Doing More Than

This article is printed in Research in the Teaching of English,
February 1995, Volume 29, No. 1, pp. 59-87.

Doing More Than "Thinning Out the Herd":
How Eighty-Two College Seniors
Perceived Writing-Intensive Classes
Thomas L. Hilgers
Ann Shea Bayer
Monica Stitt-Bergh
Megumi Taniguchi
Mānoa Writing Program
University of Hawai‘i at Mānoa

More and more college campuses are offering one or another form of "writing-intensive" classes across the curriculum. This study investigates what students perceive to be the effects of the writing-intensive requirement at the University of Hawai‘i at Mānoa where students are required to take five courses designated as writing-intensive. To identify the potential composite effects of taking three or more writing-intensive classes and to identify evidence of learning that may have resulted from these multiple experiences, we interviewed 82 randomly selected seniors. Using interview transcriptions, we developed a scheme for analysis of the data. These analyses revealed several areas of self-identified improvement associated with writing-intensive classes: writing skills, knowledge acquisition, and problem-solving abilities. Students also reported that they had become better writers through interaction with their professors during the writing process, although they also reported wanting to better understand the philosophy behind writing-across-the-curriculum and the purposes of specific assignments. These student-reported effects of writing-intensive classes support the notion that writing can play an important part in learning.

At first I thought the purpose of the writing-intensive (WI) requirement was to 'thin out the herd.' I couldn't write well. I thought they threw this at me to get me out of here. But I finally took one WI class, then two, then as many as I could. Now when I go back home and my old friends who never came to UH see me, they say "What's the matter with you?" I've changed---my writing, my speech---it had a domino effect. My attitude changed. At some point I realized that the writing-intensive requirement is to help me, not to get rid of me. And it really did help!---University of Hawai‘i Senior

In 1987 the Faculty Senate of the University of Hawai‘i at Mānoa and the University's Board of Regents adopted a radical addition to the flagship campus' General Education Core Requirements: Students in all majors must complete courses, labeled writing-intensive (WI), that use writing to help students learn course content. While the WI requirement was being phased in, freshmen were required to take three WI courses to graduate. Today, students entering the University must take five WI classes and can choose from over 400 WI classes each semester in more than 70 departments.
 To administer this new category of courses, the University's Regents created the Mānoa Writing Program (MWP). The MWP oversees the designation of courses as writing-intensive and otherwise supports WI instruction. However, individual departments administer and evaluate the WI classes offered by their faculties.
 Virtually any undergraduate class can be proposed for offering in the writing-intensive mode. The processes for proposal of a class as WI, and its subsequent official designation as WI, are faculty-governed from beginning to end.
 Individual instructors first request the WI designation by writing descriptions of how they will use writing in a course they will teach the following semester. (Any instructor may request the WI designation. However, calendar requirements limit the possibility of requests from lecturers and graduate assistants, and more than 90% of the requests come from tenure-line professors.) Instructors ask their department chair to forward their descriptions to the Mānoa Writing Program Faculty Board.
 The Faculty Board nine professors from different departments appointed by the vice president for academic affairs then reviews the descriptions. Designation of classes as WI is determined by whether or not the class will meet set criteria:

  1. The course promotes the learning of course materials through the use of different types of writing.
  2. The instructor provides students with feedback on their writing, particularly drafts and revisions.
  3. The writing assignments contribute significantly to the student's course grade.
  4. The course requires a substantial amount of writing---a minimum of 4000 words, or about 16 pages.
  5. The course enrollment is restricted to 20 students to enable meaningful professor-student interaction on each student's writing.

Next, the Faculty Board approves the WI designation for proposed classes that will satisfy the criteria. If a class does not meet the criteria, the Board negotiates with the professor who is requesting the WI designation to modify the course so that it may meet the criteria. Any resulting modified course descriptions are considered in a final round of discussions by the Faculty Board.
 The director of the Mānoa Writing Program submits the list of classes approved for designation as WI to the registrar, who caps enrollments of the class sections at 20 and lists them as WI in the official class schedule used by students when they register.
 Finally, instructors who will be teaching WI classes are invited to participate in a workshop at the beginning of the semester on teaching with writing. Although the workshops for instructors of WI classes are voluntary, more than 225 professors have participated in five-day, three-day or one-day workshops on teaching with writing. In addition, shorter customized workshops have been offered in individual departments and for faculty members who teach sections of large courses. All instructors -who will teach a WI class for the first time receive extensive support materials, including writing assignments and teaching strategies that previous professors of WI courses at the University of Hawai‘i have found effective (Hilgers, Cole, Tsutsui, & Wallace, 1991; Hilgers, Bayer, Hussey, & Chinn, 1993).

