Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision		Previous revision
astro300_f11:day6 [2011/08/29 03:40] – created a_leeastro300_f11:day6 [2011/09/29 04:20] (current) – a_lee
Line 1: Line 1:
 +======AY 300 - Fall 2011: Sixth Day Lesson Plan======
  
 +=====Preface=====
  
 +Today's class will discuss writing effective test questions, the design-a-demo presentations, and continue board work exercises. 
  
 +==== GOALS ====
 +    * Review how to write effective exam questions.
 +    * Continue practicing good boardwork technique.
  
-=====Ethics Case Studies (15 min)=====+=====Section/Group Work Sharing (15 min)=====
  
-**Make sure everyone has finished the online Ethics course. **+The person on call for this week will share their section experience. Some questions include: 
 +   * What did you do?  
 +   * How did you implement your activities?  
 +   * What worked?  
 +   * What didn't work? 
 +   * What would you do differently? 
 +   * How did you assess learning? 
 +   * Did you receive any unexpected questions/reactions/etc.? 
 +   * Did anything unexpected happen? 
 +   * What were you thinking about while you were running section? Any moments of panic?
  
-**Quickly touch on the main points from the Ethics course (extended blurbs below). **+During this (and after), open the floor up for general questions and sharing about how sections are going. 
  
-Finish up with a class discussion of several case studies of situations that have arisen in C10. These could include a subset of the following (perhaps put each on a power point slide and project it). Some comments from the author are given in parentheses at the end: 
-     - The final exam was this morning, and all the GSIs have just finished grading. You are back in your office when one of your students knocks on your door. They are very distressed about the final, and is standing next to you at your desk. The conversation goes something like this: "I think I am going to fail the class. I know the final is curved, but I don't know if that cuts it. How much does HW and lab count towards the grade again? *sigh* I don't know what happened, is there *pause* //anything// *pause* I can do?" You assure the person they probably are fine and quickly look at the final grades and see this person is getting a 'C'. (//As the head GSI, I was able to access the final grades at this point. A normal GSI would not have been able to do that...//) 
-     - You have decided that Facebook-friending your students is OK. This is the 21st century and social networking is here to stay. When you're on Facebook one day, one of your students opens up a chat window. It is friendly at first, and somehow gets into him asking for help on one of the homework problems. You take the time to help him. Once that is done, he starts to talk more freely and begins sharing personal stories, including how he came out to his parents recently, etc. (//Friendship with students, establishing boundaries, etc//) 
-     - One on cheating 
-     - Another 
  
-**Ask the class to think about what should have been done and (if possible) how the situation could have been avoided. **+=====Design(Improve)-A-Demo Proposals (30 min)=====
  
-  * Some topics covered in the online course: +Each group receives ten minutes for their proposal. The proposal itself should be ~5 minutes, allowing 5 minutes for questions and critiques. 
-    * Diversity +
-      * Berkeley is a minority-majority school +
-      * Unlikely to be a major issue (especially in an intro astronomy course since the classes tend to be pretty diverse).  However, be aware that we all have unconscious biases and try to make an effort to be open and inclusive.  For example, try to equally call on males/females, different races, //etc.// +
-    * Disabled Students +
-      * Berkeley allows students with documented learning disabilities (or physical disabilities affecting their ability to learn) to have extra time (and occasionally a 'quiet' or private room, //etc.//) on quizzes and examinations.  Students must have their disabilities diagnosed or approved by the Disabled Students' Program, or DSP. +
-        * For examinations, this is typically dealt with by the professor.  If a student comes to you about a disability-related problem, tell them to contact the professor and/or Head GSI. +
-        * In Alex's class, students take quizzes in section twice during the semester. Alex usually lets DSP students know that taking the quiz normally has turned out successfully in most cases, since it is designed to last well under the allotted time. However, if students insist on it, then you must accommodate them. If possible, the exam may be extended in situ, or a small group can repair to another room. Otherwise DSP can help with arranging for a proctored exam with additional time  (Most general solution: borrow another GSI's quiz and give it in your office during office hours.) +
-        * If a student in your section is having major difficulties with the course but seems intelligent and really is trying, consider talking to him/her about being diagnosed for a learning disability. Obviously this is something to be dealt with delicately. +
-        * [[http://dsp.berkeley.edu/sbin/dspACCESS.php?_page=dspFAQ|FAQ for faculty and staff about DSP]] +
-    * Sexual Harassment +
-      * Should be common sense - don't get involved.  Power difference -> potential for serious problems. +
-      * [[http://ccac.berkeley.edu/policies.shtml|Policies on Sexual Harassment]] - Title IX: Sex Discrimination and Harassment (UCB Equity Standards and Compliance) +
-      * If you see something between your students that you think might qualify as harassment, you're not required to do anything (as I read the policy), but keep in mind that you're the authority figure in the room. You can report instances of harassment, but that's a fairly severe step. A reasonable start might be talking to the parties involved informally and individually. You should probably find someone with experience in dealing with such situations. +
-         * One person we suggest is Colette Patt (colette@berkeley) who "works directly with the dean on issues associated with diversity in science" and is supposed to be an excellent resource for harassment-related issues as well diversity issues. +
-  * Tons of information can be found on [[http://gsi.berkeley.edu/ethics/documents/Maggie_AnnotatedResources.pdf|this]] list of resources (all of which are mentioned during the Ethics Course). +
-  * It is very unlikely that a serious issue will come up as long as you use basic common sense.  If it does, know where to turn for help!+
  
 +===== Exam Questions (35 min) =====
  
 +==== Activity ====
  
 +  * Break into groups of two.
 +  * Instructors will distribute exam questions
 +  * (5 min.) Answer and Critique your partner's questions
 +    * Note that evaluating others' exam questions is equally or more important as knowing how to write them, since GSIs almost always vet exams but only contribute a few questions at most.
 +  * Ask them to evaluate question stem and distractors. 
 +    * Stem - Is this clearly worded? Does it concern a "major topic"?
 +    * Distractors - Are these worth the words they're written with? Capture common misconceptions? Clearly worded?
 +    * Think about where students may get tripped up, if anywhere. 
 +    * How can they be //improved//?
 +  * (10 min.) Discuss/Argue your critiques with your partner. Reach a consensus on what makes an effective test question. 
 +  * (10 min.) Discuss with entire class
 +  * (10 min.) Critique some of Alex's "proven to be good" questions. 
 +    * Questions should be difficult because of the concepts tested, not because of tricky wording. 
 +    * How might an undergrad read the question. How can you get in an undergrad mindset?
 +
 +Main take-aways:
 +1. The stem of the question should be meaningful but not wordy. (e.g., More than just "The seasons... (a)(b)(c)(d)(e)" but less than a paragraph about the Earth and its seasons (and then five responses).) 
 +
 +2. Avoid double negatives. In general, write the stem of the question in the positive. Students read negative terms and forget to reverse the logic of the relations being tested.
 +
 +3. Irrelevant difficult is unnecessary. What are you testing them on? If you want to probe their knowledge of subtle points, make sure it is warranted (e.g., "Is this subtlety necessary to understand a bigger theme of the course?")?
 +
 +4. All distractors (incorrect answers) should be plausible. I really liked Therese's suggestion of having someone not in the course read your responses (but not the stem). If they can pick out the correct answer (or obvious wrong answers), then you have work to do. That said...
 +
 +5. All responses should have a similar tone and length.
 +
 +6. Avoid "none of the above" and "all of the above", especially when all possibilities are listed in parts (a)--(d)! 
 +
 +Other questions to critique, if time:
 +  * Now critique some past exam questions
 +  * //Use old Alex/"proven to be good" questions//
 +    * Have the practice final from the 2007 AY C10 reader
 +    * Have some MC questions from the back of Slater & Addams.
 +    * LSCI questions: http://aer.noao.edu/auth/LSCIspring2006.pdf
 +    * ADT 2.0: http://solar.physics.montana.edu/aae/adt/
 +    * More concept inventories: http://astronomy101.jpl.nasa.gov/tips/index.cfm?teachingID=32 . (Most of these don't publicize the questions online; you have to email the author to get them.)
 +
 +===== Some Notes =====
 +  * Much of this is on the handout we handed out for quiz questions.
 +  * Exams exist to:
 +     * Evaluate student learning for University-required grading
 +     * Motivate students to study and understand the material
 +     * Allow the instructor to evaluate his/her progress educating students about the material
 +     * Provide feedback to students about their understanding and study habits and illustrate specific gaps in their understanding of the material
 +
 +  * What makes good multiple choice exam questions?
 +     * Not too long/wordy (neither the question nor the possible answers)
 +     * Not too much calculation
 +     * Not //too// tricky (//i.e.// there shouldn't be two extremely similar answers)
 +     * Relevant to important material (as opposed to really obscure/minute details)
 +     * **Very clearly written**, precise wording in both question and answers
 +       * //E.g.//, 'Which //best// describes...' as opposed to 'What is...' or 'How does...happen'
 +     * Pedagogical as well as evaluative (//e.g.//, some questions should probe common misconceptions)
 +     * Questions that are very easy or very difficult are **OK** as long as the test has questions with a variety of difficulties
 +
 +  * What makes bad ones?
 +     * Long answers!
 +     * Excessive use of 'all of the above' (some people say **any** use of 'all/none of the above' is a bad thing)
 +     * Questions that can be solved without knowledge of the material (usually because of the use of too many blatantly wrong or "funny" possible answers)
 +     * Multiple potentially correct answers (usually from vague questions or possible answers)
 +
 +  * Non-multiple choice questions (very similar to quizzes)
 +     * Types of questions:
 +       * Matching
 +       * Fill-in-the-table/blank
 +       * Simple calculations
 +       * Short answers and paragraph/free responses
 +       * Diagrams, plots, graphs
 +     * Most of the same points discussed above, and for quizzes, apply here:
 +       * Questions should be clear, easy to read, and unambiguous
 +       * Questions should be relevant to the material presented and emphasized (do not test on obscure passages of the textbook)
 +       * For high-value questions, allow for partial credit
 +       * Make the questions easy to grade!  Don't give students the opportunity to 'core-dump' for a problem: be very specific about what you're looking for in free response type questions.
 +     *  Can be be slightly more time-consuming than quiz questions, depending on the overall length of the exam.
 +
 +===== Board Work Exercises -- Midterm Review (30 min) =====
  

Trace: