October 12, 20214 yr I figured the number of highly educated assholes here, I cant be the only academic on here. This might make a solid landing spot for a variety of things that we either need to bitch about or ways that we can help each other out
October 12, 20214 yr Author I am a Associate Editor of an Analytics journal and right now I am looking for a reviewer that is familiar with Fuzzy AHP. Any of you fuckers need some service to the discipline for your annual PERT reviews and know about the method?
October 13, 20214 yr 13 hours ago, RDCanecutter said: Nice try, cop. I don't know man, he speaks the lingo. I think he could be legit.
October 13, 20214 yr Author Last time I try and help out my other fellow nerds in academia In truth this is a microcosm of the mindset of academia. Untrusting suspicious assholes that are easily fooled by big words and have a built in disdain for helping random strangers due to the fear of missing out on something themselves. Time to go make a midterm exam…
October 13, 20214 yr On 10/12/2021 at 8:36 AM, Laxtonto said: I am a Associate Editor of an Analytics journal and right now I am looking for a reviewer that is familiar with Fuzzy AHP. Any of you fuckers need some service to the discipline for your annual PERT reviews and know about the method? “an”
November 3, 20214 yr On 10/13/2021 at 8:23 AM, Laxtonto said: Time to go make a midterm exam… Don't you just recycle last semester's exam and change a couple of numbers? Shouldn't take more than about 10 minutes.
November 3, 20214 yr Author 4 hours ago, Fudge Nuggets said: Don't you just recycle last semester's exam and change a couple of numbers? Shouldn't take more than about 10 minutes. Hell no. Not with how much shit instantly ends up online on course sharing websites now. The days of reusing hw or exams is over. Generally it’s write new narratives to go with a stable dataset, then randomize the data a bit just in case to make sure the answers are different for each set, then change variable names and then rotate some questions in and out. Do that twice for the 8 or so major topic areas and then set up the versions to be delivered online. So 8 topics with a random choice of which of the 2 different data sources, each using a tiered selection for 5 of 10 random question stems. So any exam that I give in a digital format can be 40 questions in order we covered in class and yet is almost impossible to have the same exam as someone else in class… Face to face is similar but do 4 for each topic area and just use the same question stems. Make it technically easier to have the same exam, but I then control where they sit so that helps get around that issue because the exams are no longer randomly assigned.
November 4, 20214 yr Author 3 hours ago, Fudge Nuggets said: So yeah... about 10 minutes. Most exams, to do it right takes 3-6 hours maybe more depending on the topic to build and then about the same time amount to build out online once you are done with all the disability accommodation and image descriptions that are now required to make anything online compliant. The hard part is since it’s almost all applied analytics you actually have to run all the analysis for topic, so 8 models or models sets that include 8 different imports of data x 2 for the two different data files. So “building” the question stems isn’t too bad (outside of coming up with new bullshit narratives) but then running the models to include the correct answer for each question plus at least 2 others that are plausible takes a while. Essentially I’m making sure that all 160 questions with the exam question bank has correct answers based on their data file plus at least 2 other distractors. That takes a shit ton of time to do it right. The only bonus is it is auto graded. I could always just use the book questions (except I generally don’t assign a straight text) but that really doesn’t teach the students anything but to memorize the textbook/slides/hw.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.