 Theoretical Framework
 The study reported here continues the assessment research undertaken during the Mānoa Writing Program's first five years (Hilgers et al. 1991; Hilgers et al. 1993; Marsella, Hilgers, & McLaren, 1992). That research, coupled with comments made by students and instructors on end-of-semester surveys, provided a significant base of descriptive data that included descriptions of teaching strategies followed in 24 WI classes, each from a different field. Commentaries and intention-statements from the professors and students who constituted the observed classes usually accompanied the descriptions. The database also included a few "case studies" that looked at the strategies students employed to deal with assignments in WI classes (Marsella et al., 1992).
 Although these studies yielded a broad sketch of teaching strategies in WI classes and some sense of how students handle individual writing assignments, none of the studies considered possible effects on students who enrolled in more than one WI class, in more than one discipline, and in both upper and lower divisions. The authors decided to augment the earlier research by building a data base that would identify the potential effects of taking three or more WI classes. Identifying these effects could benefit teachers and local program decision-makers.
 This study also sought to elicit information that would be useful to the far larger audience concerned with educational reform. More and more college campuses are offering one or another form of WI classes: Surveys from the late 1980s showed that 30 to 50% of postsecondary institutions had writing-across-the-curriculum (WAC) programs (Kinneavy, 1987; McLeod, 1992). WI classes are perhaps the most visible manifestation of higher education's efforts to institutionalize renewed concern with the effectiveness of teaching practice in undergraduate classes (Herrington & Moran, 1992). Yet many interrelated components of such classes remain unstudied (Marsella et al., 1992; Stotsky, 1992), in part because of the difficulties associated with research that aims to demonstrate learning as a result of particular activities (Schumacher & Nash, 1991). Indeed, Ackerman (1993), in the most comprehensive review of research on writing and learning to date, found no compelling case to support the commonly held beliefs behind the WAC movement. And one common assumption, that teachers can "teach" students the particular genres of writing in their own disciplines, is under attack (for example, see the discussion in Freedman, 1993a, 1993b; Williams & Colomb, 1993).
 Given such questions about basic assumptions underlying writing-to-learn pedagogies, the study reported here attempted to illuminate some heretofore dimly lit areas of the WAC landscape. The University of Hawai‘i provided access to an extensive program for examination. Although only its fourth year, the overall program had achieved stability. During the program's brief history, professors have taught more than 2,000 WI classes, all of them following guidelines mandating the frequent use of writing to promote learning, the provision of feedback, and a minimum quantity. And although 1992 seniors needed no more than three WI classes to graduate, the typical senior from the sample had actually taken five WI classes.
 This study's primary goal was to answer the question, "How are we doing?" from two perspectives." The first perspective responded to local professors' interest in how their students assess teaching efforts to improve students' learning and writing in undergraduate classes conducted in a writing-intensive mode. The second perspective sought to understand how students described the effects of their multiple experiences in WI classes, particularly as they related to student learning.
 Method
The relative dearth of findings in support of writing-based instruction may in part result from an inappropriate "fit" between the complexities of human learning and conventional research designs. Ackerman (1993) pointed out that conventional quantitative designs are ill-equipped to deal with the varieties of practices and expectations associated with writing in different academic disciplines, and they are insensitive to social and cultural factors associated with writing and learning. Others (Guba &Lincoln, 1989; Mishler, 1979) have gone further, raising a chorus of questions about the validity of the positivist assumptions behind much assessment research.
 The problem of design is twofold. A research design that fits less than optimally into the phenomenon of interest is likely to miss potentially critical elements that play into the phenomenon. Second, findings derived from such a study can mislead, particularly when relevant "variables" are many, poorly defined, or poorly understood (Hilgers & Marsella, 1992).
 Clearly, the critical components of writing-intensive instruction in different disciplines have not yet been identified. It is equally clear that the particular configurations of components in the Hawai‘i WI classes are not stable (and may never be). In addition, students have never been asked to reflect upon their learning experiences other than through end-of-course evaluations. There has been no effort to consider possible cumulative effects of WI-class experiences. We needed a research design that encouraged students to think critically about their experiences. These realities suggested that conventional quantitative research designs, while potentially helpful in different circumstances, were unlikely to provide answers to our research questions.
 Consumer-Based Assessment
 To identify the potential composite effects of WI classes, we decided to borrow a page from consumer-based research and focus on the students as "consumers" of WI classes. The students were, after all, "consumers" as well as intended beneficiaries of the WI classes required for their graduation. A review of approaches to consumer research led us to give special attention to interviewing. It appeared that semi-structured interviews would be more likely than several other approaches to yield valid answers to our research questions. True, there are several reported difficulties associated with the use of interviews for purposes of assessment research. How to devise and ask questions (Lofland & Lofland, 1984; Berg, 1989) is a sticky issue when associated with interviewer neutrality (Dross, O'Brien, & Marsh, 1987; Lee, 1991) and the desire for information rich responses (Charmaz, 1991; Nelson-Gray, Haas, Romano, Herbert, & Herbert, 1989). Interviewee expectations of what constitutes appropriate behavior during an interview can lead to misleading or otherwise invalid responses (Charmaz, 1991; Garretson & Teel, 1982; Mostyn, 1985; Richardson, Dohrenwend, & Klein, 1965). Cultural differences and interviewer ignorance can lead to problems in analysis of responses (Dross et al. 1987; Lee, 1991). Finally, interviews are labor and time intensive (Doll & Jacobs, 1988).
 The potential advantages of interviews, however, make attempts to overcome the difficulties worthwhile. The interview setting excels in its potential for "naturalness:" When the setting is appropriate and rapport is achieved (Embrey, Mondy, & Noe, 1979; Lee, 1991), the interview is akin to a conversation. As an open-ended research technique, the interview allows complexity to emerge (McCracken, 1988). A well-executed interview allows the interviewee to probe his or her memories, articulate tentative judgments on experiences, and then, after additional consideration, to expand or even modify the judgments. A well-conducted interview, then, is more likely to yield a valid consideration of complex phenomena than is, for example, a questionnaire which assumes that 1) all respondents have thought about the phenomena of interest; 2) their thinking on the phenomena is finalized before they completed the questionnaire; and 3) new thinking will not occur as a result of responding to questions.
 For guidance in overcoming the difficulties associated with interviewing, we turned to assessment projects that had used interviews. "Exit interviews" with consumers are not unusual in the non-academic world. They occur frequently in conjunction with product (including movie) "sneak previews," and they are referred to in television coverage of every major election. In-house researchers in large corporations who are interested in personnel matters use exit interviews, and researchers use interviews to gauge consumer satisfaction with both new and longstanding products (Schmidt, 1984).
 The use of exit interviews in higher education, however, is infrequent (Light, Singer, & Willett, 1990). Several small, mainly private and sectarian, colleges associated with the Appalachian College Assessment Consortium (administered at the University of Kentucky) use exit interviews as a capstone educational experience for random samples of graduates and to assess student accomplishment of member colleges' particular goals (Street, 1991). Large-school assessment involving exit interviews is perhaps most advanced at Harvard College. The primary method in Harvard's assessment research has been peer-conducted interviews with graduating seniors (Light, 1990, 1992a, 1992b).
 Interviewees
 Since our primary questions could best be answered by "experienced" students, we limited our investigation to seniors, as measured by accumulated credit hours, who had taken at least three WI classes. The University Registrar's Office compiled a list of students who met this criteria, and we mailed letters of invitation to random students from the compiled list as need arose; invitations were sent out in lots of 50. Sets of invitations were to be sent out until approximately 80 students had signed up for interviews. To compensate interviewees for the time they provided, and to signal the importance of what they were being asked, each interviewee was paid $20.00 for the hour-long interview.
 Developing Interview Questions
 We formulated specific research questions based on instructor concerns, research in the larger field of composition studies, pedagogical theory, and the concerns of University administrators. The assessment team was concerned with the techniques for effective interviewing. We wanted to invite expansive responses from our interviewees to give them an opportunity to reflect constructively on their experiences. We wanted to minimize the "say-what-will-please-the-interviewer" characteristics of student responses. To enhance the likelihood that the comments would be valid, we committed ourselves to making our interview settings as conversational as possible. We used graduate students as interviewers; they were close in age to the undergraduate seniors and were trained to encourage identification between themselves and those with whom they talked. Interviewers were also trained to articulate follow-up questions for eliciting a maximum amount of information and to check the reliability of responses.
 A review of existing literature yielded several procedural hints that were tested in a series of pilot interviews (interviews not included in the final sample). We conducted the pilot interviews in three segments. The first segment built upon a set of questions created by the full research team after brainstorming sessions. The two student interviewers tested both the questions and techniques by each doing one interview. After transcripts from the interviews were prepared, the full team reviewed them and decided upon new questions and strategies. These new questions were employed by each interviewer in another interview. Questions were again modified, primarily to reorder questions and to eliminate jargon. During the 6 subsequent pilot interviews, interviewers were allowed to modify the questions and then discuss reasons and effects in debriefing sessions following each interview. It was during this phase of pilot testing that the final battery of interview questions emerged (see Appendix).
 Coding System
 The next task was to develop a set of codes that could be applied to transcripts, which ranged from 9 to 18 single-spaced pages. We developed the codes to help the research team make sense of the cumulative record provided by the full set of interview transcripts. The coding scheme, therefore, served to disentangle students' responses that might address concerns not expected to be cued by specific question and to deal with comments that shifted readily from topic to topic. One student, for example, responded to question 7, "Now how do you feel about writing?" with "It's a pain;" another with "Fine." But other interviewees took time to explain their answer: "I feel good. Because now when I write like in my science lab I write down what we do and what happens and it helps me understand the experiment. And then it makes studying for the final easier because I've learned a lot already by writing up lab reports every week." Because the interviewee discussed improved knowledge acquisition, reducing this latter response to only a check-list item for question 7, positive feelings, would not accurately reflect the interviewee's answer. The codes, in sum, needed to highlight topics of interest that might be mentioned by students at any point in the interviews.
 Guided by the five primary questions, the interview team created a list of topic areas that emerged from the interview data. But we also wanted to be able to capture individual differences among interviewees. So, using the general categories, we went to a sample of four transcripts, coded them, and created sub-codes for idiosyncratic responses under the appropriate general category. Samples appear in Table 1. (Copies of the full coding scheme are available from the authors.)
 The transcripts were then coded on a computer using Textbase Alpha (Sommerlund & Kristensen, 1989) by the two student interviewers. Twenty percent of each coder's transcripts were also coded by the other coder as a check on coder reliability. After all 82 transcripts had been coded, a new paid coder, who was unfamiliar with both the purposes of the study and the coding procedures, joined the research effort. She was first trained in the meanings and markers for each of the code items used by the two primary coders. She then was given a random sample of 10% of the transcripts to code independently of the primary coders. When the outside coder identified a minimum of 80% of items coded to a category by one of the primary readers, the code category would be deemed stable and identifiable. When the outside coder's efforts suggested that category modification would more accurately reflect interviewee comments, the research team would attempt a redefinition of the category; if all agreed on the new definition, transcripts would be recorded on that category. Where cross-coder agreement did not reach 80% of items coded to a category, the category would be deleted.

