January 15, 2014

Group Testing-taking Options to Consider

By: in Teaching Professor Blog

Add Comment

I’ve been doing some reading on group test-taking (often called cooperative or collaborative testing in the literature). I am stunned by the number of studies and the many ways the strategy has been used. I’m not going to summarize the research in this post, but rather offer a collection of options. Most of these ideas appear in more than one article so I’m not citing references.

When first confronted with the idea that students might do an exam collectively or be able to consult with others on their individual answers, our teacherly eyebrows raise. How can this be ethical? Don’t grades measure individual mastery of material? Valid concerns, yes, but check out how the design features described below respond to those concerns.

This leaves the question of whether there are any good reasons to consider the option. The research is consistent on one point: collaboration reduces test anxiety, especially for the very anxious. And there are important lessons to be learned when consulting with others, like figuring out who is correct and how to ascertain the merits of an argument. When I used the strategy, I couldn’t believe the intensity and passion with which students discussed the content as they collaborated. Then there’s the fact that in most professional contexts, if you need an answer, consulting with others is almost always an option.

Let’s start with the basics: students can collaborate on an exam or a quiz. For faculty worried about the viability of the approach, quizzes are a low-stakes place to experiment. The collaboration typically occurs in small groups, often with just a partner.

If the fear is that some students will “free ride” and not study for the quiz, you can have everybody prepare as if they were taking the quiz on their own. When students arrive in class, randomly pair half of them with a partner. Both partners still take the exam individually, but they are allowed to interact, during or after they’re both done— their choice. They still hand in individual exams, but they can change answers based on their collaboration. Next exam or quiz, follow the same procedure. This means a student may take all, some, or none the exams with a partner.

Maybe some students won’t take the group interaction part of the activity seriously, but most will if there’s a chance to improve their grade. How about this approach: Students take the quiz individually, hand it in, and then convene with a partner or a small group to complete the same quiz again. Their individual grade is the average of their individual quiz score and the group quiz score. Or, the group grade is transformed into bonus points added to the individual score. Group grades of 100 earn five bonus points, those above 90, three points, etc.

And still another approach to group test-taking finds teachers limiting the collaboration time. Students take the exam and 15 minutes before the end of the period, they consult with others in their group for 10 minutes and then use the final five minutes to reconsider and possibly change any of their answers. If Scantron forms are used, students turn in their individual form first, then consult with their group and turn in a second Scantron indicating any changes. These changes become the answers that count. This option provides feedback that makes it easy to see whether changing answers based on input from others helped or hurt their scores.

Given the attitudes a lot of students have about group work, some may strongly object to group testing. Fine. Let students choose. They may take the test alone or with a partner. The partners do one test and both get the same grade. In one study, the percentage of students who chose to go solo didn’t change much for the next exam, even when students were told that partner scores were significantly higher than individual scores.

What if the quiz involves a skill, such as doing a set of tasks on the computer? Here the group is allowed to designate one member as their coach. The coach can help the group during the quiz but with one significant caveat. The coach’s hands are tied (yes physically, but loosely) so the coach can’t do the task but can still give directions, offer advice, and provide feedback. How about letting the group provide feedback, maybe even grade the coaching they were provided?

Interesting options abound! If you’ve tried group test-taking in any of your courses, please share your experiences in the comment box.

© 2014 Faculty Focus, Magna Publications. All Rights Reserved. Use of any content without permission is strictly prohibited.

email
Add Comment

Tags: , , ,


Comments

matthew | January 15, 2014

Sources please, I would really like to read the literature myself. This is a really interesting perspective

Elena | January 15, 2014

Your suggestions assume the instructor has a lot of time to grade/regrade these tests and offer various options. We don't have those resources at my univeristy. In addition, if collaboration to learn is the point and not just inflating grades, why not let the students collaborate after the grades are posted? Those who did the work and earn the good grades get what they earned, and those who didn't can still learn if they want to.

Clinical Professor | January 15, 2014

I do use this approach which results in a rich learning environment. Students complete an activity individually and submit this. They then collaborate and discuss with others and submit this second version also. I use the scores from the individual effort to group students for the next activity. Instead of grouping a "mix" of students, I group the high achievers, prepared students together and the unprepared students together. Some of the originally unprepared students quickly realize that they are not able to be a passive learner in their present group and for the next activity their individual effort shows a remarkable improvement. These students which I will refer to as "individual improvers" are then grouped with like students and a prepared or a high achiever. The individual high achievers or prepared students flourish in this learning environment and they don't feel like they are being taken advantage of by the unprepared.

Carol KIlian | January 15, 2014

We do what we call collaborative testing after every student completes the exam individually. They are randonly assigned in a group of five and they retake the test as a group effort. If they achieve the 93-100 percent range, they receive 2 "extra collaborative points" each. If the test percentage is 85-92 percent, each in the group receives one collaborative points.
The caveat is that they must have already achieved the passing rate on the individual test, and the points are not added into the final grade unless they have a passing grade. This promotes group work, and will often move student up one letter grade in the final calculations.

Melinda | January 15, 2014

This is the first time I have seen this topic discussed. I'm also interested in the research on it. For years, I have been allowing collaboration on low-stakes quizzes. My students take the quiz individually first, then a second time in a small group, and I average the two scores. My rationale for it doing this is to help students prepare for the licensure exams they will take in my discipline. As the students collaborate, they share not only information about the content but also test-taking strategies. My hope is that the less-skilled test takers will learn some techniques that will help them face the exams with more confidence and less anxiety. I have never offered a collaboration option on high-stakes exams, but these ideas are intriguing.

Linda | January 15, 2014

I was introduced to this technique through a faculty workshop presented by L. Dee Fink, author of Creating Significant Learning Experiences. I have used this strategy for low-stakes quizzes that address comprehension of the assigned text reading before students attend class. Students complete the quiz individually through our course management system and then at the beginning of class they use the Intermediate Feedback Assessment Technique (IF-AT) scratch-off forms. This process usually requires about 15-20 minutes for a 10-question quiz. I combine the individual and group scores as weighted totals, but I think several possibilities exist for implementing and grading this strategy.
http://www.epsteineducation.com/home/about/

I have observed (informally) much richer discussions and interactions among students. I appreciate the idea of grouping students by level of achievement and this would be simple to do when students submit their individual quizzes online prior to the scheduled class date.

Christopher | January 15, 2014

Even though I haven't been a student in almost 20 years, I still cringe a little when I hear other faculty discussing group work and team collaboration. As the overachiever (like most faculty), I was always the one who ended up doing all of the work while the group benefited. I even conducted my own little (unscientific) experiment during my last year of undergrad. I had the same faculty member for two courses. Both courses required a group assignment. In course #1 I approached the group project in the same manner that I approached group projects in the past: I did the majority of the work and the entire group received an A. In course #2, I forced myself to use a different approach and accepted the contributions (if any) from the rest of the group without editing. The end result was my first C. Lesson learned.

Amelia | January 15, 2014

Why not cite legitimate sources? Not citing them makes your column less useful and much less academic (terribly journalistic?). It lends much less weight to your suggestions and sets a poor example.

Env. Prof. | January 15, 2014

I have done it once, in response to a much lower mean on an exam than I was expecting. Without handing back the first (individual) round, I had students retake the test in groups of 4-6 WITH resources (told them to bring books, notes, filled out reading guides, powerpoints, anything non-electronic), and averaged the grades. I also told students that they should ask others in their group with answers that were different from theirs to back up their answer using the resources. I was hoping that those who had not successfully connected what was in their notes, reading guides, PPTs, etc. to the test, would learn from those who had. Averaging the two grades improved most scores, soothing anxious students. More importantly, the average on that section of the final was higher than it had been previously, suggesting that all students may have learned that material a little better by going through the review process. I might just do this again!

Tech Teach | January 15, 2014

I use this in an introduction class, where my students are given a quiz and are put in small groups depending on the size of the class. When everyone is finished I have them move into bigger groups until there is just one big group with all the same answers. So only one test has to be graded. I change it up at times and when the small groups are finished they write their answers on the board. If any answers are different, then we have each group justify their answer. Each group has an option of "jumping ship" if another group convince them that they are correct. I call these quizzes "community quizzes". After we finish, I give the students individual quizzzes over the same material. I only had one class that overall did not do better, and some of that class did not take part in the discuss. So the individual quiz caught them.

Jim | January 15, 2014

I always like to see sources too, making it easier to look up further information if I desire – but I don't agree that it lends less weight to Ms. Weimer's suggestions, nor that it sets a poor example. This is a very interesting idea, and I'm anxious to try it. Thank you for the information.

Robert | January 15, 2014

Resources is a valid concern–our time is limited.
I like to give "one exam to the group" so that the number of exams is less than if taken individually.
So, in that sense, it helps with the resource constraints.

It may also help in another way. When a student is absent, they want to "talk to the teacher" to get caught up. In my classes, I mandate the first few exams are taken as groups–and (I think) this approach removes some of their worry about getting caught up. That is, they can watch their peers do the problem on the test, and learn as a result.

Is there a free rider problem? Yes–and occasionally I get complaints. (Overall, I think the pros outweigh the cons…)

Engineering Prof | January 15, 2014

I've tried most of these variations and had problems with all but one of them.

When students were allowed to pick their partners, the top students in the class immediately found one another and gained an unfair advantage, and the weakest students whom nobody wanted to work with had to work with each other and suffered an unfair disadvantage. If I paired the students trying for ability heterogeneity and put the strongest student in class with the weakest one, second strongest with second weakest, etc., in the pairs with heavily mismatched students the best ones would do all the work and the weakest ones would high grades that they didn't really deserve, and both students in those pairings would (unfairly) do better than two average students paired with each other. Any other pairing scheme inevitably led to some pairs being much stronger than others and so getting unfair advantages. When the students could gather in groups initially and briefly go over the test and then separate and take the test individually, the groups with the best students got quick and accurate rundowns on how to approach every part of the test, giving them an unfair advantage over the other groups. The same thing happened when the students took the test individually and then convened in groups to review and revise their tests. Each of these schemes led to vigorous and (I believe) justified student complaints.

The only system that worked for me was having students take the tests individually, then get in groups to rework and revise, with the individual grades getting much more weighting than the group retest grades. The upside of that approach was that most students ended up knowing how to work the test to a much greater extent than they would have without the retest; the downside was that it required the instructor or TA to grade two sets of tests instead of just one, which is a serious drawback in a large class.

Lolly Ockerstrom | January 15, 2014

For the last 2-3 years now, I have encouraged students to work collaboratively on exams. On exam day, we discuss the exam questions (essay prompts for English literature) as a class, and I leave the room to allow the students to talk among themselves. They have about 10-15 minutes to discuss the exam questions. Then they settle down to write individual essays. The first time I did this, the students were utterly surprised–but they were much more relaxed and confident after they had had time to discuss the questions, and their essays were stronger essays as a result of this initial conversation. And: their essays were, overall, much more detailed and overall more thoughtful than if they had not been able to discuss the questions ahead of time.

Clinical Professor | January 15, 2014

From my experience, the weak students are not innately cognitive impaired. They are weak because they are unprepared and in some cases outright lazy. These "weak" students are the passive learners. I have seen dramatic "attitude adjustments" toward learning when "weak" or passive learners are paired together. They no longer can rely on someone else to carry them. The top students are not innately cognitive superior, they are the ones that are active engaged learners and instead of an unfair advantage, they gain a rich learning experience with students like themselves and are rewarded for their efforts.

Eunice | January 15, 2014

I have a collection of resources. If you email me directly, I will send you the file.

Eunice | January 15, 2014

The intent is to use assessment FOR learning, not just OF learning.

eunice | January 15, 2014

Effective group work is not a group assignment but rather the opportunity to collaborate throughout the term, in class. It may involve a group assignment but it works better if the focus is learning and their is ongoing formative feedback as well as group interaction guidance provided. Having said that, most of life involves an unequal distribution of work.

Christopher | January 15, 2014

" most of life involves an unequal distribution of work" Very true; however when it comes to assessing student knowledge we are also credentialing those students. Therefore, based on your statement, in theory a "D" student could receive the same credentials (via transcripts) as an "A" student. When the D student enters into the workforce without the knowledge implied by the credential issued by the University, what does that tell employers about hiring that institutions graduates in the future?

Allison | January 15, 2014

I do this for my introduction course. Students are put into teams at the beginning of the semester and can work together on the final exam (they have had to complete the previous tests individually). I have a class just for them to discuss their strategy and perhaps even divide up the chapters so that one person is the expert on one or more topics. Interestingly, teams don't always have the same score. They do decide individually if they agree with the other team members. Also, I am clear that if a team member does not feel like sharing all their answers with other team members, they are exempt from doing so. I also stress the importance of being a contributing member of the team. By the end of the semester, the teams have a good relationship and grasp on each team member's capabilities. They like this opportunity so much – it relieves their stress, and they actually study more because they are part of the team effort. I don't tell them that I'm going to do this at the beginning of the semester. I announce this two weeks before finals…as a surprise gift.

Shana | January 15, 2014

I do this in an intro/survey course as well. The students are grouped at the beginning of the semester and always sit with their group. I have a once/week 2 1/2 hour class. The students are assigned a reading before class. When they arrive in class they take a ten item, multiple choice quiz on the reading individually. They turn in just the answer sheet and keep the question sheet (marked with their answers) to then take the same quiz with their group, producing a single group answer sheet. I grade the individual answer sheets while they take the group quiz and quickly enter them into Blackboard after class. After the group quizzes are turned in I go through the correct answers. During the semester they take ten quizzes like this. Rather than ten quizzes at 10 points each taken twice totaling 200 points, the group quiz scores are weighted 60% and the individual scores 40% so the total quiz score for the semester is out of 100 points. Every semester I ask the students if I should keep doing my quizzes like this or change it and the majority always say keep it. After the quiz I use my lecture to elaborate on this topic. I expect they can learn from reading alone and then what they couldn't learn from the reading is what I use class time for. I was taught this method from a mentor in my Ph.D. program, PJ McWilliam. She suggested that it helps to have the teams/groups have a project over the course of the semester to complete together as well, that the quizzes are not enough alone. I have them do a bit of a book club on the last class session and then they use that book for the final. Each group reads a different book. It has worked for me. In a 2 1/2 hour class it is nice to have them talk to each other rather than just listen to me lecture too much. I rarely have to go over answers on the quiz, the group discussion take care of most of that. I would appreciate a reference list though, if anyone has one.

Mohammed Mohaidat | January 16, 2014

I like this technique. I want to conduct a study to investigate its efficiency and see if it's reliable. Could you kindly provide me with studies and any hints might help.
Best regards,
Mohammed Mohaidat
Jordan

Amelia | January 16, 2014

No sources means no way for the reader to follow up or confirm or learn more. We wouldn't accept that from our students. A professor should know better. This isn't just an opinion piece, it is a description of many options, mentioning that studies exist, but citing none.

Thom Mansen | January 16, 2014

I have used this method of testing for quite a while for the very reasons that have been cited. I would have groups of 4 randomly assigned just before the class. The tests were numbered and I had groups of 4 numbers on the overhead so students could find their own groups. My biggest fear when I first tried this was that everyone would get an A and there wouldn't be a wide distribution of grades. That fear never materialized. I found that the mean of the tests did not vary greatly when compared to the means of tests for students taken separately. I consider this format a way for learning rather than just a demonstration of what they had learned. I remember taking tests and having only myself to argue with or discuss the options with and I think for my style of learning, I would have gotten a lot more out of taking the test by being able to discuss the answers with someone else.

I did this for a Physical Assessment class and a Research class. When I would walk around the room I was pleased with the discussions that I was hearing from the groups. This was a great way to demonstrate and learn critical thinking skills. It was also interesting when students would ask me the answer to a certain question after the tests were turned in and many times they would indicate that they disagreed with the group answer and put their own and it was about 50/50 for those who were right in their decision and who were wrong.

I also give students the option to take the test by themselves if that is more their style, however, I've never had a student take me up on that.

I didn't do this, but I think it might be helpful to have the group of students grade their peers on their participation in the group discussion. This would encourage everyone to participate and not let anyone coast.

Mary Bart | January 16, 2014

Thanks all for your contributions to this discussion and for your interest in the literature where the author read about the different group test-taking options she’s highlighted here. Please watch for a follow-up post, complete with references, in the near future.

Mary Bart
Editor, Faculty Focus

Libby Pace | January 17, 2014

I teach an introductory course leading to a paralegal certificate. The course is heavy with terms, concepts and processes that are entirely new to all the students. My goal is to use the terms, concepts and processes as the content to develop critical thinking and communication skills. For the last few quarters I have done more and more collaborative testing. I started with collaborative quizzes covering terms and concepts. Students have 15 minutes to work individually on the quiz, followed by 10 minutes working together in small groups. Then we switch a few people in each of the groups for another five minutes of collaboration. Finally we discuss the answers as a large group. My theory is that this process ensures everyone has and understands the correct answers, exposes them to the concepts multiple times possibly creating longer retention, and encourages critical thinking and communication skills as they discuss different points of view/approaches to each concept. (If you are in class and take the quiz, you are assured full credit for it.) Last quarter, I decided to do a collaborative final exam. All questions are short answer or essay. Students have 20 minutes to read the exam individually and decide what they know vs. what they could use help on. Then they divide into small groups, where they can use their notes, PowerPoints, textbooks and each other to discuss concepts and approaches. They could also cross-pollinate between groups. In the end, everyone writes their own exam answers. Feedback from students was that the exam was "hard," which I took to mean challenging. Grades were best for students who worked with at least 3 others. Those who chose 2-person groups still did well. The one person who chose not to collaborate got a D; all others got mostly As or Bs (with one C). My sense is that the collaboration benefited everyone (who actually did collaborate), but the individual answers allowed individual preparation to count too. (My "midterm" is an individual, non-collaborative oral presentation; homework is individual and involves a lot of writing. Each class includes at least two small or large group exercises.) A big part of my reasoning for collaboration is the idea that there is a lot of collaboration in the profession the students plan to enter. I also think collaboration emphasizes and reinforces critical thinking skills in ways that individual work cannot. I discuss the course format and my reasoning for it throughout the course. I just started my second quarter using this format and the students are clearly excited about it. I look forward to refining my approach over the future quarters.

kincaiddonovan | January 17, 2014

Interesting!
I've always believed in collaboration, but I never thought of actually promoting it on class – not to mention on class exams. I give group activities but never on written exams.

So yes, why not?

I think this is cool. Really.
Thanks for sharing!

kxcn | January 17, 2014

surprising test
i am always try to develop a similar type group by all are enjoying his study.
kcn

FSPAC | January 17, 2014

We do not use group testing per se. We do have projects, presentations, and research topics that we assign to groups. We assign the groups to make sure everyone learns to work with others regardless of learning styles, likes/dislikes or other pre-conceived ideas. We also found that if we allow students to self select their groups, it soon becomes apparent that some groups merely reinforce ineffective studying and testing, repeating the same errors rather than correcting tem. We do have an occasional take home "exam" and we assume the students work collaboratively. Since we are a master's level health professional program, we need our students to learn to work effectively as part of the healthcare team as well as learn study and test-taking skills that ensure they pass boards.

mef24 | January 17, 2014

How do I email you direactly

David Rynerson | January 19, 2014

We tried a collaborative culminating assessment for our English 9 "book clubs" this year. If you're interested, you can read more about it on this Google doc. https://docs.google.com/document/d/1GbgKor_wz_k5j

Pam Kaus | January 20, 2014

I used dyads (partners) on my exams last semester for the first time. I gave each pair three short essay questions that required the students to use important concepts covered in class/book in order to answer correctly. They decided who would answer which question so that each one answered a different question. They wrote the essay and then gave it to their partner to read, correct, add to, discuss etc. The third question they each answered individually with no collaboration. They both received the same grade for the two questions used for collaboration and I added to that grade the grade for the one essay they answered without collaboration. This created two individual, unique grades. This class is only 55 minutes long and they had plenty of time to complete. The 3rd question was the tie breaker for the student who mastered the material effectively. I will use this strategy again.

eunice | January 20, 2014

eunice.friesen@umanitoba.ca

Pete Burkholder | January 21, 2014

For those interested in procedures and literature, I have a detailed article on group testing coming out in the journal _The History Teacher_ later this year.

Rebecca | January 23, 2014

I have been doing this for 4 years. The first & second unit exams I have my students take it themselves. But for the third & final exams I have the students get in groups of no more than 4. They may not use their notes or text books, but they can discuss the exam & compare their answers with each other. I give the instructions that they do not have to agree with their group's answer & they can circle the answer they think is correct. I have found that no group has ever earned a 100%. For me as an instructor it is so interesting to listen to the different groups say things like—I read this, I remember that, or the instructor said this. It allows the students to teach one another what was said in class, what was read in the text book, or what to look for in a question. And it takes the pressure off test taking. Education is about learning & application, not failure. This method makes learning fun, enjoyable, & exciting for all students. However, if a student does not want to take the exam with a group he or she does not have to. In today's business world we go to staff meetings & other group discussions so why not learn from group testing. I find my students have enjoyed the times they have spent in my class with pleasure & the students learn.

Gary Stark | January 26, 2014

I'm not sure, but given some of Maryellen's descriptions, she may have made some reference to my paper on group exams. See full cite below, either way. A few quick comments…
The concern about "too many A's" — if it's a reasonably tough, valid test, isn't this a good thing?
The concern that "one person does all the work" — not if the group is properly structured and rewarded and given an appropriate task (see my paper).

Highlights of my paper:
How I do them (after students are tested individually)
When groups are appropriate
Group testing is a method of improving learning
Group tests are easier for you and more enjoyable than "going over" the exam.
Eliminates students' surprise at their grades as they determine how they did as the groups progress through the exam.
A way of determining the validity of your test questions.

Stark, G. (2006). Stop “Going Over” Exams!: The Multiple Benefits of Team Exams. Journal of Management Education, 30 (6) pp. 818-827.

Audrey | February 5, 2014

I have been using this method for a number of years and student grades for individual tests have improved and student engagement is high. Students write both an individual test and a group test in the same class time. First is the group test which are situation analysis and essay type questions related to the unit outcomes. After a short break the students then write an individual test which is predominantly the T/F/MC type questions on the same unit outcomes.Overall Grading is: 80% of grade is from the individual test, 20% is group grade. Structure:First and most importantly, students learn how to work in groups during the first few weeks of classes and the skill is integrated into the entire course. Students work with the same group throughout the unit. Students must actively participate in 75% of classes/activities in order to earn the ability to write with their team. Students who do not qualify write the same (group) test on their own in addition to the individual test. Attendance is excellent in this course as students don't like missing an opportunity to write a test with their group and share a commitment to their colleagues. Students identify that the group test and being able to converse with colleagues is like a 'review' for the individual test. '2 heads are better than 1' and in my case '4 teammates are better than 1'. As mentioned, individual test grades have improved with the use of this system. This has become a great tool for both students and me as the faculty. I would love to meet with others at the May conference to discuss further ideas along this idea.

Jeanne Mullaney | February 5, 2014

A colleague of mine sent along this reference that compares student learning from group exams vs. individual exams. It also has a great how-to as an appendix, although it doesn't summarize various options.

Hollis Gilley, B. and Clarkston, B. "Collaborative Testing: Evidence of Learning in a Controlled In-Class Study of Undergraduate Students" Journal of College Science Teaching, 2014, 43 (3), 83-91.

(http://www.nsta.org/college/ — it's a featured article)

Mary Bart | February 5, 2014

Hello Faculty Focus readers,
Maryellen Weimer has written a follow-up article to this one, and it includes links to eight different studies.

A Follow-up to the Group Testing Article: Feb. 5, 2014 http://www.facultyfocus.com/articles/teaching-pro

Thanks for your continued interest in the topic.

Mary Bart
Editor, Faculty Focus

Paul | February 5, 2014

As a Geology prof, I use collaborative “testing” in two contexts: 1) graded results from laboratory exercises where the students are engaged in a discovery learning process; and 2) in tests of complex conceptual material. I do not use collaborative testing where the grading depends on retention of a significant component of information or relatively simple concepts. When I use the collaborative approach to a graded assignment, I am doing so because discussion among group members (and with me) aids in the learning process.

In Geology, the laboratories are mostly discovery exercises, not the “cook-book” exercises commonly used in the other traditional sciences — the discovery process is often aided by the multiple perspectives of a small group working together. There is commonly not a single route to the answer, and the sharing of ideas on how to get there usually means they cover more intellectual ground than if I were to prescribe the most efficient path … and come out with a deeper understanding of the system as a result.

In addition, we are commonly asking them to develop spatial and visual-tactile learning strategies — the development of any new learning style or learning mode is greatly enhanced by understanding the issues (i.e., difficulties, different approaches, etc.) that others are having with the same material and the mutual explaining of one’s own approaches and (mis-)understanding, coupled with explanations (note – not directions) from the instructor. Where there is a process to be learned (such as drawing a well-constrained cross-section), each student ends up producing a final product to be graded.

I use collaborative testing (usually as take-home tests) in certain upper-level courses where I want the students to be able to synthesize concepts when dealing with real-world-type problems that require the application of multiple elements of the course material. By allowing the students to collaborate (and ask me for help after they have worked on the problem for some time), I am really using it as a learning experience in how to approach and conquer large, complex problems … as well as making them revisit a wide range of course material (and hopefully internalize it better.

Note that I am using this when there is not a single right solution. To use it when there is a single right answer is to allowing students is to allow students to slough off in their learning of how to get that single right answer. For me, a collaborative graded experience is best viewed as a graded learning experience rather than as a pure testing experience.

@pamsong1951 | February 12, 2014

I have used collaborative testing with my senior nursing students for their fourth unit exam for several years now. Because a large part of my course is preparing them for their licensure exam, I used to spend a fair amount of time in exam review and, in some cases, in heated arguments over the rationales. The students now take their exam individually on Blackboard and receive their grade. Collaborative testing is completely voluntary. Any student may opt to take their individual exam grade and leave at that time. If they choose to do collaborative testing, they come to me and receive their group assignment. Randomly assigning the group numbers prevents the buddies from working together and usually keeps the groups fairly well balanced between strong and weak students. Once everyone has completed the individual exam, they go into their groups, one person in each group logs back into Blackboard, and together they take the collaborative test, which is identical to the test they just took on their own. If the group scores >90, each person gets 2 points added to their individual grade, 80-89 gets 1 point added, and <80 gets zero points added. I keep the group size no less than four, no more than six, as this seems to allow the best level of collaboration. I have seen many benefits to this practice. First, there are no more arguments with me about exam item rationales, since all the arguing is done during the collaboration process. Second, they are able to put into practice an important skill they will need in their professional life. Third, the students state they gain a greater understanding of the content and test taking strategies from the collaborative process. Finally, as an educator, I thoroughly enjoy the excitement and the interactions between the students, particularly when I see that usually quiet, introverted student confidently advocating his or her position during the discussions.

Pamela Rutar | February 21, 2014

I have used different forms of collaborative testing in my classes, primarily with quizzes. I have found it to be a valuable form of formative evaluation, not only for the students, but for myself.
The information I gather from the collaborative quizzing gives me feedback on student understanding of concepts that need to be revisited as well as those that are well understood. It also offers the opportunity for students to become engaged with the material in a different way. Some students like it and some don't. However, as I have also had grades suffer due to group work, I always offer the opportunity for the student to turn in their own quiz work. I encourage discussion and collaboration on items of which they are unsure or have difficulty selecting answers.
I still use summative evaluation in the form of exams, where students do not collaborate. I think there is value to both methods.
I have not tried the random assignment to pairs, mostly because of the size and configuration of our classrooms. I do think it would address the idea of student preparation and may give it a try.


Trackbacks

  1. There are no trackbacks to this post yet.

Add a Comment

Logged in as . Logout »


website security