the time allotted. And to not either over expect or under expect for them. I mean try to give them something that's going to make them feel good about themselves by the end.

R: Well that was real important to me [to have a project formed that could be done by a student during the program]. Whatever area they picked to work on I wanted them to be able to pose a question and have as their goal to answer it by the end of the summer. Or you know to make some progress on it by the end of the summer that would be tangible that you could write a report on. Not just to write a report, I mean I wanted it to be an important question that other researchers in the area would be interested in. So maybe they couldn't answer the whole question but maybe they could get some new data that nobody had seen before that would help to answer the question eventually.

I: Why did you think that was an important thing for them?

R: Well I guess I feel like research is a lot more fun if you solve your problem or if something comes of it, you know if you feel like you've made a contribution somehow.

In the student section we discussed that students derived satisfaction from contributing to the mentor's research. In fact, many mentors themselves discussed that the students wanted to feel their project was useful to others in the research community.

R: Because both of [my DMP students], I found, if they didn't have something solid by sort of midway through that they could say, "Wow, this is what I did." Both of them were like, "Gee, I feel like I haven't done anything." So I didn't want them to go away with that feeling.

I: What did you do? Did they bring that up with you?

R: ... I'd give them some training materials and met with them, and then after like four weeks she didn't like what I had given her. She did the work, and then she decided, "If I had something else it would be better." So I switched her project over, and she wanted something in a little bit different area and more solid. So they really did want to contribute. They didn't want to just kind of sit around and play on the machines all summer.

V.A.3. Multiple projects provided relief when students were bored, frustrated or had technical difficulties

In addition, many mentors suggested having multiple projects prepared at the beginning of the program. These mentors expressed that the student would be more motivated when she was able to choose a project that interested her.

I: How did you decide what the students would work on?

R: I always have kind of a mental list of problems that I want to work on and for both summers I kind of looked at it and came up with a list of five or six problems that I thought they might be able to make some progress on it for a summers worth of work. So my plan was to start them with say one problem per week until they found one that they made some progress on or that they liked a lot. ... Some of the problems they look at and they write a program maybe to find out something about it but they don't see anything, they don't see any direction to go from there. And maybe I don't either from their preliminary work and maybe they didn't get very excited about it either. So one of the things when you're trying to solve a problem it helps if you really care about what the answers gonna be. So maybe in some of the ones that didn't catch they either felt like they couldn't get anywhere or they weren't very interested or wasn't very much fun to manipulate these particular mathematical objects or something like that.

I got to know [my first DMP student]. I tried to find out what she wanted to do, what kind of research she thought she might be interested in. And I tried to give her that particular type of research. Like, I didn't peg them into a single project. I made sure that I had a couple of things available that they could do and even switch mid-stream if they decided that they hated it. Which is what happened with [my second DMP student].

These mentors emphasized that having multiple projects would prevent the student from getting too frustrated or bored by one project and allow her to learn different material.

I: What if the [DMP students] e-mailed you and said, "HELP!" How did you deal with that?

R: They never did. I always gave them several things you know to think about from one week to the next. So if they weren't making progress on one they could focus on a different one.

R: If [the DMP students] wanted to try something else we'd try to think of something else for them.

I: Why did you structure it that way?

R: Well, I think we didn't want to have an experience where someone would come and just hit a brick wall and feel frustrated that they couldn't do the particular thing or this was the task they had to do for the summer and they couldn't do it. I think they wanted to give them flexibility. Everyone is different. There are certain things you like doing or feel more confident doing than other things. We just felt it was good to give people options. In particular, not let them feel like they were having a negative experience and that they hated the whole situation.

We had some projects for them to do, and I don't know how this happened the first year, but we sort of all knew by the second year that the way to do it was to assign a couple different projects so that if they got stuck with one or if the computers went down they could look at something else. And so this year we definitely have had about three different projects for them to bounce around in their heads. It just gets boring working on just one.

V.A.4. Having multiple projects benefited the mentor by contributing to her research

In addition to creating a successful experience for the student, most mentors stated that a well-defined and doable project would also benefit the mentor in that it would contribute to their research program. When we asked mentors to give advice to other mentors who were considering participating in the DMP, many commented that having a well-thought out project benefits both the student and mentor.

I: What advice would you give to someone who was considering participating in this program as a mentor?

R: Okay. Two things: first, the mentor should really realize that her project -- that an undergraduate student who does not have much background in it should be able to pick it up easily, so that the mentor can get something out of the work that she put in. That's the major thing.

V.B. Immersing students in a research community

Another strategy for a creating a successful DMP experience was to immerse students in a research community that can include the mentor, graduate students, undergraduate students, and other faculty. Many mentors stated that involving a student in a research community would benefit the student by providing multiple resources and social interaction while also benefiting the mentor by decreasing the time demands of mentoring an undergraduate.

V.B.1. Involving students in a community that included graduate students gave students a more accurate view of graduate school

Given that all of the mentors felt that the purpose of the DMP was to encourage female undergraduates to consider graduate school, many commented that it was important to have a community of graduate students with whom the students could interact. These mentors stated that through observing and interacting with graduate students, their DMP student would develop an understanding of graduate school that she may not receive through her interactions with her mentor alone. Some mentors purposefully chose projects that ensured that students would interact with graduate students in order to give the DMP students more experience with graduate school life.

R: But I think it's really important that they have an environment and that they feel like they're part of the group as opposed to just sort of coming and being by themselves for the summer and maybe talking to one faculty member. So I think that's really a useful part of what happens here. And that might be harder to set up at a place where you have a smaller graduate program or something like that. But I don't know. That's important.

I: Why is it important?

R: Because I think that is really what gives the students more of a feeling for what graduate school would be like. And I think that's one of the goals of the program is to give them a feeling for what it would be like if they went to graduate school. And if they go in and just work with an individual faculty member, it's not an accurate reflection of what graduate school is like. The graduate students learn a lot from each other you know, probably more than they learn from us because the other students are all very smart and are learning things at the same time, in the same areas. And so I think that interaction is a really important part of graduate school. And if they don't see that I think they will get an inaccurate idea of what it would be like to do research.

R: Basically I was trying to come up with something that would help my current students. What she did last summer was helping a student that I had then. That student's now graduated so now she's working with a different student. So I wanted her to be able to not just be working with me but also be working with a graduate student. So I try to get the project close enough to that graduate students research that she can communicate with them.

I: Why is it important for her to work with graduate students?

R: Well I thought part of the purpose of the program was for them to get a feel of graduate school so I purposely wanted them to work with a graduate student on a personal basis so they learn a lot just from having that student tell them about graduate school.

R: I think the graduate students were much better role models than I was.

I: Why would that be? R: Because they're closer in age and they're closer to the kind of experience they're doing. They kind of were the next step, I mean I did that too when I did what the graduate students do when I was a grad student.

V.B.2. Involving students in a research community gave them a better understanding of the collaborative research process

Many mentors stated that through their interaction with other graduate students and faculty, students not only learned about graduate school life, but also about the nature of a collaborative research process. By participating in meetings and working with other researchers, students were able to observe the ways in which graduate students and faculty interact professionally when discussing their research.

Many mentors commented that their DMP student initially did not feel comfortable providing input at research meetings, preferring to defer her opinion to graduate students and faculty. These mentors stressed to their student that they did not know all of the answers in research and the student was capable of contributing valuable information to research discussions. In the following quote, one mentor explained the benefits of her student participating in their research meetings.

R: The two [DMP] students were in our parallel lab, and there were also graduate students on the same project, so we had weekly meetings, or biweekly meetings, or sometimes it would be three or four people, a couple other students, and sometimes it would just be with them, or one particular undergrad. They saw this interaction going on, and I think that was probably something they had never been able to participate in before.

I: Right, kind of an interaction of ideas as opposed to a one-sided --

R: Yeah, right. They were saying things like, "Oh, well, we're stuck here, what do we do?" And it was okay to say, "We were stuck."-- for faculty or for the advanced grad students to admit that. Or sometimes the grad students were looking for things, or they needed some piece of code or something that the undergraduates had done, so I think they felt good that their stuff was being used, that there were questions coming from other students, not at their own level, but even higher up. The reverse was also true, so it worked out pretty well.

Many mentors commented that when students participated in a research community, they experienced research as an open-ended process in which faculty and graduate students work collaboratively to solve problems. Many faculty stated that they perceived a change in their student's confidence level about her ability to contribute valuable information through these meetings.

Because when I had seen her resume, I mean, she looked really good on paper, but when she first arrived, she seemed to be very very unsure. I think it was an intimidating experience for her initially because she had never done research before, and what we tried to do was we had the project, [and] we knew exactly what needed to be done. But to [give her] a good feeling for what research is, we tried to engage her in research discussion. So we would sort of sit around and talk about what needed to be done in the project and hoped that she would sort of come up with her own ideas. Initially she was very very shy, and she would speak to me all right, but when we would sit down in a group she would really not say very much. And I think that towards the end that changed a lot. She was speaking up more and she was able to contradict us, which I think is a very good sign, especially in a research environment, and I think that happened. So when she finally left she said that aside from her skills, her programming skills improved, she learned about simulation, she learned more about database systems, and I think all of that will help her in her career, but I think she became a lot more confident, and I think that was the biggest achievement of the ten weeks.

V.B.3. A research community provided multiple resources for the student

Most mentors stated that a community of graduate students, undergraduates and/or faculty members afforded multiple resources for the student that the mentor may be unable to provide by herself. In addition to being a resource for research-related questions, a community could also provide an element of social interaction for these students who often knew no one at their mentor's university.

I mean the one problem I can see that might happen for the program like this is if they assign the mentees to a person who doesn't have grad students working for them or for which there isn't some sort of social environment. Because I'm at a really different place. You know, I'm over 40, I have children. I'm not going out every night and dancing like I think a lot of the undergraduates were interested in doing. But the graduate students were doing that stuff, so they could kind of fill in the social vacuum for them. And I think that if the mentees worked at a place where the social vacuum couldn't be filled in they could of course be with each other and do stuff with each other, but it wouldn't be as good of an experience.

It is nice that [the DMP students] fit into a group and I know with [one of my students who also participated in the DMP at another university] did the same thing. And there was a group working already of grad students, so you aren't just left alone. And I think that's important, that if you're looking for places for the students to go, the fact that there's a working group there of some form. You know, even if it's a group of grad students that are not working on this project but working through the summer, that they can interact with. I think that that's important.

I think it also helps if you're in an environment where you either have your own lab or at least have several students working for you so that you are not the only source of inspiration and help for that student. I think it's very difficult to take somebody in for a summer and expect that you're going to instantly have a really good one-on-one working relationship. So you have to provide other opportunities to them.

In the following quote, a mentor discussed the multiple resources that a community can provide for the DMP student new to the city and university.

The one awkward part of it was setting up living arrangements for the student. And that was my primary concern going into it is [that] you don't know what the student is going to be like, what they want to do, how aggressive they are in taking care of themselves, whether they need to be hand held, that kind of thing. And if you bring in one student as we did, it can be a little lonely. And we tried to make sure that both of us had graduate students, and I in fact had an undergraduate working for me that summer. So we tried to make sure that she had people to interact with, to ask questions about the area, to go to lunch with, that kind of thing. But those sorts of logistics are something that we don't think about on a daily basis, and that was a concern for both of us when we brought her in. And was she going to feel extremely isolated and was she going to be shy or have difficulty getting around the new environment.

Many mentors noted that by having the students work with or be in close proximity to other graduate students, they were alleviating some of the pressure the student may feel in interacting with a faculty member. These mentors felt that it may be less stressful for the student to be able to ask more technical questions of a graduate student than a faculty member.

I: Did they have grad students that were supervising them or that they were working closely with?

R: Yeah, they had grad students that they were working with. There was one that sort of had finished last year, but she was still around. She, actually last summer, was here full time and helped [my 1994 DMP student] get started. In some ways I think that was a comfortable transition for [this DMP student] because she felt less intimidated talking to [the graduate student] first. ... Plus they were physically located together. They were in the same room, so it was easier for them to talk than to make a trip around and look for [my co-mentor and me] in our offices.

V.B.4. A community decreased the time commitment of the mentor

Many mentors who provided a research community that included graduate students for their DMP student reported that it decreased their time commitment with the student. As stated throughout this document, the students had little experience with research and thus needed a high degree of guidance. This meant that a mentor working without a research community of graduate students may need to spend a considerable amount of time assisting the student. Since many of the mentors had multiple responsibilities which at times made them inaccessible for the student, they felt that a community of graduate students would be more available to answer students' questions. In fact, many mentors stated that since the graduate students were around more often, they would provide a more consistent resource for the students' day-to-day technical questions.

R: I think it's really important to have a graduate student who is also acting as an advisor or [the students] just won't get enough advice. Sort of supervision, whatever. I: Because they can't get it all from you?

R: Yeah. Because it's not reasonable for me to assume that I'm going to be here at every time of the day. [That] they can just drop in and talk to me just because even when I'm here I have meetings and meetings with other graduate students. Meetings with faculty. Different kinds of things. So I think it's really important. The graduate students tend to be in town more and they're just more sort of constant sources of information if they need that kind of immediate answers to questions.

R: I think it would be ideal to set it up so that the student is working very closely with a graduate student and a lot of the questions go to the graduate student and then the faculty member gets involved on maybe a little bit less constant basis. But you know that then depends on having a graduate student who's gonna do a good job of being a mentor and of involving someone else in a project and that's not something that a lot of the graduate students necessarily, you know they're sort of still trying to make their own way in the project that they're working on. And I don't think that every graduate student is necessarily gonna be good at playing that role and being involved in that way.

I: And why would that be ideal to have them working closely with a graduate student?

R: Well I think because they would have a lot more, I mean the graduate students are closer to them in terms of experience and you know, you can quickly forget what it's like to you know to be thinking about research and just starting out with it. And I think that having a graduate student around that a student, an undergrad can work closely with would just provide sort of probably more opportunity for interaction and more frequent interaction than a faculty member can usually provide. Just because of time constraints and you know I think I was fairly accessible but I was probably not as accessible as, you know a grad student working at a nearby desk would have been.

The amount of time I would have to work with any one person, even the grad students is relatively small and if they're here to spend all their time with me, that can be a potentially discouraging setting. So that's where the grad students being there to be there on the kind of you know, day by day, hour by hour basis really helped [my DMP students] a lot. But in terms of the program is set up as faculty -- undergrads. Unless that's kind of built into it, I think there is some potential for it not working out so well.

Almost all of the mentors who had a graduate student work with the DMP student did not view the graduate student as replacing them in their role as mentor. Rather, these mentors stated that the graduate student provided a different function: the graduate student answered the day-to-day questions, while the mentor provided assistance on the "big picture." Many mentors felt that the two roles complemented one another and ensured that the student had access to the help she needed.

Evaluator Question Should the CRA-DMP require that the mentor have a community of graduate students with which the students can work over the summer?