Table 1
Sample Codes and Subcodes

M. Writing in classes within their major
    M.p Purpose
    M.p.1 Prepare them for career-specific writing tasks
    M.p.2 Does not prepare them for future writing tasks
    M.p.3 SO instructor can evaluate/grade the students
C. View of restricted class size (20 students)
C.s No difference between class size of writing-intensive classes and other classes
C.i Small class size in writing-intensive classes results in more attention from instructor
C.d Small class size promotes interaction among students
SM. Suggested modifications to writing-intensive classes
SM.1 Train teachers to teach writing-intensive classes better
SM.2 Improve the image of writing-intensive by changing the name
SM.10 Teachers should provide more information on writing resources (e.g., reference books, workshops)

Representativeness of Interviewees

 From the initial pool of seniors meeting the criteria, we sent invitations for interviews to 300 randomly selected students. There was no way to ensure that positive responses to the invitation---ultimately made by 82 of the invitees---would come from the representative sample guaranteed in a true random design.
 However, three indicators provide approximate gauges of representativeness. Table 2 shows the distribution of the 82 interviewees by chosen major and distribution of undergraduate majors at the University of Hawai‘i at Mānoa. Analysis revealed no significant differences (X^2= 9.6, df = 5).
 Because our personal interview format did not seek self-described ethnic information, we could use only an approximater---interviewees' family names---as an indicator of apparent ethnicity. This approximater, although somewhat crude, often coincides with self-described ethnicity, which individuals, in any event, sometimes change as part of the shifting political and social climate. Analysis revealed no significant differences between the apparent ethnicity of the 82 interviewees and the self reported ethnicity of the University of Hawai‘i at Mānoa undergraduate population (X^2 = 7.4, df = 5).
 Finally, we compared the grade-point ratio of interviewees with that of all seniors. The 82 interviewees' grade-point average was 3.04; for 5628 seniors, it was 3.0. The grade-point averages of our interviewees were not significantly different from that of their class (t = 0.8).

Table 2
 

Majors of Interviewees and of Undergraduate Population

Major

Interviewees (N=82)

UHM Undergraduate
Population (N=12,560)

Arts and Humanities

23%

18%

Education

11%

7%

Human Resources

16%

13%

Natural Sciences 

9%

12%

Professional Schools 

16%

29%

Social Sciences

25%

21%

Interviewer Fidelity to Question Battery
 Fidelity was initially defined as asking all questions; after pilot testing, fidelity was redefined as eliciting information in all areas covered by the questions. To maintain the conversational flow of the interview, interviewers were allowed to omit asking a particular question if an interviewee had addressed the question's content in response to an earlier query. Some leeway in question order and wording was allowed to accommodate interviewees' digressions and responses that anticipated future questions. To help ensure fidelity to the question battery, interviewers were instructed to discretely check questions as respondents addressed their contents. Analysis of transcripts showed that the interviewers were faithful to the final battery of interview questions over the course of 82 interviews.
 Stability of Code Categories
 As reported earlier, cross-rater identifiability and reliability were achieved on about 80% of the coded categories. Analysis by the outside coder led to modified descriptions of two categories and a consequent re-coding of all transcripts using the modified category description. Acceptable agreement could not, however, be achieved on the remaining 20% of code categories. In most cases, disagreement involved categories devised to highlight subtle, usually idiosyncratic, differences between similar responses. What is reported in the remainder of this article reflects only those code categories that were found stable and reliably identifiable.
 Pattern of Enrollment in WI Classes
 As already mentioned, interviewees had typically taken 5 WI classes although they were required to take only 3. Figure 1 shows that typical students take more than half of their writing-intensive classes during their senior year. (However, the "senior" year may be atypically long: Students become seniors once they earn 89 credits, and more than half of UH graduates accumulate 40 or more credits as "seniors.")

Figure 1. Typical percentages of writing-intensive classes
taken during college years.

Results
 The present study was intended to guide local program development by looking for evidence of learning that may have resulted from multiple experiences in WI classes. The study was designed around these five questions:

  1. What do experienced students identify as the benefits, if any, of WI courses?
  2. How do these students see relationships between WI instruction and course work in their majors?
  3. How do experienced students describe successful WI instructors (if they judged some as successful)?
  4. How do students view class-size restrictions in WI classes?
  5. What would experienced students do to improve WI instruction?

The report of what we learned is organized into sections that address our findings regarding each of these questions.
 What Do Experienced Students Identify as the Benefits of Writing-Intensive (WI) Courses?
 Improved Writing Skills.
 Responses related to improvement of writing appeared in transcripts from all 82 seniors. Of the interviewed students, 78% indicated that WI instruction had improved their writing or their confidence in their writing abilities (or both). Twelve percent were somewhat ambiguous in their responses. And 10% stated that WI instruction did not improve their writing.
What exactly did the students say? The following excerpts from the transcriptions illustrate typical responses in each of the above categories. (Some excerpts have been lightly edited to remove the overly frequent like and you know). The first excerpts are from the 62 students who talked about how WI instruction had improved their writing or their confidence.

Many people come to college, get this far not being able to write well, or not being able to write at all sometimes. And I don't think I had that problem but I am better now even though I think I could write quite well when I got here. Now I'm even better than before. Everybody benefits from these courses. I believe WI classes have improved my writing because I have to do it more than once and so I keep getting feedback on what I need to improve on. By the end of the semester I shouldn't be making those mistakes that I did at the beginning and probably at the end of the semester my writing should be a lot better than what I had started with.
 [WI classes] forced me to write more and I think as far as writing goes that's the only way to learn, at least for me. Just having to write more and think about writing, it improves vocabulary and improves your structure, it improves, how you put things together, and one paper builds off the next. You think, 'Oh this is a great way to do this' and the next time you try that and maybe try another way to approach it---you just have little breakthroughs every time you write. I think it's helpful.
 Yes, [WI courses] definitely [helped me improve writing]. [Q: Which courses?] All of them. I'd say all of them. Because---mainly because the teachers will give you direct help. They'll---they know that the requirement's there to help improve your writing so they'll give you help, you know, if you need help writing a draft. Like my teachers would always tell me, "Anytime you need any help, bring in a draft and I'll help you out."

Several students talked about increased confidence in their writing:

I used to be very afraid to write anything down, and now I'm like, "Eehh, I'll do it"---hand in a rough draft and work on it from there.
 In a way I got more confidence in myself because they [WI instructors] show you your strengths and also showed me what I should improve on my own. In some classes that are not writing-intensive they'll just give you a grade, but you don't know why you got that grade.

Ten students waffled in their responses to the questions, "Do you think writing-intensive courses improved your writing?" and "Which courses helped your writing?"

Certain ones [improved my writing]. But others it---wasn't as if it was a writing-intensive class, so no. [Q: Which classes did improve your writing?] The ones where they gave me feedback on my writing. The other two - didn't.
 I don't know if they improved my writing, but I can say that they improved my understanding of what I was writing about. Except for this semester---like I said to you. The professor didn't sit down one-to-one with each of the students and say, Okay, this is what needs to be improved and why I think so.'

Eight students stated that WI instruction did not improve their writing:

It improved my ability to think; they have not improved my grammar or spelling one bit.
 I don't know. Improved? I think once you get out of high school your writing style is pretty much set. So I don't know if it necessarily improved it. Keeps you in practice. But I don't know about necessarily improving.

Improved Knowledge Acquisition.
 A second major finding involves the role of writing in helping students to understand and retain course content. Of the interviewed seniors, 83% reported that WI classes helped them understand course material better retain course material longer (or both):

When you have to write a paper, you have to actually read [the material]; otherwise you can't write the paper. Some tests---like objective tests---sometimes you can fudge your way through, but really, you might not have gone to class the whole semester; maybe you just skimmed or whatever. But to write about it---I mean you actually had to read it well, and understand it, and then you give your interpretation of it while you analyze what was written. I think that's helpful. You can learn a lot that way.
 The teacher would always say that they can't cover as much as they want to cover or that students' interests always lie elsewhere from the lectures, so they want an opportunity for the students to pursue something in their own interest in that particular subject or field of study. So the paper is really not so much for the teacher or for the professor's benefit, it's more for the student's learning or experience. So students get more out of the class than usual.
 [WI classes] make you think about the subject a lot more than you would normally. If you have a paper due every week, it sure helps you prepare for that class, that's for sure. It encourages you to do outside research instead of just reading your textbook and coming to class and taking notes. It helps a lot.
 If it's possible to take writing-intensive classes in your major that would be good, because the more you write the more you understand the subject. When you write, that's how you understand. If you're just doing calculations and stuff and you just watch the teacher, it's not as effective as writing it down for yourself.

In general, comments supported a finding from Harvard's assessment projects: "[T]he relationship between the amount of writing for a course and students' level of engagement---whether engagement is measured by time spent on the course, or the intellectual challenge it presents, or students' self-reported level of interest---in it is stronger than any relationship we found between student engagement and any other course characteristic" (Light, 1992).
 Improved Problem-Solving Abilities.
 Fifty percent of the students went beyond talking about knowledge retention and reflected on how writing helped them to become better thinkers:

If you can put your ideas down, if you can take that time to transfer your ideas from the mind to the paper, you're not being a passive thinker anymore. That's what's good about writing because it changes a person from being a passive to an active thinker, being critical, being able to analyze, to be precise.
 More than help my writing, I think [writing-intensive classes] helped my thought processes---to develop my ideas, to think about them fully and then to express them on paper. I don't necessarily think of that as helping my writing but more my thinking processes, developing my ideas for that class.
 The more writing I do, the more I'm forced to think about the subject matter. And the more I'm forced to think and the more I'm forced to exercise my brain, I feel like my thoughts are a lot more cohesive and coherent.
 I think people write to help them develop a really clear idea. I think of writing as a process, like a way of thinking. Because when I start writing something, I don't have a clear idea of what I'm going to end up with. I mean I have a general idea but I never specifically know what I'm going to come up with until the end.... It becomes even more clear to me and then I have to write a second draft because you know my first one isn't good enough. But I think writing is like a process, like a thinking process, and you discover all these really neat things while you're in the midst of writing a paper. It's hard to explain.
 It's a challenge. For me writing isn't easy. I don't think writing is really easy to any person. Well, actually that's an overgeneralization. But the reason I write is because I think it's one way for you yourself to improve your way of thinking. I think getting things on paper, writing it, helps the thinking process.

How Do These Students See Relationships Between Wl Instruction and Courses
in Their Majors?
Writing as a Practical Job Skill.

 The interviews revealed a particularly high degree of student satisfaction with WI courses taken in their majors. The source of the satisfaction for 89% of the seniors was their perception that these classes served to prepare them for future career writing tasks:

The writing that I've been doing so far is actually a glimpse of what I'm going to be doing in the industry when I go to work for some firm. I mean everything that they'll be asking you to do, like preparing proposals or contracts or specifications for a project, is directly related to the type of writing we're doing in class. It's just the groundwork.
 I hope to become an architect and there's a lot of programs that need to be written. So the work that we're doing in school definitely relates to the profession. Actually a written program is one of the areas that you can charge extra for---it's not part of the basic service that an architect provides for the client, and that's good!
 I think [this writing] is going to help in the future because we're going to end up doing a lot of writing. I know that because I worked summers in the hospital and you do a lot of charting. And the better you get at reporting facts, the better I think your charting is going to be. You know, clear, concise, to the point. As far as taking statistics and writing them down for information and translating it into complete sentences, I'll be doing that in charting. Assessing the family and writing, yeah, I'll probably be writing assessments of individuals or groups, including that in the charting. And it's the same with the quality assurance paper I'm doing right now---I'll probably be doing that later, too. So it's useful.
 I think [the writing] helps a lot in the sense that in the business world you have to write a lot because you have to communicate. So right now the business college forces us to write a lot. So I think we're learning. I will probably be in accounting. If I became an auditor, I'll have to write reports, so I think the courses are training us to know how to write a standard report.

But there were contrary voices:

I'd like to go into medical school and I'd like to become a physician, so I think as far as writing on patient's charts and writing down diagnoses, that's where a lot of my writing is going to come in. I don't think the writing in my major is going to help me too much because I have had no practice in my writing with scientific terms and scientific papers.
 Writing helps me right now. I don't know how it will help me in the future but it helps me now to become the person that I am---like what I believe in. In the future I don't know if I'm going to keep on writing myself. If I go to graduate school maybe I'll keep on writing.

