MobLab Valentines

A Date to Remember: How MobLab Can Find You Love

We are showered with love everyday from both instructors and students. However, this one particular story, recounted by Texas A&M instructor Noah Bacine, leads us to believe that perhaps MobLab can help you find love as well.


Let me preface this story with my aside about chat. Student’s absolutely love the chat feature, it may cause a lot of noise in your data, but it is very effective in keeping students actively engaged. That being said, I never know what will happen when I turn on the chat feature in these games. I would like to believe that telling the students that I can see the chat would lead them to focus solely on the game but in many cases it only seemed to exacerbate the desire to say off the wall things.

My favorite MobLab chat story is about one student trying to use the in-game chat feature in a partner’s game to get a date. Dictator partner’s were anonymous and there was a pretty uneven split in groups so that there was way more males in the class then females. This did not deter one student who offered the entire pie in exchange for a phone number. It did not matter to them that they were talking to a stranger who had a good chance of not even being their desired gender or that I would be able to see all of this afterwards. Luckily, the other student was a good sport and gently tried to push the student back towards playing the actual game. I wonder what kind of story they’ll tell their future kids about how they met.


Happy #EconValentines. Stay tuned for our full interview with Noah!

Competitive Market: Updates

The Competitive Market Game (Continuous Double Oral Auction) is a workhorse and played by over 50 percent of instructors to help highlight a variety of market related learning objectives. Therefore, it is big news to many users whenever we make modifications. In the most recent server update there were two modifications to the Competitive Market Game:

First, we have added an “Average” row in our tables. This is part of an effort to include measures of central tendency across all of our games. This will allow instructors to quickly compare the central tendency of markets to the equilibrium benchmark.

competitive_market_average_row

Second, we have added a new treatment “Buyer Subsidy” that can be used to illustrate the effects of  buyer subsidies in housing, education, healthcare, etc.. We hope many classes will find this useful.

Buyer Subsidy

We are always looking for new ways to make the platform better for our instructors. We hope these changes improve the effectiveness of your teaching with games and allow you to cover more of the areas you want to explore with students. Happy Playing!

 

kevin_mccabe

Conversations with Educators: Kevin McCabe

This past summer I had the great pleasure to meet with Kevin McCabe, Professor of Economics and Law at George Mason University. His research has appeared in journals such as the American Economic Review, Science, Economic Theory, Games and Economic Behavior, and many more. He is the Co-Director of the Center for the Study of Neuroeconomics at George Mason and he is in a small group of people who could potentially win the Nobel Prize in Economics for his pioneering work in Neuroeconomics. Beyond being a top notch scholar he is a terrific person. He and his wife Kathleen were great hosts to me on my visit last summer and gave me a detailed explanation of their work with virtual worlds and related outreach. They have some wonderful things going on!

Q: Your research has primarily used experimental methods to study economic problems. How did you become interested in using experiments in research?

A: In 1985, as an assistant professor at the University of Arizona, I became interested in experiments after attending a seminar by Vernon Smith. I spent a year learning the Plato computer language and working on the design of my first experiment on fiat money as a store of value. After coming back from a fellowship at Washington University in St. Louis, I joined the Economic Science Laboratory to work with Steve Rassenti and Vernon Smith on ‘smart’ markets for Natural Gas and ‘smarter’ auction designs for two-sided call markets. At this point I was pretty much convinced that the scientific method was a great way to study economics and I realized that the combination of theory, experiments, and computational thinking was a perfect fit for me.

Q: Were you doing classroom experiments before you started in your experimental research or was there a moment where you transitioned from experiments for research to experiments for education?

A: A requirement for my fellowship at Washington University, in 1986-87, was to teach a micro class. The class had very smart students, but they had never experienced how markets work. I decided the answer to this lack of experience was to run market experiments which I programmed on my luggable Osborne computer. It was a great way to read the literature and attempt to replicate research in the classroom and I found students enjoyed this learning by doing approach. When I got back to Arizona I taught a monetary course built around the design and testing of monetary institutions. For me the enjoyment was connecting monetary theory to experiments. For the students the enjoyment was asking what will happen as opposed to hearing about what does happen. Since then I’ve used experiments in every class I teach.

Q: Other than the Osborne computer what technologies have you used in the past for classroom experiments?

A: I’ve used hand run double auctions, public goods, bargaining, and game theory experiments. I moved early on to more sophisticated experiments where I could program the institutional rules and environmental conditions, first on an Osborne computer and later, on laptops. In these experiments the messages were still collected by hand and input into the computer. I also started using networked computers running a simple database in BASIC and then VISUAL BASIC and later using the common shared data format available in FORTRAN. Later, I used the PLATO system, pioneered by Arlington Williams, which was designed to efficiently use the idea of a shared data space. Currently, I program laboratory and web experiments in Python and Virtual World experiments in the Linden Scripting Language. At some point, all my research experiments and results end up in the classroom. I also use Charlie Holt’s Veconlab which implements many experiments by other researchers. Through Veconlab, Holt has helped pioneer the availability of web based parameterized demonstration experiments.

Screenshot 2017-01-16 15.06.41

 

Q: When did you start using MobLab and what attracted you to it?

A: In spring 2015 I decided to teach a Managerial Economics course for undergraduates. At the time I had just learned about MobLab at a conference and decided to use it in my course. One major attraction was the shift in student access to technology in the classroom. Today every student has a smart phone, tablet, or laptop of some type. MobLab had been designed to work with all of these devices making it an exceptional teaching tool. A second reason I wanted to use MobLab was to introduce a more fluid way of moving back and forth between theory, experiment, and practice. MobLab has short video instructions that allow the instructor to run a demonstration experiment in less than 15 minutes and it provides useful data displays for immediate discussion. Third, Moblab has a survey tool which I use instead of a clicker system to ask in class questions to gauge student comprehension.

Q: Now that you have used MobLab a while, what do you like about it as an instructor? What have the student reactions been?

A: The value of MobLab is the ease with which experiments can be designed and run on many different devices. MobLab has an instructor interface that makes it relatively simple to manage students and experiments.   The data from experiments can be downloaded as a comma separated file with anonymized decisions which can then be made available to students for further analysis. Most students report that they like MobLab experiments in the classroom. MobLab also seems to act as an icebreaker opening more students interest in participating in classroom discussion.   I am also very impressed with the continuous improvements and additions made by the MobLab team, their customer support when problems occur, and their willingness to listen to their customers.

Q: Do you have a favorite MobLab game to play with students?

A: MobLab has many of my favorite experiments making it difficult to choose just one. What I like a lot is the ability to run a sequence of experiments to create a narrative for students. For instance when we talk about how one sided auctions work students first participate in a first price auction followed by a second price auction followed by an English clock auction. They can see how changing the rules of the auction affects bidding behavior. In the next class they are put in common value environments to see how changes in the environment affect bidding behavior. We can then discuss optimal bidding strategies and the importance of expertise in different auctions and auction environments.

Q: What advice would you give a new instructor who wants to use games in their classroom?

A: My main advice is to use experiments to explore how institutions and culture work to shape incentives and coordinate actions to produce different outcomes. In doing this I would highlight the lessons being learned by the research in economic science. Most of the experiments in MobLab have been replicated many times. Acquaintance with the research literature behind the experiment will prevent the students or the instructor from telling ‘just so’ stories about the data. One feature of MobLab to keep in mind is that classroom experiments are likely to be run without monetary incentives while the research behind the experiment was done with monetary incentives. Without monetary incentives, some experiments in MobLab will not replicate the behavior produced in incentivized experiments.   The only way to avoid misinterpretation of the experimental data is to be able to compare what students did without monetary incentives to what subjects did with monetary incentives.

 

5 Tips to Starting the Semester Off Right

We’ve built our platform to be easy as pie. But, we still have some advice about using MobLab well. After all, the technology is a complement and not a substitute to good teaching. So how can you get the most out of MobLab?

Tip 1: Run an out-of-class survey for at least 48 hours before your first in-class use.

You don’t want to stall the learning and fun of your first MobLab use. Use our surveys to generate some pre-class questions. You can use our pre-class survey which contains questions on student motivation, course interest, anticipated grade, and more. Conversely, you could design your own survey to assess math background, learning styles, syllabus comprehension, etc..

Tip 2: Preparation is the key to success

We want you to feel confident when you run our games in class. The best way to feel confident is to prepare. It doesn’t need to be much time (20-30 minutes) and if using our modules preparation can be even easier:

  • Watch the video instructions or read the instructions. This is a starting point to familiarize you with the decision situation and the screen elements students need to keep in mind.
  • Read through the module guide. Better yet, print it out and highlight/annotate parts of the script (enlarge graphs before printing). In particular, it might help to
    1. Highlight important actions you need to take,
    2. Annotate the example graphs and ideas to prompt discussion,
    3. Make notes about the reflection questions
  • Preview the game. The preview feature will improve your understanding of game flow. Once you feel you grasp the game click End Game and press View Results to ensure you feel comfortable presenting the results.

