The fact: my students write badly. The source of the problem is, most likely, low quality primary and secondary education, with little emphasis on essays.

Long term solutions are beyond the scope of this question. My focus is on the short term, as a teacher of a course that asks for essays and reports. How can I help my students to improve their writing skills, even if just a little, during the course of a semester?

A potentially relevant fact: there is no awareness among students (and I would even say, among teachers) that good writing is important for their future.

Other context: I'm in a business department. My class has an average of 20 students. They come from high school. I'm new to the university so just getting to know the type of student.

This question (or its answers) seems focused on graduate students, with many examples from lab work or theses, which is not my case. This question is about us, academics. The focus here is on undergraduates who will not develop an academic career, but require such skill nonetheless, and unfortunately my university is not providing it, at least not systematically.

  • 3
    'there is no awareness among students... that good writing is important for their future' Has your institution's pastoral care or employability support team presented the students with the evidence that good writing is important for their future? Are you even sure it's true? Commented Dec 7, 2020 at 16:57
  • 2
    @luchonacho I mean, are you at a CC? Are most of your students directly from high school, where they performed well? Or, are there a lot of transfers? Do they mostly speak English (or the local language) as a first language? STEM heavy? SLAC? Commented Dec 7, 2020 at 17:10
  • 2
    @SamuelRussell I see exactly the same issue with physics - our students don't have as many essays, but they still have to be able to write clearly and well for essay-type exam questions and especially lab reports. As with business, my time in industry (after my Masters, before my PhD) demonstrated that good writing skills as an engineer were very useful.
    – Chris H
    Commented Dec 8, 2020 at 9:27
  • 2
    @ChrisH I'm also in the UK, and what I had in mind is the difference between a student who doesn't have (a good grade in) GCSE English language, and a student who does have a very good grade in GCSE English language, but has forgotten (e.g. through lack of practice during two years of a pure STEMM set of A levels) the skills listed in assessment objectives AO5 and AO6 of the subject content and assessment objectives for that qualification. Commented Dec 8, 2020 at 11:33
  • 2
    @DanielHatton I suppose a reasonable passing grade in GCSE all we can expect from our students, but getting a level from that description is hard: most of AO5 and all of AO6 are applicable in junior school (I have a child in year 3, and similar wording is used) and are requirements for the rest of life. The distinction you make is clearer. My A levels were some years ago, so things have changed, but we were still assessed on our ability to communicate ideas in writing in physics and computer science, less so in maths. My written communication also benefited from studying German to A-level
    – Chris H
    Commented Dec 8, 2020 at 13:03

11 Answers 11


Several thoughts.

  • Give writing assignments (goes without saying). Probably more short ones rather than fewer long ones.
  • Make sure students know that form matters as well as content. To encourage students who start from a disadvantage, count improvement as well as mastery.
  • Collect and critique first drafts before grading the final work.
  • Allow students to resubmit essays.
  • More problematical: have students critique each others' work before submission. Those critiques could even be private, so that students wouldn't be calling each other out before the teacher, only noting when some argument was unclear to them.

Of course this will be time consuming for you and for them. To do it properly you may have to revise the course objectives to include work done to improve writing - at the expense of "more content".

  • 2
    Peer reviewing may introduce psychological issues (privacy, competition, personal feelings ...). That said, it's common and students may be used to it. Just be aware of potential problems. Peer review need not attempt to deal with standards, just with clarity. That will go a long way. Commented Dec 7, 2020 at 15:32
  • 3
    For peer reviews, I have seen a tool from UCLA cpr.molsci.ucla.edu used in a math class. I'm not sure if student peer review from students who cna't write is worth the effort, but if you want that, then cpr solved some logistical problems. Commented Dec 7, 2020 at 15:45
  • 4
    @DimitriVulis it's often easier to spot the flaws in others' writing, so peer review should work well. The process is instructive for the reviewer as well as the reviewee. It won't catch everything; I'm sure some errors are very common, but I do find that when students take up my recommendation to peer-proofread they do better than those who don't (OK, correlation!=causation, selection bias, confidence, etc.). This even extends to students whose first report is badly written and respond to feedback for the second, but of course they have an incentive in the form of many important marks
    – Chris H
    Commented Dec 8, 2020 at 9:36
  • 1
    I'm not sure I agree with the "start from a disadvantage" advice, but that's perhaps because I don't understand exactly what the proposed advice is. It seems to be a grading advice, but students have known for ages how to game such grading. Offering more flexibility with deadlines, extra coaching or other similar help could be a better form of compensation.
    – MSalters
    Commented Dec 8, 2020 at 12:05
  • 3
    An anecdote about what can be problematic about non-blind peer-reviewing: I once had a class where Student A did not want to say anything negative about Student B because A worked at the company owned by B's father (I found this out while chatting with the A after the semester was over). So, there can unexpected conflicts of interest.
    – Aditya
    Commented Dec 9, 2020 at 5:14

This will require effort on your part, but it's a system that has systematically improved my writing from a D grade on average to an A grade. Often times I see professors engage in 'feedback overload' with each assignment, and your target student will typically take that feedback, understandably not know what to do with it exactly, and continue writing poorly.

Much of your work starts at the beginning of the semester if you want to improve writing in your students. When you grade papers, you need to be looking for specific systematic flaws that persist in an individual s writing --- if you're lucky, enough students make the same mistake and you can address it all at once to the class in a single document.

For example, suppose you have a student that abuses non-specific language. This might take the form of a paragraph like:

Big yellow dogs are preferable to big red dogs because big red dogs are lucky (Clifford, 2020). They say that big red dogs are better, but they don't account for the fact that big yellow dogs tend to be passed over for their appearance, but often have great personalities.

The abuse of 'they' in this paragraph is a common example of students being non-specific in their language and it's often systematically in their writing. Who is 'they' after all? Is it Clifford? Perhaps it's the red dogs? Maybe it's the editors at the Wall Street Journal? Who knows. But if you focus your feedback on bringing the students attention to this same one or two repetitive problems then you make the feedback digestion process more manageable for the students.

So in editing their work you want to bring their attention to at most 1-2 systematic problems that are persistent in each assignment. By focusing on just 1 or 2 systematic problems, the next assignment should have a majority of those habits improved. Then on the next assignment you pick 1-2 things that are systematically wrong in their writing and you keep 'layering' this procedure over the semester. You might recognize multiple systematic problems, in this case, I still suggest you show restraint on addressing all of them at once.

As a result you end up with a compounding interest problem and you will see improvements over time by being incremental and targeted in your feedback. Of course, you can only do so much, a lot of work rests on the shoulders of the students and you should not bare their burden more than you need to.

  • A lot of effort indeed! Interesting idea but sounds like one to be implemented after significant experience. Unless I invest heavily, as you say.
    – luchonacho
    Commented Dec 7, 2020 at 18:48
  • What do you mean by 'after significant experience'? Commented Dec 7, 2020 at 18:57
  • I'm new to the university and course. So I mean after many iterations, until you get to know their style and shortcomings.
    – luchonacho
    Commented Dec 7, 2020 at 19:22
  • 5
    I don't believe this to be correct. Good writing principles transcend courses (you said this yourself ;) ) and the students change every single year, so there's no better time to start then right away! Try to experiment with it, otherwise, you'll never really know for yourself. Besides, what you seem to be doing now doesn't seem to be working the way you want, so what do you have to lose? Commented Dec 7, 2020 at 20:02
  • I like this idea, but I'd suggest looking for common weaknesses across the class for the first writing assignment, and then make the appropriate improvement a specific, explicit requirement of the remaining assignments. Repeat for each subsequent assignment. That would generally help most of the class without shifting too much of your time and work from the primary subject matter. Commented Dec 10, 2020 at 18:47

One of the struggles that I've always had is getting students to understand how badly they write, and why it matters. If they don't understand that, why would they work at improving anything? (For context, I'm coming from a maths background, so the specific things I do may not help you, but perhaps the idea can inspire something?).

What I like to do is essentially a peer-review exercise, but the review element is more about the students using what's been written - they have to interact with it on a deeper level, so that they have to struggle against the consequences of poor writing.

For example, I divide students into teams. Each team gets a box of lego. In the first session, they have to design a model and write instructions for how to build it. They are not allowed to draw pictures. The idea is that the sequence of precise instructions mimics a mathematical proof. In the second session, they get a different team's instructions. They don't just have to read the instructions and assess them. Instead, they are challenged to build the same model as was originally designed by following the instructions. This is the point where they really understand how unclear the instructions were and why it's a problem. (If a mathematical proof isn't repeatable, it's not much of a proof!)

Could you find a similar sort of exercise that models what your students could be expected to do in the future? Where they produce a document, and that document has a purpose and whose success is affected by the quality of the writing?

