Alternatives to Lab Reports?

Posted by Chad Orzel on May 5, 2011
(15)
More »

An academic email list that I’m on has started a discussion of lab writing, pointing out that students in some lab classes spend more time on writing lab reports in a quasi-journal-article format format than they do taking and analyzing data. This “feels ” wrong in many ways, and the person who kicked off the discussion did so by asking for alternatives to the journal-article style lab report.

This is a recurring discussion in physics education, because everybody who teaches lab courses struggles with this issue (guess what I’m procrastinating from grading right now…). It’s made much worse by the fact that many students, particularly in intro courses, write so badly that it’s painful to read their labs. Even more annoying are the students who clearly could do better, but have chosen not to for some reason (I get really annoyed when students email me lab reports that are riddled with spelling errors, because I know that they can see the exact same squiggly red underlines that I do when I open their reports).

The problem is, technical writing is an important skill for a scientist, and something that science majors need to learn. Which means we have to teach it in some way, and most of the alternatives I’ve seen proposed don’t serve that purpose– oral exams are a great way to tell whether a student understands the lab, and lab-specific questionnaires are an okay way of getting at that question, but they don’t teach writing. Blog posts or non-technical write-ups involve some writing, but it’s not technical writing. And anybody who thinks it’s easier for students to write pop-science style articles (Scientific American is the usual suggested template) has never tried to write good pop-science articles. It’s really hard to do well, in some ways harder to do well than technical writing.

The best compromise cop-out seems to be having students write pieces of lab reports– just an Abstract, say, or just the Conclusion. Which sorta-kinda teaches technical writing, and does at least produce less volume of material to grade, but it’s not ideal by any stretch. Again, writing disembodied sections of a report can feel harder than writing the full report would be, particularly if it’s something like an Abstract or Conclusion, which require you to know what you would have put in all the sections you’re not writing.

So it’s a Hard Problem, and I don’t have a good solution. Anybody out there with brilliant ideas along these lines should feel free to leave them (or pointers to them) in the comments, though.

(15)
More »

Comments

  1. #1 Steve
    May 5, 2011

    I teach a two semester sequence of chemistry lab classes. The first course has 60 students, and the second one has only 15 since only our BS degree seeking majors have to take it. In the first class, I have the students write only part of a paper (either the Introduction, the Experimental, or the Discussion), and of course do the data analysis. In lieu of a final exam, they take one of those lab reports and turn it into a full journal article.

    In the second class, they graduate to full journal article style lab reports for every experiment. It works all right, in that many of the students who make it through both classes have noticeably improved writing skills. I’ve found that, no matter how much writing students are asked to do, they always complain that it’s too much.

    That thing about students not proofreading and turning in results with glaring spelling errors drives me absolutely insane, especially by this point in the year. And I’m about ready to put a line in next year’s syllabus stating that anyone using the word ‘utilize’ in the Experimental section of a lab report automatically fails the class and gets deported to the Business school.

  2. #2 Pam
    May 5, 2011

    students in some lab classes spend more time on writing lab reports in a quasi-journal-article format format than they do taking and analyzing data. This “feels ” wrong in many ways,

    I don’t know, that sounds about right to me. In my experience doing the experiment (collecting the data) is always the least time-consuming part. At this point in time, data analysis takes longer than writing up the results, but that is because (1) I do more complicated data analysis, and (2) the reports are for internal (to my company) communication, not for journals. Back in my academic days, writing a paper took FOREVER, especially when the submission-and-review process is factored in!

    Regarding “alternatives” to proper lab reports, I agree with you that there is no better way of learning technical writing than DOING technical writing. A lot of science & engineering students do not appreciate that being able to communicate your results is a crucial element of experimental work, whether you are in academia or industry. No alternative to writing reports will be as effective at teaching how to write reports as practicing doing just that.

    Now, I *can* imagine situations where teaching technical writing is not a main goal of the lab course. I’m thinking of general-ed or introductory courses where the labs primarily function as demonstrations of principles discussed in class. In those cases, providing the students with template to fill out as they do the experiment is, I think, more effective (and less painful all around) than making them write a report from scratch. But I don’t think that sort of thing is appropriate for higher-level students who are (presumably) interested in pursuing science or engineering as a career.

