When do TAs do their marking?

[St Andrews Lynx]

As I find myself awake past midnight with a pile of lab reports in front of me (now covered in red pen...because I'm a traditionalist like that) I'm wondering where other 1st year or more experienced TAs set aside time to go through their marking?

 

I suspect it's better to try and mark them all at once (so you can remember why you're deducting 0.5 marks on Question 2)...but would fitting in bits and pieces during the day time around other tasks be equally effective for simpler short questions?

 

Also...do you do your marking the day the assignments are handed in...or leave it to the last minute...or pick a time somewhere in between?

[fuzzylogician]

I grade assignments in one go, in an afternoon I keep clear for that purpose. Normally I do it a day or two before I need to return the assignments, not right after the deadline because students get extensions or otherwise turn assignments in late and I don't want to deal with it twice. I start with some students I expect to do well to get a feel for how well the assignment went over. If several good students messed something up I will probably want to address it in class rather than on individual assignments and I may not even take off points, depending on how pervasive the problem is. I then do the rest of the assignments, and always save some good students for last, to keep morale high towards the end of each cycle. I grade the same question for everyone, then move to the next, instead of grading whole assignments for each student. I also normally have a grading key and I add to it if I need to deal with unexpected mistakes so I can remember why I took that 0.5 point off.

Most importantly: I decide in advance how much time i am going to spend on grading, based on what I am paid and how much time I invested in other TA related work. I don't mind working more on some weeks, but not on a regular basis. It might mean that the students get fewer comments on their papers, but they instead get a solution or we discuss things in class together. Either way, it's important not to let TAing take over all of your time but instead keep it under control.

[rising_star]

I wouldn't necessarily do it all at once. For multi-question assignments, I'll grade 1-4 questions at a time (depends on the length of each one), take a break, then grade other questions. That way you keep the consistency across questions but also give yourself a break. Or, make a quick cheatsheet/list of why you're deducting that you can reference later when you're more tired. The latter has the benefit of being something you could share with students if they want to know more about their grade or question your marking.

 

I don't put grading off until the last minute. I try to do a little bit a day if I have a stack of essays. For example, last semester I had to grade for 60 students and for each essay they had 3 choices of which prompt to answer. So, day 1 I'd divide them based on the prompt, check Turnitin, and skim for obvious plagiarism issues. Then, after that, I'd do a quick skim of a few per prompt to get a sense of how they did and where I'll likely need to deduct. After that, I grade. I personally set a timer per paper because otherwise it's easy to get bogged down making lots and lots of corrections on a student's paper. The time is sufficient for reading, commenting, and writing the grade down. (For last semester, I used 6 minutes because papers were 2-3 pgs double-spaced and I needed to fill out a rubric. Any time left I used to check email, grab water, eat a snack, etc. And then after every 8, I took a longer break.) For me, this strategy works well but it's not for everyone. It took me a few semesters to figure out what works best for me and what I've ended up with is something that may not work for you.

 

If you're literally leaving a pile of red ink, you might want to rethink your grading/commenting strategy. I try to give only big picture comments that summarize the key highlights and flaws at the end. In the paper, I tend to focus on things like missing evidence or citations, bad argumentation, or lack of a thesis statement. It There's so much I could correct but then it'd take me all day. Hence the timer thing to make sure I don't fix all their problems (there's a writing center for that!).

[TakeruK]

I grade pretty much the same way as fuzzy indicates. I'm only usually assigned something like 3 or 4 hours of grading per week, so it's definitely something I can do in one afternoon. Also, the courses I was grading for usually only had 1 to 2 questions per week (but they take a pretty long time for the students to do and for me to grade), or they were things like lab reports or lab notebooks.

 

I try to make a full detailed marking rubric for each type of work I'm marking (e.g. problem sets, lab reports, lab notebooks) and hand this out to all students at the beginning. I explain it the best I can so they know what's expected of them. Then, I make an abbreviated (1/4 page) rubric with just the keywords/criteria (and their point value) and when I grade the work, I just circle whatever the mistake was, or tick whatever they did right. I write their final score on the rubric (it also shows the points breakdown) and staple it to their work. This saves me a lot of writing the same stuff over and over again on multiple assignments! (e.g. I can just circle "sig figs" or "show units" or "label plot axes" etc.). And it makes sure that I always take the same number of points off for the same type of error!

 

I also make sure to keep time at the end for doing "admin" things, such as entering grades into spreadsheets and writing up an email to the prof with a summary of the average, the distribution, and common mistakes. 

 

As for scheduling marking time, if I was able to set all of my due dates etc. I would choose to have homework due at noon on Wednesday, allow students to hand in late homework with a X% (usually between 30% and 50%) late penalty until noon on Friday, and then do my marking on Friday afternoon and/or the weekend. 

Edited September 10, 2013 by TakeruK

[csibaldwin]

I grade them as they come in, they are on blackboard as .doc files, so I grade them with in-text comments. I try to get them back to the students as soon as possible. They take forever to grade, even though they are only 2 page essays, because of the horrible grammar. It is an intro online class with 70 students, and I am the sole instructor.

Edited September 11, 2013 by csibaldwin

[Nerd_For_Life]

I start grading once the assignments come in, or a day or two after, with the goal of a 1-2 week turn around time depending on the length of the assignment. The amount I grade at once and for how many hours in a row changes depending on what other stuff I have to work on that might be of a higher priority (or what other stuff I have to work on that I'm avoiding working on )

[comp12]

When I started, I did the whole attempting to mark in as few sittings as possible thing too.

 

Then I sound that I was going crazy. And that students whose assignments were towards the end of the pile showed a correlation to lower grades

 

So now, I make conscious efforts to space them out.

[TakeruK]

When I started, I did the whole attempting to mark in as few sittings as possible thing too.

 

Then I sound that I was going crazy. And that students whose assignments were towards the end of the pile showed a correlation to lower grades

 

So now, I make conscious efforts to space them out.

 

That's a good point! For my weekly marking, it usually only took 2-3 hours to finish so I don't think that was long enough to severely affect the last assignments graded. Like fuzzy said, I try to keep track of the time per student and try to keep it equal. I generally have 20-30 students, so it's about 6 mins per student!

 

When I graded their final lab or project reports, which took several times longer, I did space them out. To compensate for potentially different moods affecting score, I tried to make sure my time/student was still the same and I made a more detailed/strict rubric for myself to follow. Also, at the end, I also did random checks to make sure the points deducted/awarded were consistent across marking periods.

[danieleWrites]

I grade as they come in or all at once, depending on my schedule and the assignment. Almost everything is turned in online and stuff starts trickling in several days before it's due. In-class, graded writing/work is an all-at-once thing. I have to comment on essays many times throughout a semester. Oh for an answer key! Can't they learn composition with multiple choice?!

 

Any graduate program will tell you that you're a student first and teacher second. I do what fuzzy & rising_star do. I have specific amounts of time set aside for all grading activities and don't do grading outside of that time without a good reason. I don't set a specific schedule, because I do not grade papers when I'm either upset or really happy (the assessment tools I must use are almost entirely subjective); but I do set a specific amount of time aside per week. Since I make the assignment schedule, I can plan clearly. I also use a timer (for the papers that I inevitably want to spend a lot of time with) and I have a place where I do nothing but grade and a specific ritual that I do to start grading (computer there, assignment instructions taped to the wall here, a hand-written list of the very specific things I'm looking for in the assignment on sticky notes on the right side of the monitor, and their reference book open so I can quickly refer them to specific pages) because I'm all ADHD like that and these kinds of things help me maintain my attention on grading rather than on minesweeper or (as my current frustration levels with Blackboard dictate) hanging out on the internet.

 

One thing to note: students rarely look at everything you mark on their work; some won't look past the grade. The more you write, the less of it they'll read. Write less on the papers in terms of pointing them back toward the right track during future study and review sessions. Use questions, rather than statements. Don't write in complete sentences. Have a few specifics in mind about what you want them to understand from the work (the big picture stuff) so you've already got a framework to make quick decisions about what to annotate on their work and what to just mark incorrect. You're an intelligent, thoughtful, and organized person, so you've got your own way set, no doubt. To save myself some commenting time, I'll sometimes bring up the most worrying trend in the errors. Sometimes many students make the error, sometimes it's a few.

[St Andrews Lynx]

The rubric sounds like a great idea - for me as well as the students! Thanks for all your input, folks.

[surefire]

  Then, I make an abbreviated (1/4 page) rubric with just the keywords/criteria (and their point value) and when I grade the work, I just circle whatever the mistake was, or tick whatever they did right. I write their final score on the rubric (it also shows the points breakdown) and staple it to their work. This saves me a lot of writing the same stuff over and over again on multiple assignments! (e.g. I can just circle "sig figs" or "show units" or "label plot axes" etc.). And it makes sure that I always take the same number of points off for the same type of error!

 

I also make sure to keep time at the end for doing "admin" things, such as entering grades into spreadsheets and writing up an email to the prof with a summary of the average, the distribution, and common mistakes. 

 

 

An abbreviated rubric! I love that idea! I might have found it difficult to implement in my first year of TA-ing, as it can be difficult to predict the ways in which students will go astray; but now that I'm TA-ing some courses for the second time, I might go back and reverse-engineer some abbreviated rubrics, based on the common feedback I gave on assignments LAST year that will be given out to a new batch THIS year! Great idea!

 

Setting aside "admin" time is also a great consideration! I initially tried to allot every paid hour that I had to being attentive to student's work, but there will always be some kind of "time suck" like spreadsheets or e-mails or plagerism reports, so it is definitely good to plan for Murphy's law!

 

I would add to the advice provided by multiple awesome posters by saying that it is OKAY to ask senior students who may have TA-ed the class before you if they could share a rubric. I typically always make my own (unless the prof has one she is committed to), because I find the process of making one helps me to be very knowledgeable/accountable/consistent when doling out marks. This is helpful when students show up to ask for elaboration on a mark and you KNOW what standards you had in mind, because you built them yourself. That being said, I have asked for rubrics from previous TAs so that these can serve as a starting point. This, of course, all depends on what kind of prof you are TA-ing for, or if you are instucting the course yourself OR if there are multiple TAs in the course; in these cases, you may well be using an existing rubric, or co-building it with the prof/other TAs or you might build it yourself but have to submit it somewhere for approval before implementing it. If you're not sure, just ASK, it will save time!

 

A note on marking when the course has multiple TAs: I have found that an early marking meeting is helpful, ideally within a few days of the due date, when some assignments are still trickling in. In my experience, before these meetings, each of the TAs mark a handful of assignments based on the agreed-upon rubric - I usually mark until I have one or two high-scoring assignments, one or two low-scoring ones, and a handful of middling ones. I issue these marks on the rubric, not on the actual assignments. Then we meet, pass the assignments out to each other - keeping the marked rubrics to ourselves - and then we re-mark to see if different TAs come up with the same-ish mark. Usually, we fall in the same range. If we don't, it's an opportunity to parse out and come to an understanding on descrepencies.

 

Finally, I am really diligent about tracking my TA hours. Spreadsheet it. If you can't seem to get through the assignments adequately in the time allocated for each (and you don't hit a good stride after the first dozen), this does not mean that YOU are the problem. Any number of factors could be at work. Last year, I had a class of 60 and found that I was going over on assignment time. It turned out that, while the class was capped at 60, my hours were allocated based on my department's assumption that a few people would drop and that the class size would re-adjust to around 50. I substantiated the need to make a change with the prof by providing my spreadsheets, and she got someone else to do the invigilations so that I wasn't going over my time. My institution has union stewards that are willing to help negotiate strategies to help with this, so it's worth looking into and staying on top of!

 

Not all the advice on this thread will work for you, but you'll cobble together a strategy!

[TakeruK]

I'm glad to hear in this thread that there are other people that log their TA hours and keep track of how much time they spend on it! My Masters institution (in Canada) just signed their first bargaining agreement when I had started and I volunteered as one of the union stewards. I found that many of my colleagues really disliked the idea of logging hours. They either didn't want the extra hassle, didn't want to change their ways, or found the idea of being an "hourly worker" distasteful. I can understand reluctance to change, but I think keeping track of your hours is a responsible thing to do in order to properly manage your time without overworking yourself. Also, as we move forward in academia or even in professional settings, we will have busier and busier schedules and we will have to prioritize and manage our time. As for the "hourly worker" idea, I think in academia, we are often pressured (from ourselves or from our bosses) to work above and beyond the minimum requirements. But TAing is a job, and I think it's incorrect to say working more hours = more dedication / passion for teaching. 

 

I don't log my hours in a spreadsheet, but at my Masters school, the union encouraged all TAs to sit down with their supervisors at the beginning of the term and fill out a "hourly budget", which we would normally divide into very rough sections, like X hours for tutorials, Y hours for marking, Z hours for office hours, and perhaps some hours set aside for major marking events like projects or exams. I have a notebook I keep for TAing (I jot down reminders to tell the class next time, names of my students and maybe something about them so that I can remember who they are, etc.) and I set aside one page per category above and then I just write down stuff like "Week of Sept 9: 3 hours" etc. 

 

At my PhD school, in the US, there aren't any unions here and I think that's pretty common at US schools. The TA handbook here even explicitly states that while we might be assigned a TAship of X hours, that number is meaningless and it's up to you and your TA supervisor to determine how much hours you actually work. We also don't get formally paid for TAing, either! Since it doesn't make a lot of sense for me to try to "rock the boat" here, as the system works pretty well simply because I think there is an existing mutual respect between professors and students so that (for now anyways), that we are not taken advantage of as "free labour". Although it does make me uncomfortable that there aren't contracts protecting my right as an employee (actually we're not even considered employees, which also sucks!). But I digress -- I only brought this up because a lot of what surefire and I said about logging hours might not actually apply to many US schools.

 

Finally, multiple TAs for one course can make things difficult in terms of logistics! In addition to meetings like surefire suggested, another way to standardise marking is to split up the assignments so only one TA marks any one assignment. You would have to work extra hours in your weeks where you mark, but hopefully fewer in weeks where you don't so that it still averages out. One of my lab courses had two TAs and we had to mark lab notebooks every week. The class was split up into small groups that rotated through different experiments each week, so instead of each of us having to know all 8 experiments, we just split them down the middle and we just had to really know 4 experiments very well. We didn't try too hard to standardise our marking styles (other than making a common rubric together and comparing our averages) because we felt that over the course of the semester, each student would be marked by each of us an equal amount of time, so it should average out. Sometimes people decide to split the class up into two marking groups, but if you do this, make sure that the marker assigned to each group changes over time so that certain groups of students are not positively/negatively affected by different marking styles!

[Between Fields]

I'm the instructor of record, so I grade everything. Since I only have 20 students (yay for small, liberal arts and sciences institution!), I can usually grade fairly quickly.

 

For short assignments, I grade them immediately after class during my office hours. That's easy for me, though, because they're completion (as long as they actually followed the directions and I can read it). That's probably not so possible in fields other than composition.

 

For long assignments, I read through all of them, and then sort them by how they "feel," in terms of strong, middle, and weak, and then start grading from the bottom and work up. It's best to save something good for the end, so you're not completely worn out and grumpy. I do use a rubric, but I also scribble comments on their drafts. I did a seminar project on student feedback, and as someone else said: the more you write, the less they read. It depends, though, what the learning outcome from grading is: are you wanting them to revise, or are you justifying a grade? For my first two papers, I don't let them revise and so my minimal marking is even more minimal than normal, just enough to point out a few good things they did and a few bad things they did. Later papers get more detail, because they can revise them.