  • 2
    That is a very cool and clever exercise! Thanks! Will see how can adapt it to my context.
    – luchonacho
    Commented Dec 9, 2020 at 13:14
  • 1
    This is indeed pretty elegant and cool. Commented Dec 10, 2020 at 4:42
  • This sounds like a lot of fun. I could imagine this as a game people played for fun outside of class.
    – Brian
    Commented Dec 10, 2020 at 19:11
  • 2
    Did anyone use this or could refer to a similar example that could be applied in social science or medical context?
    – sophar
    Commented Jan 14, 2021 at 8:29

You should look into Writing across the curriculum. Even if you get no support from your school, there are things you can do as an individual instructor.

You should discuss your writing remediation plans with your "management" (assistant chair or someone "in charge" of your course) in advance, so they would not be surprised. You should not introduce any new requirements in the middle of the semester.

Assign work where the students will have to write about the material in your class. Several short narratives (5-10 sentences) will work better than one long essay per semester. If their writing is not to your satisfaction, then make corrections (note that will be a lot of work for you if done properly) and ask the students to re-write and re-submit.

Make sure that the poor writing in the final version does not affect the class grade, but a failure to submit the work will affect the grade at least 1 notch.

Consider using one of the automated tools that detect plagiarism in student writing.

  • 5
    "a lot of work"... indeed! That is why I guess no one around here seems to put effort on it. You say writing not to affect class grade. What about writing change? I could as for an initial ssay and see improvement over the term.
    – luchonacho
    Commented Dec 7, 2020 at 15:19
  • 2
    I'm confused why poor writing in a final version wouldn't affect the class grade, did you mean the first? Commented Dec 7, 2020 at 15:22
  • 1
    @luchonacho What would you do with a good writer who didn't improve at all? Commented Dec 7, 2020 at 15:22
  • 3
    @AzorAhai-him- this is not a "writing" class. The class grade assesses their work in another subject. Just requiring students to turn in written narrative, and giving them good feedback (again, this is a lot of work for the grader) will help some students a lot. Commented Dec 7, 2020 at 15:40
  • 5
    @DimitriVulis I suppose I don't agree that "their work" is completely separate from "the writing." It seems like many students wouldn't put in the effort to improve if it doesn't affect their grade. Commented Dec 7, 2020 at 16:18

For us in the French system we had a very rigid way of writing which ended up being pretty handy. I think it's the thing that taught me how to write well. You could teach your students how to write in this way. I will present the method and after say how it was taught to us.

So, the method is the following, and I'll try to be quick. You take the question you were given and analyse its keywords. Do some brainstorming on how to answer.

For the introduction, you give a general idea to catch the reader's interest. Then, depending on the question, you give context and/or definition of keywords. You may also explain why the question is interesting. Then, you reformulate or just write the question again, and you announce the big ideas your development.

For each big idea, you make a "part". You start by a logical connector (first, secondly, etc.), then by providing a summary of your idea, then by announcing the sub-ideas it is made of.

Then, for each subidea, you start by a ~one phrase summary, then the bulk of your argument, then your examples, then you either summarize again or just explicit how the examples are relevant, and then you transition for the next subidea or idea.

For the conclusion, you summarize and maybe then you "open" the question.

Now, the thing is, to teach us this, the teacher went by parts. So every week (which is possibly too much for uni) we had to turn in something. It could be only the introduction, or only one big-idea, or just the structure (so the "titles" for the big ideas and so on). And we had a clear barometer of which things we weren't doing well (conceptualizing the question, organizing our ideas, etc.).

And also, with time I have incorporated some things of this method into my "normal" writing, but not everything. It can be a bit too boxy. Also, I was kind of quick here. If you want more information, let me know.

  • 1
    In English, we have the five paragraph essay which is a similar structure. en.wikipedia.org/wiki/Five-paragraph_essay Commented Dec 9, 2020 at 15:28
  • 1
    I like this answer, simply because good writing is dependent on good form. When assigning an essay, be SPECIFIC about what is required structurally, just as in this answer. The forces your students to organize their thoughts before even writing. Afterward focus on general deficiencies, such as mixed idioms, loose grammar, etc.
    – Dúthomhas
    Commented Dec 10, 2020 at 0:19
  • Might be a different generation, but: My native-born-French partner says that when she moved from France to America, she was completely bewildered by the American structural expectation of introduction/topic/support/conclusion, because she'd never heard of it growing up in France. Instead, the emphasis had always been on thesis-and-antithesis -- which itself got her graded down in American college as incoherent (not arguing for a single position). . Commented Dec 10, 2020 at 4:40
  • 2
    @DanielR.Collins That's also true! The French system has its flaws, but it certainly teaches you critical thinking —and that has a lot to do with taken different points of view and positions into account. However, I was used to thesis-antithesis only on certain subjects, not all. Commented Dec 10, 2020 at 16:00

Writing is learned by reading

Maybe give them some easy papers to read during the course of the semester and have them write summaries. If they don't know how scientific literature is supposed to be worded, as a start, having them read some good examples is likely gonna be more helpful.