  3. #3 Miss MSE
    May 5, 2011

    As Steve said, no matter how much writing it is, there will be complaints about it being too much. In one lab course, we were required to write full lab reports, but only one section would be graded for any particular lab. We didn’t know which section until after we’d turned it in. This allowed to professor to give us more detailed feedback, but also forced us to practice writing more. Over the semester, every section type was graded at least once, and the final lab was graded in its entirety. Even though it was a lot of writing, the course required a lot of qualitative analysis, so I didn’t particularly mind it.

  4. #4 Sherri
    May 5, 2011

    I’ve never seen any workable way of teaching writing that didn’t involve writing, and lots of it. One of the deficits I’m seeing in my daughter’s high school education is a greater emphasis on timed, in-class writing, rather than essays. It feels like this is focused more on test prep; these kids will face this kind of writing not only on state high school exams, but on the SAT and AP tests as well. This kind of short-form basically unedited writing doesn’t lend itself well to learning longer-form, edited writing, in my opinion.

    What’s the ratio of time you spend collecting and analyzing data to time spend writing about it? Is it wildly different than what you’re requiring from students? Or is it different because the students are better at collecting and analyzing than writing?

  5. #5 TheBrummell
    May 5, 2011

    I’ve met this discussion before, and I agree that it’s a Hard Problem without any obvious easy solutions (kinda the definition of Hard Problem, I guess).

    One partial explanation, though by no means excuse, for the annoying simple errors that show a lack of proofreading & editing, is the student’s sometimes rational decision-making about how they spend their time. For an experienced scientist, such as the professor teaching the course, or the graduate student TA grading the midterms, proofreading is an integrated part of writing, the only good writing is re-writing, and these things take no additional time as they are part of the process of composing sentences and paragraphs. For the inexperienced undergraduate student, proofreading represents a chunk of time that could potentially be better spent on something else; “better” in this case doesn’t necessarily mean “brings one closer to long-term goals”, I think it often means “more fun”. There’s a very clear point of diminishing returns on the graph of effort vs. reward; why spend another hour on this paper when it’s only worth 10% of one class out of five you’re taking this semester, and what you’ve got done so far is almost certainly good enough for a 70% grade? Most of the time, another pass with your tired eyes isn’t going to boost that mark by much, if at all.

    Changing how marks are allocated, to focus more on writing and less on regurgitating facts is one way to directly address this time-management decision-making by the students. From what I’ve seen, mostly as a TA in some classes with various levels of emphasis on scientific writing, it doesn’t take long to get most of the students to a “decent” level of writing skill; really great writing is obviously something that requires years of practice, like any skill, but eliminating the majority of the annoying little things in the majority of the students takes about 3 to 4 months, in my (limited) experience.

    Threats work, too – perhaps a toned down, easier to take seriously version of your fail-and-deport threat would be useful. When I gave a class presentation about how I’d marked their first papers and what I wanted to see on their second paper (and where I was going to concentrate the marks), the improvement was dramatic. Most of the lowest marks on the first paper were largely due to stupid little mistakes, spelling errors, grammatical weirdness, ugly and poorly described figures, etc., and those were virtually eliminated when I said “I’m going to really hammer those things, if you riddle your paper with spelling mistakes the maximum mark you can get will be not a good mark”. Communication is, once again, key.

  6. #6 Rhett
    May 5, 2011

    Yes, this is a tough problem. I think in a lab, I can focus on two of the following (maybe three):

    • Lab skills
    • Experimental design
    • Modeling
    • Physics concepts
    • Error analysis
    • Lab report writing

    I can’t decide which of these are the most important to focus on (in fact I change each semester). Ideally, they would all get some focus. For different labs, it is also different. If it is a lab for non-science majors or physics majors, I will emphasize different things. Typically, for the non-physics and chemistry students, I like to emphasize the the error analysis since they will probably never see it in another course.

    But I guess the discussion here assumes that you will focus on lab reports. I do have something to say about this. A long time ago, I worked on getting students to review other students’ lab reports. It really had some nice promise. Basically, students would write a report and then several other students would evaluate it.

    It was funny that students will write more carefully when they know that other students are going to read it. Also, when they see mistakes on other students’s papers, it helps them understand why it is a mistake.

    Perhaps what stopped me was technical difficulties. At the time, students were handing in stuff on paper and I would copy it without names. I used to also do this with WebAssign, but it was still a little difficult. I might bring this idea back alive. Maybe if students just post stuff online and other students evaluate it, it might work.

  7. #7 Alan
    May 5, 2011

    When I was in my first year of University, we had a lab instructor who made very clear the importance of being able to write clear, accurate and grammatically correct reports. One of his methods was to give us a journal paper and analyse how it was written. It was as if we were doing the peer review of the paper. It was a mental struggle at first and very efective. When we got to the point of recognising ways the article could have been better, we knew we had learnt something.