How Do Experienced Students Describe Successful WI Instructors?
 Attitudes.
 Two major categories of responses related to instructors of WI courses The first category described the characteristics of instructors who, in the eyes of the students, did the best job. Sixty-one percent of the seniors specified instructors who were perceived as encouraging, caring, and accessible.

She doesn't intimidate and she just lets you---she ives you real positive reinforcement, and she's real open. She'll criticize, but in a nice way and she'll offer [suggestions]. It's like she's on your side to help you to produce the best paper that you can.
 So many relationships with professors are real removed and you're afraid to approach them. The WI instructors were real approachable, real enthusiastic.
 He always encouraged us. He never put us down in any way or made us feel bad for writing how we wrote. And then at the end of the semester, he made us compare our last paper with our first paper and it was like this really big improvement.
 [Q: What did she do that was helpful?] I guess mostly the encouragement because even if my idea's different, she would always, "Well, write about it, and maybe I can understand what you're trying to say and see where you're coming from."

Thirty-four percent specified the value of professors who were seen as challenging, strict, and pushing for excellence. Another 33% related teacher success to an instructor's genuine concern with good writing. Finally, 22% of the responses stressed the importance of a professor's knowledge of course content.
 Teaching Strategies.
 The second category involved teaching strategies the instructors followed in their WI courses. Students frequently described strategies that made the classes work for them: strategies that improved their writing abilities, helped them learn course material better, or improved their problem-solving abilities. (A number of students mentioned more than one strategy.) The most-mentioned strategies were those that provided students with feedback: Ninety-four percent of the students emphasized the importance of getting feedback on their writing from the instructor and from their peers, feedback that would often encourage students to revise. Forty-three percent of the 82 seniors mentioned particular satisfaction with specific written feedback from the instructor; 30% talked about the role of peer feedback; and 21% highlighted teacher-student feedback conferences.

Other students read my drafts. It felt good to see good comments from peers, 'cause usually only your teacher will see it. It's good to see what your peers say about your writing. If they agree, I must be doing something right they know what I'm saying. They write comments. For me, that was really satisfying.
 The comments were pretty good on each of my papers. It was pretty extensive. For the Education class, we started out with a lot of feedback, and we started out having to do a lot of freewrites also, and I thought that was good in that not only do you get the feedback, but you get a chance to try and improve on your writing using the comments that were given to us.
 Most professors criticize whether things were clear or not clear---like if I needed to elaborate on certain things because they had no idea what I was referring to. And, you know, when they catch that, then I begin catching it more. That's really important.
 I notice a lot of other teachers don't really say much and they just slap a grade on with no comments. These teachers will really sit down with you and go through and show you how you can improve and what's wrong and everything.

Another strategy, closely related to provision of feedback, is the provision of frequent opportunities for students to hand in drafts and then revise. Indeed, only 2 students reported never having revised, while 74 commented on their revision experiences, nearly all positively.

You're always allowed to pass [drafts] back in. I love it because you do it and try it out and you can try new things without a feeling of, "Oh God, what if they don't like it? I'm going to fail." It's like I can try this and see how it goes over and if it doesn't work I can try something else. So I think that's really helpful when professors let you do that because it gives you the chance to take risks, to try a different style. Those classes are good for that.
 A lot of times I could [revise] and I found that I, like I said, I hate that. I hate having to go back to something that I've worked so hard on and re-do it again and again and again. But a lot of classes have allowed me to do that and they've encouraged that and I really think that's a good idea. If I know that I have a chance to improve my grade, then that's enough motivation sometimes to go through that painful process and make it better and in the process I'm learning from it.
 If you have a very hard teacher and they nitpick about your paper saying, "Oh, you should develop this more," I think they should also give you a chance to rewrite it. I think revision is a lot more effective in making you a better writer than just writing all these dozens of papers and you keep making the same mistakes, so T think revision is a lot more effective in that way.
 I took a marine geology class and that was pretty intensive. We didn't have a final so a lot of weight was going to be on those papers. It was due the day where our final was supposed to be so I spent a lot of time on the paper. I turned in three or four drafts to her and that was a lot of work, but when I was done with it, I could say that I was really happy with the work that I did. Any other paper that I encounter here at the University similar to that I know I'll be able to handle it and not panic and not put it off till the end. I turned in all those drafts; that-way it would get better and better every time I'd get it back. That was good.

Finally, two other positive strategies are worth noting because of frequent mention. One was the use of writing to involve students in thinking through the relevance of course content to career choice. The second was related---teacher attention to the design of relevant assignments:

He'd bring in a lot of his personal experiences because he's been in the educational system for a long time. And the things he had us write about really made us reflect on our choice of career. A lot of the papers came out sort of personal.
 They did a good job because of setting assignments into real-life application. Sometimes if it's just an exercise and it's not real, you know it's an exercise and you don't get as into it as you do if they say, "Okay, we're going to do this assignment, a letter to the parents. Just write this like it's a real one." After you're done, it's not just a paper. You might actually use what you wrote later. That's good. A lot of teachers did that.

How Do Students View Class-Size Restrictions in WI classes?
 To allow for the types of interaction that seem to promote learning and writing, registration in every WI class at the University of Hawai‘i is limited to a maximum of 20 students. Specific interview questions asked students if they saw this restriction as making any difference in the quality of their educational experience.
 Eighty-five percent of the interviewees indicated that the small class size had made a difference. Most of them saw this as critical to their obtaining the feedback that they described as so important to their learning. Fifty-three students also made the more general observation that small class size increases the amount of individualized attention a student receives. However, 15% said they had experienced no particular differences attributable to the size of their WI and their non-WI classes. Here are some typical comments:

WI classes are small. That's another thing I like. I don't like big classes. I think it's important to have maybe 12---10 to 12 students in a class and not make it so large that you won't have time to go over your writing. I like to break down into little groups and have mini-workshops. I think an important part of improving your writing is to get other people's feedback.
 I like the class restriction. I really like that because I think you can get the most out of it. The smaller the better. I know some teachers out there don't want to assign so many [papers! because they can't read them all. A size restriction gives teachers more time to look at your paper and be more critical and spend some time with the student.

What Would Students Do to Improve WI Instruction?
 While students in general reported satisfaction with WI classes, no student was equally satisfied with all WI classes. In commenting upon satisfactory elements, most students focused on one or two particular classes. The same focus on individual classes occurred when students were asked about possible improvements. Suggestions for improving instruction were made at many points in the interviews as well as in response to the question: "Do you have a suggestion for improving writing-intensive courses?" The suggestions can be grouped into four categories.
 Fuller explanations.The most frequent suggestion---made in one way or another by 61 students---was that instructors be more explicit in their explanations of WI course requirements, the purposes of specific assignments, and the purpose of the WI requirement.
However, when asked specifically about the adequacy of instructors' explanations, 26 students expressed satisfaction and another 21 felt that there was nothing to explain. Student descriptions of instructors' explanations were revealing: "Yeah, I remember her explaining it. That we need so many pages of written work or something." Many students felt that a professor's highlighting the syllabus or noting the number of writing assignments was an adequate explanation of WI classes. The contradictory responses hinge upon individual judgments of what constituted adequacy. Students who expected more than references to logistics frequently suggested that fuller explanations were in order.
 Administrative changes. Forty-three students recommended changes in the practices or policies related to WI course offerings. Suggestions included standardizing WI course requirements rather than specifying minimums (19); increasing the number of WI courses offered each semester (13); distributing WI classes more evenly across departments and across upper- and lower-division offerings (13); improving the image of WI classes, primarily by relabeling them with something less intimidating than intensive (12); and increasing the number of WI classes required for graduation (4).