Note: If you do not use the modules you can use the Guide button to look through information. However, the modules might still be useful to give you ideas about how to think through the presentation of results and some of the debriefing questions you might ask. Also, you can ALWAYS contact us at MobLab if you have any questions in preparation.

Tip 3: Reference the game later on in class

MobLab games and surveys are a fraction of what you do in class. However, the games are powerful at anchoring student understanding. Hearken back to the game with phrases such as, “This is like …” or “Remember when …” and so on. This also helps improve the continuity of games in the overall course.

Tip 4: Use our MobLab Experience Survey

Learning is about feedback. Just like students need feedback, we as educators need feedback. Use our MobLab Experience Survey to gauge how well a given game-lecture pair worked together to improve student understanding. You can run these after class and give students 24 hours to complete them so this feedback doesn’t eat into your class time.

Tip 5: It’s all a learning opportunity

Sometimes strange things happen. Don’t say things like, “Normally […] happens but for some reason […] didn’t happen.” If you get odd results in a game make some lemonade out of those lemons! Ask students, “Tell me about how you made the decisions you did.” It will open up the conversation and help you to identify where misunderstandings might be happening, incentives might have gone awry, or otherwise. No matter what it will produce some teachable moments.

module_guide

MobLab Modules

You may have noticed three tabs in the instructor console: Games, Surveys, and Modules. In our Year in Review 2016 we wrote about forthcoming modules, “These modules are pre-built playlists with instruction comprehension questions, an A/B test (e.g. market followed by market with intervention), and final reflection questions. We are also pairing these with detailed procedures that estimate time for each segment of the module and provide instructors with ideas about how to present the results, questions to prompt students with, and more.”

modules_tab

Currently we have four modules posted in this tab. Within the next week or two many more will populate the Modules tab (e.g. Comparative Advantage, Convergence, Markets with Price Controls, and many more). Let’s turn to our Bank Run + Deposit Insurance Module. It includes a stack of surveys and games with reflection questions. One item that might not be self-explanatory is the Open Session (for more information, we have written about that here). If you click a specific module you not only see the stack of surveys and games but a Guide. 

the_stack

Clicking on the Guide you will see a standalone window. While there is an overview of the module and objectives there is significantly more detail than the guides you are accustomed to seeing. There is a timeline given to the instructor and detailed procedures on how to run the game in class and present the data to students. There is also a discussion of the reflection questions.

module_guide

 

We hope you use these modules and find them helpful. Thank you for your enthusiastic support of MobLab and please, please, please provide your feedback on them (or suggest new ones!). They will undoubtedly get better across time because of your help.

highway_2017

Year in Review: Where have we been? Where are we going?

What are the greatest stories? From classics like Iliad and Odyssey to modern stories like On the Road and To Kill a Mockingbird, ask the question, “What makes them great?” A hero on a journey. I think we like them so much because we are never more human than when we embark on a journey to achieve something outside our reach. In this post I outline our journey at MobLab and what we have been able to accomplish in 2016. Writing this has been an extraordinary reminder of how far we have come. Thank you for all of your help and joining us on this journey! I list in italics where we are planning to go. 

Improvements in Game Offerings

Here are the games that we have developed this year:

In the spring we know we will be developing the Beer Game, Monty Hall Game, Centipede Game, Simple Closed Economy (links labor and goods markets), and Principal-Agent game. Other possible games in development are to be determined.

Beyond new games we are also taking a critical look at all of our existing games. We are reviewing our game screens, summary results, and guides. As we review these we are asking the question of whether modifications can be made to improve student understanding, help prepare instructors, and help improve the clarity and usefulness of the graphs. In this effort your advice has been invaluable. Thank you to you and all the instructors who have given us feedback!

Improvements in the Instructor Console

The engineers are constantly working on items that you might not notice like improved performance but here we outline some of the larger changes to the instructor console. This list is not exhaustive:

  • Playlist Scheduler – With the playlist scheduler you can set the date and time to run whole playlists outside of class.
  • LMS Integration – We’ve made signing up through an LMS simple and added grade sending capabilities which include previewing and sending different types of grades to the LMS.
  • Standalone Windows –  Guides, instructions, and summary results all pull up in standalone windows that make refreshing easy. This makes the transition between instructor console and these pop-ups seamless.
  • Live Survey Results – A new standalone window that shows live survey results streaming in (will be released in our next update)
  • Enhanced copying features – You can now copy individual games/surveys to different playlists within the same class or across classes.
  • Improved organization of console – We separated our game library into Games and Surveys (this is laying the groundwork for our upcoming “Modules” – more on this later). We redesigned our playlist column, added the ability to sort through playlists, and more.
  • Improved Onboarding Process – Instructors can schedule a demo upon first sign up. We added computer-guided tours to improve understanding of the MobLab Instructor console.
  • Borda Points – This provides instructors with another option for how to assign points for classroom experiments.