    I still think that analysing good writing is a good way to learn to write well.

  8. #8 John Novak
    May 5, 2011

    I think this is going to depend on what type of class you’re teaching, and what sorts of careers your students are likely to end up with.

    If you’re teaching upper level physics labs for predominantly physics students, then it makes perfect sense to concentrate on journal-style lab reports. If you’re teaching (say) sophomore level physics classes that also serve (say) various flavors of engineers plus maybe general requirements for chemists, biologists, etc, then focusing on the journal-style aspect may be a bad joke for several reasons:

    First, your engineering students will just hate you, because engineering industry reports and engineering academic styles differ from physics styles, so they’re just not applicable. Second, even your young physicists may not be getting much out of it– it doesn’t matter how well they communicate, if they have nothing worth saying due to lack of fundamental physics skills.

    Maybe you’re thinking of something different than I am when you say journal-style, but for a lower level class I would be focusing on lab reports that sneakily reinforce the primary skill of physics rather than the secondary skill of technical writing. (Note, I do agree, absolutely, that being able to communicate is critically important. I just think it’s difficult to focus on communicating when you barely have a handle on the primary material you’re trying to communicate.)

    For upper level students, of course, I say just flog them harder. The beating will continue under morale improves.

  9. #9 Dr M
    May 6, 2011

    Rhett picks up an important thread here. This was one of my main points of criticism when I was an undergraduate taking lab classes myself, and also later as a grad student when I taught lab classes. As Rhett says, a particular lab exercise may serve any of the purposes he lists. For practical and/or pedagogical reasons any given lab exercise can only serve one or two of those goals well. Most ill-designed labs I’ve encountered have as their basic problem that the person who designed the lab did not have a clear idea of what purpose to concentrate on, or tried to cover all of them resulting on neither purpose being served very well.

    So the first thing to do is to let go of the idea that every lab should always result in a quasi-article-style report to be graded. If the purpose of a lab is to drive a physical concept home, skip the report writing, or make it very simple. If the purpose is to teach report-writing, concentrate on that and make sure the students understand that this is the purpose and it is expected that it take some time, and that yes, it is quite all right to ask questions about how to go about a particular piece of writing.

    The other thing is that, at least where I did my undergrad, labs where always part of a larger course teaching some field of physics (electromagnetism, say). Passing the lab classes (handing in reports which were graded as either pass or fail) was required for completion of the course, but the final grade was entirely decided by the final exam. This way, the labs became hurdles to jump, and nothing more. Putting in an extra effort didn’t pay off. The conclusion here is to make lab classes mean something. Don’t just tack them onto a course: make them a course! Give students incentive to work, and get them to feel they are actually learning a necessary skill.

  10. #10 Bruce Fowler
    May 6, 2011

    Back when I was an undergraduate, when television was monochrome and dinosaurs roamed the planet, I did spend more time writing a lab report than doing the lab and analyzing the data. And this was back before either personal computers or electronic calculators. But the amount of times decreases with experience. By the end of the semester my time-to-write would have decreased by about 0.5 or so. With a new semester, new course, the time would go up again because of differences in report format. There was also variation in discipline. Biology lab reports were almost trivial and chemistry reports were more difficult than physics ones. But all had a decreasing time demand if one was learning.

    Incidentally, the same applies to writing drafts of papers (presentations.) When you start doing so it takes a lot of time but after a few get published and you learn the mechanics, the time decreases. Or it did for me.

  11. #11 CCPhysicist
    May 6, 2011

    (I get really annoyed when students email me lab reports that are riddled with spelling errors, because I know that they can see the exact same squiggly red underlines that I do when I open their reports)

    Are you annoyed with yourself? You should be.

    The reason they turn in those reports is that you don’t take off one point for reach red squiggly underline you see, even if it means giving them the negative total score for the report that they earned. (You don’t even have to read for content if the net score will be negative no matter what they “wrote”.)

    Rhett’s suggestion is one that I am thinking about implementing. You have to do some extra work up front to design a grading rubric that the students use when doing a blind review (and dedicate part of a lab period to this), but I am told it can be very effective.

    PS –
    I’d like to see a poll on what is most important in the lab. Or maybe two or three polls, one for non majors, one for the first majors class lab (generally mechanics), and one for the second one (generally circuits).