I don't know how they choose the writing-intensive teachers, or if the teachers volunteer to make their classes writing-intensive, but they really have to have something standard where you have to make sure that the professor can give the students helpful feedback.
 Make all classes writing-intensive. Don't call them writing-intensive, just say they're normal classes and make a change on the department side saying, "Look, we need more work. People cannot write nowadays. We need more emphasis on writing." And just institutionalize it.
 Writing-intensive---that word. Maybe you guys can change the word. Intensive sounds like they're going to do something to you!

Feedback and revision. Given the many comments about feedback, it is no surprise that the third highest percentage of recommendations involved increasing opportunities for instructor and peer feedback. Students recommended more interaction between students and instructors so that students might receive extended feedback on their drafts. Several students suggested that revision should be required or at least made a standard option in all WI classes.
More student "ownership." The final student recommendations clustered around the idea of giving students more choices over topics, or giving students free choice of topic. Several also suggested that student authors themselves should have some say in determining appropriate lengths for their essays and reports, as well as the timeline for composition.

I think they should give you, allow you, enough time to work on your papers . . . and they should give you at least a couple choices of what you could write on, not just one.

Discussion
 What Student Perceptions Suggest About Writing-Intensive Instruction and Student Learning
 The results provide confirmation for teachers whose experience convinces them that writing has an important role in learning---a role not well documented in research. Because the study did not involve classroom observations, observations of students while they wrote, or samples of student writing, the results are not direct evidence of connections between writing-intensive instruction and students' learning, behavioral change, or attitudinal change. Nonetheless, the results do constitute evidence that most students perceive positive changes: Most interviewees reported that they were learning more or learning better in WI courses; in addition, many reported increased confidence as writers. Several theories of human behavior stress the importance of perception---along with content and context---as a component of how we perform (Bandura, 1982;Meichenbaum & Cameron, 1982). To the extent that perception reflects learned behavior or mediates subsequent behavior, our findings of perceived positive change are highly important.
The areas in which students perceive themselves to be learning are consonant with the claims of writing-to-learn advocates. First, students said they are learning course content better. Repeatedly, students said, in one or another way, "By writing about it, we are learning more about it." (Student statements about the other side of the coin are truly depressing: Too many noted how easy it was to cruise through courses that required no writing.) Second, some 78% of the students reported that they are becoming better writers. Significantly, the key factor they pointed to is not the amount of practice they got or the quantity they wrote; it is the amount of feedback that their course instructors and their peers gave their writing. Finally, students comment, in a variety of circumstances, that they are becoming better thinkers and more attentive to problem solving processes as they learn through writing.
 The students interviewed in this study suggested that the ways of learning and the extent of their learning were somewhat different in WI classes. Schumacher and Nash (1991), after demonstrating the inappropriateness of most instruments used in attempts to document learning, suggested that a primary benefit of learning through writing may be a reorganization of the concepts and processes used in critical thinking. They went on to note that the reorganization is accomplished cumulatively and is to date best measured by relatively primitive instruments currently under development by cognitive psychologists. In this study, students occasionally mentioned new confidence in themselves as potential career professionals and as writers, suggesting that new or modified schemata for behavior are resulting from WI classes. It was not uncommon for a student to refer to a specific WI class in which the new awareness occurred; interestingly, that class typically occurred later in a student's educational career or involved a particularly liked instructor. Such references hint that the cumulative effects of WI classes may be as significant as the effects of any single experience or course---or even more significant. Further, WI courses taken early and late in a student's academic career, as well as within or outside of the chosen major field, may well have different effects.
 Students' perceptions linking career advantages and WI instruction reinforce the dominant finding from the Mānoa Writing Program Faculty Board's annual review of survey forms distributed toward the end of many WI classes. The survey forms also show that student satisfaction is greatest with WI classes in their chosen majors. This is probably related to the frequency with which students in this study reported satisfaction with "relevant" writing assignments but not with "exercises." To write well may be the goal, but our students seem to tell us that writing that is directly related to their profession is even more important, and more satisfying, than mere "good writing."
 Advocates of education for its own sake may be dismayed by students' concerns with career-related relevance. A counter to such dismay may lurk in something we did not find in student statements: complaints about the quantity of WI classes required for graduation and the likelihood that students have to take at least some WI classes outside their chosen fields. And behind student testimony is another---quantitative---fact that speaks equally loudly. The seniors who were interviewed for this study had each typically taken five WI classes. To graduate, the University required that they take only three, and in the case of interviewees who were taking longer than 4 years to graduate, perhaps even fewer. The fact that students were typically taking 67% more than the required number of WI classes underlines what can be inferred from student statements: Experienced students seek out WI classes because they get more out of the classes. This occurs despite student perceptions that the classes require more work and despite a typical hesitancy of first- and second-year students to enroll in classes that require writing.
 What Experienced Students Want in WI Instruction
 Given that the typical interviewee had taken five WI classes, the number one suggestion for improving instruction is troubling. Students want more explicit statements from the University's faculty and administration about the purposes of the WI course requirement and of individual writing assignments. Many seniors were less clear than they wanted to be (or ought to be after five WI classes) about the philosophy underlying WI instruction.
 Student confusion over the specific purposes of WI instruction is probably a reflection of professors' confusion. For many---perhaps most---professors a commitment to use writing to promote student learning is a more radical departure from previous teaching methods than they expected. Several professors have told the authors of this report that they did not really understand their own responsibility to help students with the writing in their discipline until they had taught two or three WI classes---even when the professor had spent a day or two at a workshop on WI teaching methods. Why? Like their students, professors have been conditioned to think of writing instruction as the responsibility of the English department and to think of "good writing" as a unimodal construct. It is only through experience with their students-as-writers that professors come to learn that good writing in their own field is in some ways quite different from good writing in, say, physics or history. Student comments throughout the interviews suggest that students would themselves like to have clearer understandings of such differences.
 The topic of how we learn to write in the genres of different fields is a subject of debate (e.g., Freedman, 1993a, 1993b; Williams & Colomb, 1993). For the moment, teachers may be guided by attending to how their students navigate new waters; teachers can also benefit from recollecting their own experiences as they entered a community of specialists, a community that was once quite foreign. In the future, we can suggest more research to identify what students need to know. What we have learned from this study is that students want more explicit discussion of issues related to the functions and methods of writing in different disciplines.
 Finally, student responses reported here suggest that students also want their professors to act toward students in specific ways. The recent literature on teaching with writing frequently uses two metaphors to describe teachers: coaches, and masters to their apprentices. Student comments summarized above make clear that students value both. They want their teachers to explain the special characteristics of writing in their fields and to help them to become good writers. They also want teachers to be coaches far more than they want referees. When teachers do become referees, students want to understand the reasoning behind the calls. They say they want their teachers to push, prod, and reinforce. Ultimately, students want a jury not only of peers, but also of encouraging experts.
 More than anything else, students in WI classes value their professors' interaction with them on their drafts, in person, or in writing. With regard to instruction, the most frequent complaint involved WI classes in which students sensed a lack of direct teacher attention to their writing---a lack of involvement and assistance. Students say they learn, and are often willing to do extra work, when a professor takes them seriously and takes their written efforts seriously. They say they thrive in environments that promote extensive interactions over written texts. They are willing to learn from one another. But ultimately, students most appreciate one-on-one learning with their instructors.
 If one-on-one involvement is what is necessary for optimal learning, research universities may have to find ways to support research that does not require professors to subtract from the time devoted to teaching. For writing-based instruction to succeed in reforming higher education, universities will need to return the spotlight to the professor's instructional role and applaud good instruction with real rewards---a point to which we will return below.
What These Data Suggest About the Writing-Across-The-Curriculum Movement
 Although when we began this study we had hoped to hear some good things about the University's WI classes, we had no expectation of hearing so many positive reports from so many of the students. What the students said was compelling. And no one, not even the most ardent advocate of unit-mastery and multiple-choice approaches to education, can help but be somewhat swayed by their comments, only a portion of which are included in this report. The interweaving of comments on what they learned about thinking more clearly, about satisfaction with professors, and on growing self-confidence are a welcome change from the more often dismal reports about the shortcomings of American higher education.
 Let us for the moment assume that through instruction in a WI mode such as University of Hawai‘i has established, students will learn more during their university years. Let us further assume that society, therefore, should invest in this mode of instruction. What would that require? First, a willingness by students, teachers, and administrators to seize the moment and do whatever is necessary to change attitudes, behaviors, and budgets in supportive ways.
 Second, we would need teacher-researcher collaborations to an ucation. We need solid reseextent perhaps never before seen in higher edarch on the ways that writing functions in the larger enterprise of learning. We do not know just what it is that leads students to perceive better learning from these classes. If the perceptions reflect reality, we do not know what is critical to that reality. The interviews suggested several possible factors: teacher and peer feedback on drafts; opportunities for students to revise; peer and student-teacher interactions (concerning students' writing, but then again perhaps it is only the interaction); increased frequency of writing; "relevant" assignments; teacher attitudes, real or perceived; and small class size with its attendant possibilities and realities. To expect that researchers in composition and education alone will figure out and fine tune the writing-to-learn complex is foolhardy. The effort will require that classroom instructors act as researchers alongside trained educational researchers. It is fashionable to talk about the desirability of professors applying their "researcher's eye" to their teaching. If we want WI instruction to be provided with the effectiveness that only research and feedback can ensure, we will have to make such fashionable talk into everyday action.
 And third, universities would have to renew their support for teaching. Many professors see teaching-with-writing as teaching-with-added-labor. Most express willingness to undertake extra work if circumstances allow them to do it successfully along with other responsibilities. The types of learning situations that our students have described as most helpful require environments and teaching schedules that allow extensive interaction. Reward systems that favor research and publication over effective teaching will not assure faculty support for labor-intensive interaction with students.
 Why have we highlighted these three needs? First, writing-intensive classes are not rafts that can be expected to carry students to brighter shores simply by virtue of being launched (see White, 1990). To be effective, writing-intensive classes require administrative and programmatic support akin to what is provided to fleets of ocean liners---support in the form of instructor training, reasonable class sizes, faculty governance, meaningful assessment, and a secure budget line. (For a full discussion of the kinds of support necessary to make a writing program work, see Hilgers & Marsella, 1992). And historically, reform efforts in education often do not survive long enough to benefit students learning unless there is a concerted effort to institutionalize carefully planned and monitored changes in a timely fashion.
 The lesson of experience may be especially applicable now that American higher education is once again a part of a larger dialogue about restructuring American educational institutions in the face of changing economic and political realities. The data reported here bring student voices to that dialogue. These voices, like all data composed of consumers' perceptions, suggest directions that might not be evident in other forms of assessment or in other conversations about writing instruction. It is up to those who seek to reform higher education to integrate what is learned from student perceptions with what is learned from other forms of research for the improvement of student learning and student writing.

Authors' Note: Preparation of this report was supported in part by the University of Hawai‘i Office of Planning and Policy, Dr. Colleen Sathre, Director.
 References
Ackerman, J. (1993). The promise of writing to learn. Written Communication, 10, 33-370.
Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37, 122-147.
 Berg, B. L. (1989). Qualitative research methods for the social sciences. Boston: Allyn & Bacon.
 Charmaz, K. (1991). Translating graduate qualitative methods into undergraduate teaching: Intensive interviewing as a case example. Teaching Sociology, 19, 38-395.
Doll, P. A., & Jacobs, K. W. (1988). The exit interview
for graduating seniors. Teaching of Psychology, 15, 213-214.
Drost, D. A., O'Brien, F. P., & Marsh, S. (1987, February).
Exit interviews: Master the possibilities: Putting exit interview data
to better use. Personnel Administrator, 104-110.
 Embrey, W. R., Mondy, R. W., & Noe, R. M. (1979,
May). Exit interview: A tool for personnel development. Personnel Administrator,
43-48.
 Freedman, A. (1993a). Show and tell? The role of
explicit teaching in the learning of new genres. Research in the Teaching
of English, 27, 222-251.
 Freedman, A. (1993b). Situating genre: A rejoinder.
Research in the Teaching of English, 27, 272-281.
 Garretson, P, & Teel, K. S. (1982, July-August).
The exit interview: Effective tool or meaningless gesture? Personnel,
70-77.
 Guba, E. G., & Lincoln, Y. S. (1989). Fourth
generation evaluation. Newbury Park, CA: Sage.
 Herrington, A., & Moran, C. (1992). Writing
in the disciplines: A prospect. In A. Herrington & C. Moran (Eds.),
Writing, teaching, & learning in the disciplines (pp. 231-244).
New York: Modern Language Association.
Hilgers, T., Bayer, A., Chinn, P, & Hussey,
E. (1993). Perspectives on writing and learning. Honolulu: University
of Hawai‘i at Mānoa, Mānoa Writing Program Faculty Board.
 Hilgers, T., Cole, R., Tsutsui, J., & Wallace,
A. (1991). Writing-intensive activities in classes across the curriculum.
Honolulu: University of Hawai‘i at Mānoa, Mānoa Writing Program Faculty
Board.
 Hilgers, T., & Marsella, J. (1992). Making
your writing program work: A guide to good practices. Newbury Park,
CA: Sage.
 Institutional Research Office. (1992). Distribution
of grades, credits earned ratios and course completion ratios, University
of Hawai‘i at Mānoa, Fall 1991. Honolulu: University of Hawai‘i at
Mānoa.
 Kinneavy, J. (1987). Writing across the curriculum.
In G. Tate (Ed.), Teaching composition: 12 bibliographical essays
(pp. 353-377). Fort Worth, TX: Texas Christian University Press.
 Lee, R. D. Jr. (1991). The use of exit interviews
in master's programs of public affairs and administration. American
Review of Public Administration, 21(3), 183.l
 Light, R. J. (1990). Explorations with students
and faculty about teaching, learning and student life. The Harvard Assessment
Seminars: First Report. Cambridge, MA: Harvard University Graduate
School of Education and the John F. Kennedy School of Government.
 Light, R. J. (1992a). Explorations with students
and faculty about teaching, learning and student life. The Harvard Assessment
Seminars: Second Report. Cambridge, MA: Harvard University Graduate
School of Education and the John F. Kennedy School of Government.
 Light, R. J. (1992b, June). Five years of assessment
at Harvard: How findings are changing teaching and learning. Paper
presented at the Seventh American Association of Higher Education Conference
on Assessment in Higher Learning Miami Beach, FL .
 Light, R. J., Singer, J. D., & Willett, J. B.
(1990). By design: Planning research on higher education. Cambridge,
MA: Harvard University Press.
 Lofland, J., & Lofland, L. H. (1984). Analyzing
social settings: A guide to qualitative observation and analysis (2nd
ed.). Belmont, CA: Wadsworth.
 Marsella, J., Hilgers, T., & McLaren, C. (1992).
How students handle writing assignments: A study of eighteen responses
in six disciplines. In A. Herrington & C. Moran (Eds.), Writing,
teaching and learning in the disciplines (pp. 174-188). New York: Modern
Language Association.
 McCracken, G. (1988). The long interview.
Beverly Hills, CA: Sage.
 McLeod, S. H. (1992). Writing across the curriculum:
An introduction. In S. H. McLeod, & M. Soven (Eds.), Writing across
the curriculum: A guide to developing programs (pp. 1 11). Newbury
Park, CA: Sage.
 Meichenbaum, D., & Cameron, R. (1982). Cognitive
behavior therapy. In G. T. Wilson & C. M. Franks (Eds.), Contemporary
behavior therapy: Conceptual and empirical foundations (pp. 310-338).
New York: Guilford.
 Mishler, E. (1979). Meaning in context: Is there
any other kind? Harvard Educational Review, 49,1-19.
 Mostyn, B. (1985). The content analysis of qualitative
research data: A dynamic approach. In M. Brenner, J. Brown, & D. Cantor
(Eds.), The research interview: Uses and approaches (pp. 115-145).
New York: Academic Press.
 Nelson-Gray, R. O., Haas, J. R., Romano, B., Herbert,
J. D., & Herbert, D. L. (1989). Effects of open-ended versus close-ended
questions on interviewees' problem related statements. Perception and
Motor Skills, 69, 903-911.
Richardson, S. A., Dohrenwend, B. S., & Klein,
D. (1965). Interviewing: Its forms and functions. New York: Basic
Books.
 Schmidt, W. (1984). The exit interview as a monitor
for change: A review of the literature. Paper presented at the Annual
Meeting of the Southwest Regional Conference of the American Business Communication
Association, San Antonio, TX.
 Schumacher, G. M., & Nash, J. G. (1991). Conceptualizing
and measuring knowledge change due to writing. Research in the Teaching
of English, 25, 67-96.
 Sommerlund, B., & Kristensen, O. S. (1989).
Textbase Alpha [Computer program]. Denmark: University of Aarhus, Center
for Qualitative Research.
 Street, S. (1991). Appalachian college assessment
consortium. In R Hutchings, T. Marchese, & B. Wright (Eds.), Using
assessment to strengthen general education (pp. 30-31). Washington,
DC: American Association for Higher Education.
 Stotsky, S. (1992). From the Editor. Research
in the Teaching of English, 26, 245-248.
 White, E. M. (1990). The damage of innovations set
adrift. AAHE Bulletin, 43(3), 3-5.
 Williams, J. M., & Colomb, G. G. (1993). The
case for explicit teaching: Why what you don't know won't help you. Research
in the Teaching of English, 27, 252-264.
APPENDIX:
"Exit Interview" Questions on Seniors' Experiences in University
of Hawai‘i Writing-Intensive Classes

  1. Why did you want to go to University of Hawai‘i?---Did anyone influence you?
    ---Did you apply to mainland schools?
    ---Did you apply to other schools on the islands?
  2. What was your best or most satisfying experience here at
    UH?
  3. What majors did you consider?---What happened to make you choose your major?
  4. Please list characteristics of good writing.
  5. Why do people write?---Why do you write?
  6. How did you feel about writing before you came to UH?---Can you give me an example of writing you did before
    UH?
  7. Now how do you feel about writing?---Was there a paper, teacher, or class that influenced
    your feelings/changed your mind?
  8. What do you like most about writing?---What do you think your strengths are in writing?
  9. What do you like the least about writing?---What do you think your weaknesses are in writing?
  10. What has been your best experience with writing or your most
    challenging or most satisfying?
  11. What has been you worst experience with writing or your most
    frustrating or most stressful?
  12. How much writing do you do in your major?
  13. Could you describe the writing you do in your major?---Was this writing helpful? How?/Why not?
  14. What was the purpose of these writing assignments?
  15. How will the writing you do in your major help you in the
    future?---How will the writing you do in your major help you
    in your career?
    ---What percentage of your job/career do you think will
    be writing?
  16. What writing-intensive classes have you taken?
  17. Why did you choose these classes?
  18. What do you think the general attitude toward writing-intensive
    classes is?
  19. Why do you think UH requires writing-intensive classes?
  20. What is the difference between writing-intensive and non-writing-intensive
    classes?
  21. How many writing-intensive classes should be required?---How many should be required in your major?
  22. Do you think writing-intensive classes improved your writing?---Which courses helped your writing? Why?
  23. What have you gotten out of writing-intensive classes?
  24. In what class did the writing-intensive instructor do the
    best job?---What did he/she do that was helpful?
    ---How was that helpful?
  25. Do you think your writing-intensive instructors explained
    writing-intensive's goals to the class?
  26. What do you like them to explain at the beginning of writing-intensive
    classes?
  27. When you have trouble with your writing, where do you go
    for help?---Anyone else? Family, friends, people in your class,
    teachers, writing workshops.
  28. What comments did the writing-intensive instructors give
    you on your writing?---What comments were most helpful?
  29. Could you make changes on your paper and turn it back in
    to the instructor?---Could you revise your papers?
  30. Can you describe the writing assignments in your writing-intensive
    courses?---What were the writing assignments like?
  31. Can you pick one writing assignment that sticks out in your
    mind?---Why did you choose this one? What makes it special?
  32. Did you learn more in the writing-intensive courses?---Did you learn more about the course subject?
    ---Did you gain a different kind of knowledge in the
    writing-intensive classes?
  33. What did you think about the number of students in your writing-intensive
    courses?---Compared to your non-writing-intensive courses, what
    did you think about the number of students?
  34. Do you have a suggestion for improving writing-intensive
    courses?
  35. You're graduating soon; do you think there is something special
    that has enabled you to graduate?---Is there something or someone that you can attribute
    your graduation to?
  36. These are all the questions I have. Is there anything you'd
    like to add or go back to and talk about more?