In the next few months, what will our engineers be working on?

  1. Incorporating elements of Team-Based Learning into the MobLab console. For example, rather than a single individual making a decision that same decision would emerge from joint conversation between two players.
  2. Improve the student console. The students will have some brand new capabilities. For example, students will have access to post-game graphs and more.
  3. The engineers are also laying the groundwork for the Modules (discussed below), improving sharing capabilities to lay the groundwork for a MobLab community (discussed below), and adding a MobLab data dashboard to provide instructors with a quick overview of student performance on surveys and across games.

Improvements in our Instructor Support

  • Created Video Instructions – Most of our games currently have video instructions available. Instructors can play these videos at the start of class. Or post the URL their LMS for students to watch in advance of class.
  • Textbook-Based Game Guides – We have reviewed popular Principles and Intermediate level textbooks and gone chapter-by-chapter to make game and survey recommendations. For each recommendation we provide the learning objectives those games aim to highlight.
  • Best Practices – Of course instructors can always email us; however, we developed a host of Best Practices to provide advice on how to teach with MobLab games.

We plan to continue rolling out video instructions, textbook-based game guides, and best practices. But we have other big plans:

  1. In the next couple weeks we will release MobLab Modules. These modules are pre-built playlists with instruction comprehension questions, an A/B test (e.g. market followed by market with intervention), and final reflection questions. We are also pairing these with detailed procedures that estimate time for each segment of the module and provide instructors with ideas about how to present the results, questions to prompt students with, and more.
  2. These modules are just the beginning. We also aim to include edited online videos of instructors teaching with MobLab so our instructors can get a better visualization of how others around the world are teaching with our games.
  3. We will be developing an online community forums for MobLab instructors such that MobLab economists and instructors can share materials internal to their institution or to a broader community of instructors. We want this to be a place where instructors can problem solve, brainstorm, discuss pedagogy, and more. 

Other Improvements

We held our first student competition with games. Find the introduction to the student competition here. Congratulations to the University of Oklahoma for their first place finish!

We have also reached out to instructors through our blog on various teaching tips with games, interviews with instructors who use MobLab, and other fun pieces like this and this.

 

 

 

 

Introducing our Loan Market Game

Shout out to all my macro people! In this post we’re talking about our recently released loan market. But, I want to let you know, we are also designing more classroom experiments including a GDP accounting game and simple closed economy (with linked labor and goods markets). You called, we answered, and you can’t let the microeconomists have all the fun.

In this game students participate in a market for a fictitious good called “Pandorium”. With our default parameters there are twelve students in a market split evenly between borrowers and lenders. Regardless of role, each student has a productive capacity for Pandorium (i.e. how much Pandorium they can produce with $100) but producing Pandorium requires money. Only lenders have the $100 necessary to engage in production. There’s the rub. Differences in productive capacities lead to a situation where exchange is beneficial. The screens of both borrower and lender are shown below.

loan_market_game

Exchange occurs in a “uniform price double auction”: Lenders submit the minimum repayment amount at which they would lend $100 rather than produce with it. Borrowers submit the largest repayment amount at which they would borrow $100. Based on these submissions the Central Bank of our fictional mining town forms supply and demand and announces the equilibrium repayment amount. Loan transactions occur if (1) The announced price exceeds the lender’s minimum amount and (2) The announced price is below a borrower’s maximum amount.

The payoffs to borrowers and lenders are given as:

  • Borrowers: End-of-year Pandorium (after repayment)
  • Lenders: End-of-year Pandorium – 100

Note: All contracts are in nominal terms. Thus, borrowers sell their Pandorium haul at nominal prices to repay their loan in nominal terms. Once the loan is repaid lenders can then purchase Pandorium at those prevailing nominal prices.

When the experiment starts all students know the price of Pandorium is $1. The price announced after loans are made depends on inflation (in the screenshots above there is zero inflation). Here’s where our manipulations come into play. Students can participate in a market with known inflation and after that a market with their is some risk over inflation.

If you’re teaching a macroeconomics course next semester we hope your students enjoy this Loan Market game. Happy Playing!

Introducing our Double Marginalization Game

The concept of double marginalization can be tricky for students to grasp. How is it possible that mergers improve both industry profits and increase consumer welfare at the same time? In a newly developed game we explore this double marginalization problem. Students act as either a wholesale coffee roaster or a retail café (shown below). Wholesalers set a price per bag of coffee. Given that price retailers decide how much to buy.

vertical_integration_joined

Beyond experiencing the double marginalization problem we also make it so instructors can allow for the possibility of vertical integration or franchising.

With the vertical integration setting each round we allow for free form bargaining. The retailer and wholesaler can both submit offers to acquire each other. Offers can be accepted or declined. When an offer is accepted the acquired firm receives payment equal to the proposal and the merged firm faces market demand as a monopolist. When an offer is rejected there are opportunities to continue to make counteroffers. However, if no agreement is reached when time expires both the wholesaler and retailer will receive nothing.

In the franchise setting each round we allow the wholesaler to offer a two-part deal to the retailer. First, the wholesaler sets a one-time franchise fee. Second, the wholesaler sets a price for the retailer to purchase bags of coffee. The retailer can accept or decline any offer. When an offer is accepted  the franchisee chooses a quantity to purchase at the wholesaler’s price and also pays the one-time franchise fee. When an offer is rejected the wholesaler can make additional offers. However, if no agreement is reached when time expires both the wholesaler and retailer will receive nothing.

Below is some data generated from the game. Rounds 1-3 were no integration rounds. Rounds 4-6 allowed for vertical integration. Rounds 7-9 allowed for franchise contracts. In the graph below the orange line indicates the equilibrium retail price under no integration (i.e. double mark up). The fuscia line indicates the monopoly equilibrium price. In Rounds 1-3 students are trending towards the double mark up equilibrium. In Rounds 4-9 when mergers or franchises are allowed students pretty quickly find the equilibrium monopoly price.

vertical_integration_graph

Another one of our graphs provides equilibrium consumer surplus and industry profits (combined profits from wholesaler and retailer) benchmarks and graphs average behavior across rounds. The first graph shows the equilibrium consumer surplus and industry profits in the case of double mark up and highlights average data from Rounds 1-3. The second graph shows the equilibrium consumer surplus and industry profits when contractual solutions to the double mark up problem are permitted and highlights average data from Rounds 4-9. You can see tendency toward the equilibria in both cases.

double_mark_up_eq

monopoly_eq_double_mark_up

Those graphs really help to show students the counterintuitive result rather than just tell them about it. So please check out our Double Marginalization game in the instructor console and let us know what you think. Happy playing!

Prisoner’s Dilemma: Matrix Form

In October I had the good pleasure of demonstrating MobLab to a 400 student class at UNC.  That day students were learning about monopolistic competition and one component of the discussion was collusion. Naturally, we decided to play the Prisoner’s Dilemma (PD) — that old workhorse for  studying the conditions under which cooperation is more or less likely to occur. The students were really excited about doing classroom experiments. I even had students shaking my hand and telling me so in advance of class!

We played a sequence of three repeated PD. The first session was a five round game with fixed partners. In the second session students were randomly rematched with someone else for a five round game with communication.

pd_joined

 

In the left panel you can see the game begins with more cooperation than defection 28.6% C/C  (Green) v. 21% D/D (Red). However, C/C declines to 8.6% while D/D climbs to 56% by the final round. In the right panel with communication C/C starts at 46.5% and by the fourth round maintains at 44.4%. In the final round there is an end-game effect to C/C of 17%. In this case the D/D equilibrium climbs across time from 15.1% to 45.7%

With such a change in cooperation rates it definitely calls to mind that old quote from the Wealth of Nations where Adam Smith writes,

“People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices.”

Finally, after two sour experiences with partners in the PD experiment we randomly rematched students with new partners. This time we ran a PD with chat and indefinite repetition. With indefinite repetition and communication 78 out of 161 groups made it through the tenth round. After the tenth round observations start to drop significantly. Despite people learning from the broken promises or “cheap talk” in the PD with communication cooperation still started out rather high. C/C started at 38% in the first round. By the tenth round C/C was 25%, much higher than cooperation in final round of the game with the finite ending. As with the previous games, D/D climbs across time from 23.4% in the first round to 51.4% by the tenth round. The .csv data files contain chat transcripts which are usually an absolute gold mine. Some examples:

  • “I’m going to trust you on this”
  • “i gotta know if ur trustworthy”
  • “If you pick D one time I will pick D every time and we will both lose”
  • “you suckkkk”
  • “Lol I gotta do what I gotta do to get the victory”
  • “YOu do realize we get the most if we both choose C, right?”

pd_communication_indefinite

 

In subsequent in-class examples the instructor was able to build on student’s improved understanding of the conventions of the payoff matrix and strategic considerations that lead to the Nash Equilibrium. It was a blast! I had a lot of fun and the students got a lot out of it.