  12. #12 djlactin
    May 6, 2011

    tl;dr, but
    How? Somebody needs to teach them how to do it! Expecting novices to produce a coherent lab report/scientific paper without instruction is ludicrous. You MUST teach them what is required. I teach a course: “How to write a scientific paper” in which I specifically demostrate (with examples and exercises) how to plan and execute the process. [I wish that I had had such instruction during my scientific writing development. I had to learn the hard way: by fucking up.] I ask each reader of this blog: “Did you KNOW what was expected when you began to write?” Then I ask: “How could you have known without guidance?” Expecting perfection without instruction is unacceptable. Teach!

  13. #13 djlactin
    May 6, 2011

    tl;dr, but
    How? Somebody needs to teach them how to do it! Expecting novices to produce a coherent lab report/scientific paper without instruction is ludicrous. You MUST teach them what is required. I teach a course: “How to write a scientific paper” in which I specifically demostrate (with examples and exercises) how to plan and execute the process. [I wish that I had had such instruction during my scientific writing development. I had to learn the hard way: by fucking up.] I ask each reader of this blog: “Did you KNOW what was expected when you began to write?” Then I ask: “How could you have known without guidance?” Expecting perfection without instruction is unacceptable. Teach!

  14. #14 MRW
    May 6, 2011

    CCPhysicist,
    “The reason they turn in those reports is that you don’t take off one point for reach red squiggly underline you see, even if it means giving them the negative total score for the report that they earned.”

    I do take off at least a point for every spelling or grammar error, and I still get far too many mistakes that I know their word processor must have underlined.

    Chad,
    A couple ideas:
    1) We’ve gone to a system this semester where some of the lab reports are close to full article format, and others are shortened to basically plotting the data, making a few comments, and answering a few questions. Even on the nearly full article format, we usually include a few questions at the end. Ideally, they’re the sorts of things that *could* be addressed in the discussion section, but…
    2) A colleague at a different university has another approach to the same course. He has, I think, two full lab reports in the semester, and he has the students revise those two reports multiple times. I’m not sure how he evaluates the other labs, but I gathered that it was something very short.

  15. #15 Gerry
    May 11, 2011

    Hi Chad,

    I am planning the assessment for my own introductory physics class and struggling to create something new after years of issues. Our institution makes use of “laboratory report books” where students fill in the blanks as they work through lab activities. I do think this is a fairly good middle ground: it introduces students to the techniques that one will encounter in the act of doing science, but requires much less writing of them than the creation of journal-style laboratory reports, which I’m okay with: for introductory physics, at least, my goal is to meet the course objectives, all of which relate to physics, and none of which deal with writing skills.

    That said, this system has its own problems, especially when the context requires that students work in pairs or groups, because the simplicity of these lab reports makes it impossible to distinguish between one student’s work and another. I’ve encountered situations where it was obvious that one student in a group was not participating in the activities or in the data analysis, but when it came time to submit the lab, all of the “blanks” were filled. Given that our institution’s lab books haven’t been altered in years, students from previous classes will often give completed lab books to future students, who can then simply copy the contents of the “blanks.”

    Obviously, lab books work for some students but don’t allow me to discriminate finely enough between students, so here’s my solution for this semester, which will surely undergo some revision: I am asking my students to write an informal, personal journal about the lab activity and their learning. After each lab, students are asked to write, not in the style of an academic journal, but rather, in the style of a blog, personal journal, or diary. This way, I get writing from individual students that is easily identified, and I can assess the students understanding of the key topics and techniques by guiding their writing, asking them to make mention of these type of things. This guidance is necessary: it might stifle a little of students’ creativity, but it ensures that we see certain points in the writing and that we can assess it in terms of the lab activity.

    That does leave the issue of assessing hands-on skills, ie. the use of equipment, which is obviously a key part of the inclusion of laboratory activities in a program of study, and necessitates a second assessment piece. I have created a rubric which assigns a grade to students based on their proficiency in the laboratory setting (ie. a range of levels of observed student performance from perfectly flawless execution of data collection techniques to an almost complete lack of ability with regard to specific techniques). I observe the students using equipment and doing data analysis as I wander through the lab, and make notes on specific instances I observe. At the end of the semester, I will take the notes I’ve gathered, and based on the rubric, assign a grade which will be a part of the final laboratory grade.

    This is a new approach for me, but it’s met with approval from the students and I’m pretty excited about it because it gets us out of a rut I’ve observed with time in the laboratory.