  • Yeah, just getting people to read more would be a big plus. It would be great if they read classic scientific papers, but I'd settle for literally anything. I suppose Bertrand Russell is out of fashion these days but maybe there's someone a little more au courant, with the same clear, concise, substantial, and witty style. Commented Dec 10, 2020 at 2:29
  • ... and practising, for sure
    – luchonacho
    Commented Dec 10, 2020 at 17:36
  • @luchonacho Sure, but for any practice to be effective, you need at least a semi-solid idea of what the end result is supposed to look like. Most of your students likely haven't read a single paper in their lives, so I'm guessing a lot of them are completely lacking this idea.
    – MaxD
    Commented Dec 10, 2020 at 19:30
  • Writing is learned by writing. You don't learn to drive a car by watching other people drive a car. Commented Dec 10, 2020 at 22:51

Here is a specific recommendation/suggestion. It depends on pairing and peer reviewing. It assumes that you don't really have the time to do adequate multi-level, multi-version review yourself.

Create a double assignment. Two different things assigned at the same time. Give a pair (self selected, random, you select, ... ) the two assignments. They decide among themselves who is responsible for which one. Each writes up one of the assignments and asks their partner for advice on it. They then rewrite, maybe more than once.

Decide on a grading scheme before you start. One way is to give them both the same grade on the assignment. Another is to give out, say, 90% of the grade yourself and let each student give the remaining points to their partner. Those points are specifically for the quality of the feedback received. Expect, of course, that almost everyone will give full points to their partner, or none. The latter result gives you a chance to advise them about teamwork, which they might not otherwise get.

If they want to "cheat" and work on both together, fine. Let it happen. It is an additional skill that will serve them later.

But, the answer of Ethan Bolker is the basis of any good scheme. Repetition with feedback is the only way it can be made to work. If you can give the feedback yourself, given the scale, and you can permit resubmissions in some form, then do that.

  • I really like the idea but did not completely understand the grading options. Is this described somewhere in more detail, do you have any reference?
    – sophar
    Commented Jan 14, 2021 at 8:42

I have a couple of suggestions. My background is that I'm an engineer. I graduated about 40 years ago with my bachelor's degree.

Back in the day, our department had two "writing" courses: "Tech Paper 1" and "Tech Paper 2". Each was a single credit course, and in each course, the papers were due a little less than a month into the semester. For each paper, there were very strict rules. Each had to be less than 10 pages, double-spaced (back in the day when we used typewriters - I don't know how many words that might be).

The papers had describe a technical problem in the "Problem Statement" (aka Introduction) section. Then they had to propose two or more possible solutions to the problem - including data, feasibility, etc. Finally, there was a Solution section that explained why one of the possible solutions was chosen.

The papers also needed a clear, less than 100 word abstract.

I can see something similar being done with management students. The problem wouldn't necessarily be technical (it might be choices for a product launch, for example).

In those days, the department's standard grading ended up with <10% A grades, maybe a little less than 40% B. The rest were Cs, with a smattering of D grades.

The rules for tech papers were special. If the student got an A or a B on the first submission, then the course was finished (though B students could resubmit and try for an A). Otherwise, the student needed needed to take the comments on paper, incorporate them into a new submission and present that. An A or a B on the second try lead to an A or a B. Otherwise, there was the possibility of a third submission (I don't think A was possible at that point). I believe that the grading on the last submission was a bit tougher (for example, if you had a C and didn't noticeably improve the quality of the paper, you could possibly slip to a D (I think - it's been a long while)).

The idea was to get students writing like professionals, not like coddled high school students. The grader's comments were nearly all in the area of making the paper more concise and more focused (well, as well as improving grammar, spelling, etc.). The idea of multiple submissions was to make the student understand that his/her work would be reviewed out in the real world, and that re-work was pretty standard. The grader's expectations were said to be in the realm of what a supervisor's expectations would be once the student graduated.

There was some allowance made for students who were writing in a language that was not their primary language (and, in part because this was in a Canadian university, papers could be submitted in either English or French).

I thought then, and still think that this was a great way to get students to write clearly.


Since that time, I've been working in industry (with the occasional detour into graduate school). I've mentored several of my co-workers on their writing skills.

As other people have mentioned, the best way to learn to write is to read a lot. But, I go further and say that when you find something whose style you like, you should read it out loud, and listen to what you are saying when you do that.

Then, when you write, you should always read what you have written out loud. Listen to what it sounds like. See if you run out of breath before you finish a sentence. See if you can muddle through that 50-60 line paragraph without forgetting what the point was. When you read out loud, you will hear the fact that you included the same word in the last three sentences.

I'd like to think that having students read an excerpt of their work out loud to the class would be a good idea. However, that would be unfair to non-native speakers as well as those who really don't do well in public speaking. But, if you decide to pair students up to review each other's writing, having part of that review being done out loud might be a good idea.

  • 1
    Sounds like a good idea, but it would require an extra writing class. Commented Dec 9, 2020 at 9:03
  • 1
    @henning--reinstateMonica: Well, yeah. And the grader spent a lot of time on the comments from what I saw from peers. The selling point is graduating students who can communicate clearly. My second point doesn't require a class though. I'm a really big fan of reading what I write out loud; I find it tightens up my writing.
    – Flydog57
    Commented Dec 9, 2020 at 19:49

To expand on Julia Sepúlveda's answer on text structure and Ethan Bolkers answer on form, provide them with instructions. Maybe even spend one lecture on this. We had a course called 'study-qualities' in my first year and writing was a large part of it. But if your university does not have such a course, you have to make do.

The instructions should be hands on, and include pointers like:

  • divide your main text in paragraphs
  • paragraphs have headings but headings are not part of the text: the reader must understand the paragraph without having read the heading
  • start each paragraph by explaining in one or two sentences what you will say in the coming text
  • end each paragraph with re-iterating the points you made in said paragraph
  • run a spell-checker
  • have someone else read your text before turning it in

The best approach I saw to this was to provide examples. One of my philosophy subject tutors wrote four different short essays (<1000 words) on the same topic. They were of varying quality with respect to structure and evidence. So the good essay was organised logically, each paragraph concerned one argument and gave reasons for the position. On the other hand, the bad essay was a mix of assertions without evidence, inconsistencies, related ideas very far apart etc. All four essays were given to the tutorial class and they were asked to grade them (in groups) with reasons for the grading.

The bad essays took him a lot longer to write. Instead of writing bad essays, I imagine you could find some examples in one of the 'homework assistance' sites and then work your way up to write the better ones.


Be ready, you will face a surprisingly strong resistance. By forcing them to write on a level which should be the most basic requirement after the secondary education, it is something which is very alien for them.

On the superficial level, their reaction will be that they do not understand what do you want. In the reality, they will actively work on to attack you back. Because it is an attack for them.

This is a radical thing. The word "radical" is coming from the latinic word for "root". Because you grab - and fix - the root of the problem.

Be ready: the problem will attack back. Imagine the worst possible attacks, like:

  • "Sorry I have dislexia" (this will be a lie in 99% of the cases)
  • "Sorry I did not know that I need to write well here" (crap talk)
  • "It is a ... class and not a grammar class" (pseudorandom scratch on a paper is still not an essay, furthermore literacy is a must-have already for the secondary education)
  • They will likely attack you indirectly, partially through their parents, and through the University administration.

It is not about politeness, it is not about beautiful suggestions. It is about that if the student is incapable to formulate round sentences, then you have nothing to grade. You might try to understand his writings - a little bit -, but honestly you are not an elementary school teacher.

I remember as I both laughed and hated the teachers who forced me to write in a good quality. Now, many decades later, that I see that a large part of the University students are in fact functional iliterate, I know that they were right and I would do the same, yet more vehemently.

So, the first important thing, be ready for sneaking counterattacks.

Second important thing, you need to have a hard hand.

Third: you will likely lose. Do not count with radical results, but target them. You have a good chance for a little improvement in the short term. But they won't write correctly, it does not matter how strongly do you "motivate" them, and you can not let fail the whole class. Your gain will be:

  • You will know that you did your best.
  • You made the Humanity a little bit better.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .