MATCH-UP 2024 7th International Workshop on Matching Under Preferences, Oxford, 9 - 11 September, 2024

 Here's the announcement and call for papers of the latest edition of the Match-Up series of conferences.

MATCH-UP 2024   7th International Workshop on Matching Under Preferences 

University of Oxford, United Kingdom   9 - 11 September, 2024

 "MATCH-UP 2024 is the 7th workshop in an interdisciplinary and international workshops in the series on matching under preferences. It will take place on 9 - 11 September 2024, hosted by the University of Oxford, United Kingdom.

"Matching problems with preferences occur in widespread applications such as the assignment of school-leavers to universities, junior doctors to hospitals, students to campus housing, children to schools, kidney transplant patients to donors and so on. The common thread is that individuals have preference lists over the possible outcomes and the task is to find a matching of the participants that is in some sense optimal with respect to these preferences.

"The remit of this workshop is to explore matching problems with preferences from the perspective of algorithms and complexity, discrete mathematics, combinatorial optimization, game theory, mechanism design and economics, and thus a key objective is to bring together the research communities of the related areas.

"List of Topics

"The matching problems under consideration include, but are not limited to:

• Two-sided matchings involving agents on both sides (e.g., college admissions, medical resident allocation, job markets, and school choice)
• Two-sided matchings involving agents and objects (e.g., house allocation, course allocation, project allocation, assigning papers to reviewers, and school choice)
• One-sided matchings (e.g., roommate problems, coalition formation games, and kidney exchange)
• Multi-dimensional matchings (e.g., 3D stable matching problems)
• Matching with payments (e.g., assignment game)
• Online and stochastic matching models (e.g., Google Ads, ride sharing, Match.com)
• Other recent applications (e.g., refugee resettlement, food banks, social housing, and daycare)

Course on Matching, in Barcelona, by Péter Biró and David Manlove, April 27-28

If you are a Ph.D. student interested in matching, here's a two day course in Barcelona, by experts in the field.

V SEIO Course on Game Theory   

The University of Barcelona, the BEAT Research Institute, and the Game Theory and Assignment Markets Research Group are delighted to host the V SEIO Course on Game Theory on April 27 and 28, 2022.

The course is targeted at PhD students and early career researchers working in areas related to game theory. Besides covering a very active research topic, it is also an opportunity to meet with other researchers working in similar areas. 

The two-day course will cover algorithmic and game theoretic aspects of matching markets. Participants are welcome to present their game theory related research during a poster session.

The course will be delivered by Péter Biró (Head of  the Mechanism Design Group at KRTK) and David Manlove (Professor of Algorithms and Complexity at the University of Glasgow).

Registration deadline:  April 1, 2022  Registration is free, please fill in the on-line form to register.

Day 1 (April 27)

 09:00-09:30 Registration and welcome session

09:30-10:20 Stable Marriage and Hospitals / Residents problems: classical results

10:20-10:50 Coffee break

10:50-11:40 Decentralised matching markets, path-to-stability results

11:50-12:40 Hospitals / Residents problem: extensions (ties, couples, lower quotas)

12:40-14:10 Lunch

14:10-15:00 Hungarian university admissions: matching with contracts, choice functions, cutoff

stability

15:00-15:30 Coffee break

15:30-16:20 Housing markets: exchange of indivisible goods

16:30-17:20 Respecting improvement property for housing markets

Day 2 (April 28)

 09:00-09:50 House Allocation problem: Pareto optimal, popular and profile-based optimal matchings

10:00-10:50 School choice and constrained welfare-maximizing solutions

10:50-11:50 Coffee break and poster session

11:50-12:40 Stable roommate problems

12:40-14:10 Lunch

14:10-15:00 Matching with payments, auctions

15:00-15:30 Coffee break

15:30-16:20 Kidney Exchange

16:30-17:20 Generalized matching games, international kidney exchange

Matchup 2022: matching and market design in Vienna, August 25-2

 Here's the announcement and call for papers

Matchup 2022 

MATCH-UP 2022 is the 6th workshop in an interdisciplinary and international workshop series on matching under preferences . It will take place on 25-26 August 2022, hosted by TU Vienna and co-located with MFCS 2022 (47th International Symposium on Mathematical Foundations of Computer Science).

Matching problems with preferences occur in widespread applications such as the assignment of school-leavers to universities, junior doctors to hospitals, students to campus housing, children to schools, kidney transplant patients to donors and so on. The common thread is that individuals have preferences over the possible outcomes and the task is to find a matching of the participants that is in some sense optimal with respect to these preferences. There has been a resurgence of activity in this area in recent years, with online and mobile computing opening up new avenues of research and novel, path-breaking applications.

The remit of this workshop is to explore matching problems with preferences from the perspective of algorithms and complexity, discrete mathematics, combinatorial optimization, game theory, mechanism design, and economics. Thus, a key objective is to bring together the research communities of the related areas. Another important aim is to convey the excitement of recent research and new application areas, exposing participants to new ideas, new techniques, and new problems.

List of Topics

The matching problems under consideration include, but are not limited to:

• Two-sided matchings involving agents on both sides (e.g., college admissions, medical resident allocation, job markets, and school choice)

• Two-sided matchings involving agents and objects (e.g., house allocation, course allocation, project allocation, assigning papers to reviewers, and school choice)

• One-sided matchings (e.g., roommate problems, coalition formation games, and kidney exchange)

• Multi-dimensional matchings (e.g., 3D stable matching problems)

• Matching with payments (e.g., assignment game)

• Online and stochastic matching models (e.g., Google Ads, ride sharing, Match.com)

• Other recent applications (e.g., refugee resettlement, food banks, social housing, and daycare)

Invited Speakers

• Sophie Bade (Royal Holloway College, University of London)
• Vijay Vazirani (University of California, Irvine)

Match-Up 2019 in Switzerland, May 26-29

Here's the program. (It contains abstracts following the schedule...)

Sunday, May 26th
14:00 - 14:15    Conference Opening (Bettina Klaus)
14:15 - 14:40“Stability Against Robust Deviations in the Roommate Problem”Daisuke Hirata, Yusuke Kasuya, andKentaro Tomoeda
14:40 - 15:05“Robust   Group   Strategy-Proofness”Steven   KivinenandNorovsambuu Tumennasan
15:05 - 15:30“Robust Design in Monotonic Matching Markets: A Case for Firm-Proposing Deferred-Acceptance”Lars Ehlersand Jordi Masso
15:30-16:00Coffee Break
16:00 - 17:20    Poster Presentation Session 1
“Preprocessing  in  Matching  Problems”  Maxence  Delorme,  Ser-gio  Garc`ıa,  Jacek  Gondzio,  Joerg  Kalcsics,  David  Manlove,  andWilliam Pettersson
“Legal  Assignments,  the  EADAM  (Efficiency  Adjusted  DeferredAcceptance   Mechanism)   Algorithm”   Yuri   Faenza   andXuanZhang
“A General Framework for Stable Roommate Problems: A Preliminary Report”Muge Fidan and Esra Erdem
“Preference Manipulation in Two-Sided Matching - Strategic Behavior and Robustness of Solution Algorithms” Christian Haas
“Practical Issues in Matching - A Case Study on Genetic Counseling Admissions in North America” Jonah Peranson
“Unpopularity Factor in the Marriage and Roommates Problems”Suthee Ruangwises and Toshiya Itoh
“School choice with priority levels:  Constrained Efficient and FairAssignment” Thomas Wouters

Monday,May 27

09:10 CSF Welcome Address

09:25 - 10:10    Invited  Talk“Parameterizing  Stable  Matching  Problems” Ildi Schlotter

10:10 - 10:35“Stable   Noncrossing   Matchings” Suthee   Ruangwises and Toshiya Itoh

10:35 - 11:05    Coffee Break

11:05 - 11:30 “Refugee Resettlement” David Delacretaz, Scott Duke Kominers, and Alexander Teytelboym

11:30 - 11:55 “Matching Problem of Civil Service”Ashutosh Thakur

11:55 - 12:20 “Trading  Networks  with  General  Preferences” Jan  Christoph Schlegel

12:20 - 13:45    Lunch

13:45 - 14:30    Invited Talk “International Kidney Exchange Programmes:  Op-timisation and Games”Peter Biro

14:30 - 14:55 “Pareto Optimal Coalitions of Fixed Size” ́Agnes Cseh,  Tam ́as Fleiner, and Petra Harjan 

14:55 - 15:20 “Balanced Stable Marriage:  How Close is Close Enough?” Sushmita Gupta, Sanjukta Roy, Saket Saurabh, and Meirav Zehavi

15:20 - 15:50    Coffee Break

15:50 - 16:15“Flexibility in House Allocation and Housing Markets” Madhav Raghavan

16:15 - 16:40 “Endowment Manipulations in Probabilistic Assignment Problem”Yuki Tamura

16:40 - 17:05“Equivalent   Choice   Functions   and   Stable   Mechanisms” Jan Christoph Schlegel

17:05 - 17:15    Short Break (no coffee)

17:15 - 17:40 “Centralized  Matching  with  Incomplete  Information” Marcelo Ariel Fernandezand Leeat Yariv

17:40 - 18:05“Simultaneous  Search:   Beyond  Independent  Successes” Ran  I.Shorrer

18:05 - 18:30 “Deferred Acceptance and Regret-free Truth-telling:  A Character-ization Result” Marcelo Ariel Fernandez

 Tuesday, May 28

09:00 - 09:45    Invited  Talk “Balanced  Exchange  in  a  Multi-Object  Shapley-Scarf Market” P ́eter Biro,Flip Klijn, and Szilvia Papai

09:45 - 10:10 “Competing for Priorities in School Choice” Greg Leo and MartinVan der Linden

10:10 - 10:35 “Information  Acquisition  Costs  in  Matching  Markets”  Nicole  S. Immorlica, Jacob D. Leshno, and Irene Y. Lo

10:35 - 11:05    Coffee Break

11:05 - 11:30 “Efficient and (Pretty) Fair Course Assignment with Quotas” Martin  Bichler,  Alexander  Hammerl,  Thayer  Morrill,  and StefanWaldherr

11:30 - 11:55 “An Algorithm for Strong Stability in the Student-Project Alloca-tion Problem with Ties” Sofiat Olaosebikan and David Manlove

11:55 - 12:20“Strategy-Proof  Approximation  Algorithms  for  the  Stable  Mar-riage  Problem  with  Ties  and  Incomplete  Lists”,  Koki  Hamada, Shuichi Miyazaki, and Hiroki Yanagisawa

Wednesday, May 29th

09:00 - 09:45    Invited  Talk “Efficient  and  Incentive-Compatible  Liver  Ex-change” Haluk Ergin, Tayfun Sonmez, andM. Utku Unver

09:45 - 10:10 “Matching  for  the  Israeli  “Mechinot”  Gap-Year  Programs:  Handling Rich Diversity Requirements” Yannai A. Gonczarowski, Lior Kovalio, Noam Nisan, and Assaf Romm

10:10 - 10:35 “Recourse in Kidney Exchange Programs” Valentin Bartier, Yves Crama, Bart Smeulders, and Frits C.R. Spieksma

10:35 - 11:05    Coffee Break

11:05 - 11:30 “Obvious Dominance and Random Priority” Marek Pycia and Peter Troyan

11:30 - 11:55 “Subgame Perfect Equilibria under the Deferred Acceptance Algo-rithm” Keisuke Bando andYasushi Kawase

11:55 - 12:20 “Optimizing Reserves in School Choice:  A Dynamic ProgrammingApproach” Franklyn  Wang,  Ravi  Jagadeesan,  and  Scott  Duke Kominers

12:20 - 13:45    Lunch

13:45 - 14:10 “Strategy-proof,  Envy-free  and  Pareto  Efficient  Online  Mecha-nisms for Fair Division with Additive Valuations” Martin Aleksandrov and Toby Walsh

14:10 - 14:35 “An Alternative Approach to Asylum Assignment” Gian Caspari

14:35 - 15:00 “Matching  with  Myopic  and  Farsighted  Players” Jean-Jacques Herings, Ana Mauleon, and Vincent Vannetelbosch

15:00 - 15:30    Coffee Break

15:30 - 16:15    Invited Talk “Pareto Optimal Allocation under Uncertain Pref-erences” Haris Aziz,  Peter Bir ́o,  Ronald de Haan,  and Baharak Rastegari

16:15 - 16:40    CSF Award and Algorithms Award

16:40 - 18:00    Poster Presentation Session 2A

ll papers as in Poster Presentation Session 1 (except for Jonah Peranson and Christian Haas).

19:00Dinner

Martin Shubik, 1926-2018

Ed Kaplan emails me with the sad news that Martin Shubik died yesterday.

Here's the first brief announcement from Yale School of Management:
Prof. Martin Shubik, Influential Game Theory Scholar, Dies

He was a pioneering game theorist, and a frequent collaborator with his graduate school roommate Lloyd Shapley. My understanding is that the two of them shared a double room, in a suite with John Nash. 

He suffered from a rare disease, Inclusion Body Myositis, and established a charity to organize research about it, Inclusion Body Myositis Registry at Yale.

He was a man of many parts. (See here, for example.)

Here are two photos I took of him at a Stonybrook conference in honor of Shapley. (In the second he must have been proving an especially difficult theorem...)

Martin Shubik in 2003

Two papers by Shapley and Shubik played important roles in areas in which I've worked:

The second (by publication date) was their landmark 1971 paper on matching as an assignment game (with all payments freely transferable), published in volume 1 number 1 of the International Journal of Game Theory: The assignment game I: The core

(Years after it was published, I asked Shapley what ever happened to part II, and his reply was "Never call a paper part I unless you have already written part II."  As I recall, he further said that the plan for the never-written part II had been to study the von Neumann-Morgenstern solutions of the assignment game.)

The first was their famous 1954 paper in the American Political Science Review, perhaps Shubik's most cited, on how to evaluate the strength of each position in "simple" coalitional games, in which every coalition is either 'winning' or 'losing'
A Method for Evaluating the Distribution of Power in a Committee System

I have many times used his model of escalation, The Dollar Auction Game, as an in-class demonstration of the importance of auction rules for auction outcomes and strategies.

Historians of game theory are sure to learn a lot from the archives of his papers and correspondence at Duke:
The Martin Shubik Papers: From Early Game Theory to the Strategic Analysis of War
************
Update: Yale SOM has now published a long, fond remembrance:
Remembering Prof. Martin Shubik, 1926–2018

Duke switches to random roommate matching

Is it a good idea to allow college students to choose their own roommates?  Duke University returns to administrative matching of first year roommates, with Dean Random making the match.

Inside Higher Ed has the story (and the url makes for an informative headline: https://www.insidehighered.com/news/2018/03/02/duke-university-blocks-students-picking-their-roommates-freshman-year 

Here's the actual (pithier) headline and subhead
Random Roommates Only
"Duke University takes away from first-year students the ability to pick their roommates. This move goes against recent trends -- and raises questions about diversity, tolerance and the college experience."

"Duke University has removed from students what has become one of the most significant aspects of matriculation at many colleges: picking a first-year roommate

"Beginning with the Class of 2022, the roommate-selection process will be entirely governed by the university, with assignments largely made at random -- a shift, officials said, meant to stem the recent movement of students self-selecting peers with similar perspectives and backgrounds to their own, fueled by social media connections made before arriving on campus."

Match-Up 2017, call for papers

MATCH-UP 2017

Overview

Date: April 20 2017 - April 21 2017

Location: Cambridge, MA, USA

MATCH-UP 2017, the fourth workshop in the series of interdisciplinary and international workshops on matching under preferences, will take place on April 20-21, 2017.

Venue:Microsoft Research New EnglandCambridge, MA 02142

Registration: TBA, February 2017

Contact us: If you have any questions regarding this event please send email to MATCHUP2017@microsoft.com

• About
 
• Submissions
 
• Schedule
 
• Local Information
 
• Committees
 
• Previous Workshops

Submissions

We solicit papers and posters on matching under preferences.  The matching problems under consideration include, but are not limited to:

• Two-sided matchings involving agents on both sides (e.g., college admissions, medical resident allocation, job markets)
• Two-sided matchings involving agents and objects (e.g., house allocation, course allocation, project allocation, assigning papers to reviewers, and  school choice)
• One-sided matchings (e.g., roommate problems and kidney exchanges)

Paper Submissions

Authors should indicate which format type their paper should be considered under. The proceedings will take the form of informal working notes distributed at the workshop and online.

Format A

• Original contribution, not previously published in (or accepted by) another conference proceedings or journal
• Not under review for a conference or journal elsewhere
• At most 12 pages (including bibliography) using 11 point font or larger with at least 1″ margins all round, and in single-column format. Any material beyond this limit should be placed in a clearly marked appendix which will be read at the discretion of the programme committee
• Accepted papers will be published in proceedings (however, this should not prevent the simultaneous or subsequent submission of contributed papers to other workshops, conferences or journals)

Format B

• Not necessarily original work
• Can have been published already in (or accepted by) another conference proceedings or journal
• Can be under review for a conference or journal elsewhere
• No page limit
• Only the abstract will be published in proceedings

Information on submitting papers:

• Submission site: EasyChair
• Paper submission deadline : Friday, December 9th, 2016
• Selected papers will be notified at the beginning of February, 2017

Poster Submissions & Submission Site

• TBA, 2017

The Econometrics of Matching Models by Pierre-André Chiappori and Bernard Salanié in the JEL

The Econometrics of Matching Models
Pierre-André Chiappori and Bernard Salanié
Journal of Economic Literature 2016, 54(3), 832–861

Abstract: Many questions in economics can be fruitfully analyzed in the framework of matching models. Until recently, empirical work has lagged far behind theory in this area. This review reports on recent developments that have considerably expanded the range of matching models that can be taken to the data. A leading theme is that in such two-sided markets, knowing the observable characteristics of partners alone is not enough to credibly identify the relevant parameters. A combination of richer data and robust, theory-driven restrictions is required. We illustrate this on leading applications.

Here is the opening paragraph:
"In October 2012, the Nobel prize was attributed to Al Roth and Lloyd Shapley
for their work on matching. Both the seminal Gale and Shapley (1962) paper and most of Roth’s work were concerned with allocation mechanisms when prices or other transfers cannot be used—what we will call nontransferable utility (NTU) in this survey. Gale and Shapley used college admissions, marriage, and roommate assignments as examples; Roth’s fundamental work in market design has led to major improvements in the National Resident Matching Program (Roth and Peranson 1999) and to the creation of a mechanism for kidney exchange (Roth, Sönmez, and Ünver 2004). While these are important economic applications, matching problems are much more pervasive. Market and nonmarket mechanisms such as auctions match agents with goods and buyers with sellers; agents match to each other in production teams, and production tasks are matched with workers; and in international trade, countries are matched with goods or varieties. Yet while the basic theory of matching was in place forty years ago, only recently has there been an explosion of empirical work in this area. Several developments have concurred to bring it to the attention of applied researchers."

Montreal conference on matching ( l’appariement), March 21-22

Here's the preliminary program...

Colloque CIREQ Montréal sur l’appariement

Programme préliminaire
 

---

VENDREDI, le 21 mars 2014
9:00 – 10:30	SESSION I – Président : Umut DUR (North Carolina State University) Peter Coles (ebay Research Labs), Yannai Gonczarowski (Hebrew University),Ran SHORRER (Harvard University) Strategic Behavior in Unbalanced Matching Markets Timo MENNLE (University of Zurich), Sven Seuken (University of Zurich) An Axiomatic Approach to Characterizing and Relaxing Strategyproofness of One-sided Matching Mechanisms Umut DUR (North Carolina State University), Onur Kesten (Carnegie Mellon University) Sequential versus Simultaneous Assignment Problems and Two Applications
10:30 – 11:00	Pause
11:00 – 12:30	SESSION II - Président : Utku UNVER (Boston College) Lars Ehlers (Université de Montréal, CIREQ), Alexander WESTKAMP (Maastricht University) Strategy-Proof Tie-Breaking David Cantala (El Colegio de México), Szilvia PAPAI (Concordia University, CIREQ) Reasonably and Securely Stable Matching Marek Pycia (UCLA), Utku UNVER (Boston College) Incentive Compatible Allocation and Exchange of Indivisible Resources
12:30 – 14:00	Lunch
14:00 – 16:00	SESSION III - Présidente : Bettina KLAUS (University of Lausanne) Shuhei Morimoto (Kobe University), Shigehiro SERIZAWA (Osaka University) Strategy-proofness and Efficiency with Non-quasi-linear Preferences : A Characterization of Minimum Price Walrasian Rule Jens GUDMUNDSSON, Helga Habis (Lund University) Assignment Games with Externalities Tomoya KAZUMURA, Shigehiro Serizawa (Osaka University) An Impossibility of Allocating Objects among Agents with Non-Quasi-Linear Preferences Bettina KLAUS (University of Lausanne), Jonathan Newton (University of Sydney) Stochastic Stability in Assignment Problems
16:00 – 16:30	Pause
16:30 – 17:45	KEYNOTE LECTURE : Tayfun SÖNMEZ (Boston College) Kidney Exchange : Past, Present, and Potential Future
19:00	Dîner de conférence (sur invitation seulement)Restaurant Laloux, 250, Pine Avenue East

SAMEDI, le 22 mars 2014
9:00 – 10:30	SESSION IV – Président : Jan Christoph SCHLEGEL (Université de Lausanne) Madhav RAGHAVAN (Indian Statistical Institute, New Delhi) Fair Assignment with Capacity Constraints Takashi AKAHOSHI (Waseda University) Singleton Core in Many-to-One Matching Problems Jan Christoph SCHLEGEL (Université de Lausanne) Contracts versus Salaries in Matching : A General Result
10:30 – 11:00	Pause
11:00 – 12:30	SESSION V – Président : Panos TOULIS (Harvard University) M. Bumin Yenmez (Carnegie Mellon University), Federico ECHENIQUE (California Institute of Technology) How to Control Controlled School Choice Péter Biró (Hungarian Academy of Sciences), Elena IÑARRA (University of the Basque Country), Elena Molis (University of Granada) Solutions Concepts for Unsolvable Roommate Problems David C. Parkes (SEAS, Harvard University), Panos TOULIS (Harvard University) A Random Graph Model of Multi-Hospital Kidney Exchanges
12:30 – 14:00	Lunch
14:00 – 16:00	SESSION VI – Président : Vikram MANJUNATH (Université de Montréal, CIREQ) * Chaque présentation de cette session sera de 20 minutes.   Ville KORPELA (Turku School of Economics) Nash Implementation of Stable Solutions in Many-to-One Matching Problems : The Case with No Complementarities Samson ALVA (University of Texas at San Antonio) Pairwise Stability and Complementarity in Many-to-One Matching Mustafa Oğuz Afacan (Sabanci University), Bertan TURHAN (Boston College) Assignment Maximization Lars EHLERS (Université de Montréal, CIREQ) TBA Alexander TEYTELBOYM (MIT) Stability and Competitive Equilibrium in Networks with Bilateral Contracts Vikram MANJUNATH (Université de Montréal, CIREQ), Bertan Turhan (Boston College) When Two Should Be One but Aren’t : Milwaukee Public Schools and the Voucher System
16:00 – 16:30	Pause
16:30 – 18:00	SESSION VII – Président : Nizar ALLOUCH (Queen Mary University of London) Claudia HERRESTHAL (Oxford University) Mobility, Learning and School Choice Albin Erlanson (Lund University), Karol SZWAGRZAK (University of Southern Denmark) Strategy-proof Package Assignment Nizar ALLOUCH (Queen Mary University of London) The Cost of Segregation in Social Networks
18:00	Réception de clôtureColloque CIREQ Montréal sur l’appariement

Matchmaking on a plane

On those long overseas flights, picking a seatmate is a bit like picking a roommate, and KLM is on the case: Mile-high matchmaking: airline to let you choose your neighbour via Facebook

"The "meet and seat" service would allow passengers to see the Facebook or LinkedIn profiles of other flyers, who are also using the opt-in service, when selecting their seat."

HT: Ben Greiner

The roommate problem on TV

James Boudreau writes to alert us to some matching theory on television:

"The NBC show ``Community'' chronicles the misadventures of a diverse and wacky group of students at a community college.  On the last night's episode (season 3, episode three, ``Competitive Ecology'') the group was confronted with the problem of dividing into pairs for the purpose of being lab partners.  When their initial pairings don't work out, one member of the group realizes that they are in a classic roommates problem and suggests that they re-match by writing down lists of ordinal preferences and submitting them to one member of the group who is unanimously selected  as the matchmaker.  Unfortunately, the algorithm that the matchmaker uses (which focuses on balancing popularity across the pairs) proves to be unstable--eventually the group is forced to share one set of equipment since they can not agree on pairings.

"The episode is currently available for free on Hulu.  The most relevant scene begins at about 8:20.  Later on, around 14:33, one member of the group even accuses others of strategically manipulating their preferences to suggest that he is unpopular. "

Judd Kessler on the roommate problem in 2002

In response to my recent posts on roommate matching for college students, Judd Kessler points me to a newspaper column he wrote in 2002, on roommates, peer effects, and the Harvard housing system: More Than A Bunkmate.

Matchmaking at Harvard, the roommate problem

The Crimson reports, In Choosing Roommates, Deans Become Matchmakers.

"The day is June 10, and Resident Dean of Freshmen Sue Brown embarks upon what will be a five-week quest: sorting 398 first-years into rooms in Grays, Matthews, and Weld Halls.


"Much of this task remains the same as it has been in years past. She sorts freshmen largely by hand, thoroughly reading incoming freshmen’s responses to roommate questionnaires. During the process, her floor is covered with a mass of paper that is eventually divided into nine distinct piles—one of the early steps in the lengthy matchmaking process.

"But in the midst of that process, technology rears its head. In Adobe Acrobat, Brown categorizes freshmen, using colors to sort the incoming students by geography and stamping the freshmen’s rooming surveys with various logos made in Photoshop. A cartoon of a chess piece marks a chess player. A compass is reserved for the “curious and adventurous” freshman, Brown says.

"Emblazoned with colors and cartoons, the surveys are printed out and scattered on a floor in Brown’s residence in Weld Hall. From there, she divides the students into three categories based on students’ self-reported levels of sociability.

"Social tendencies are one of the most important factors in determining roommate compatibility “because it’s going to be something that comes between roommates before they even meet each other,” she says.

"But other factors matter as well, Brown adds. From the sleeping habits freshmen keep and the sports they play, to the neatness they maintain and the number of roommates they want, the resident deans take into account a slew of determinants of each first-year’s personality.

"Some factors, like musical tastes, are often stronger predictors of compatibility than others, according to Brown.
...
"Resident deans often try to place local freshmen with those from outside the area, especially international students for whom going home during vacations is more difficult, according to Dean of Freshmen Thomas A. Dingman ’67.
...
"Every year, a small margin of freshmen—”about half a dozen”—are switched out of their original suites after approaching the FDO, Dingman says. Many more choose to block with and live with others in future years.

...
"After organizing the rooms, the deans turn to assembling the entryways.
“You want it to be a dIverse group, a microcosm of the freshman class,” Brown says.
But at the same time, the entryway can’t be so diverse that a student feels isolated. For example, the FDO tries to avoid placing only one international student in any entryway, Brown says. She also avoids putting two students from the same varsity sport or the same high school in a single entryway."

Roommate matching and the social internet

Roommate matching continues to evolve, with Facebook playing a key role: Students Turn to Online Roommate Matching Services to Avoid Getting Paired With a Stranger (HT: Mike Ruberry)

The Chronicle gives a good description: Colleges Use Facebook to Let Freshmen Find Their Own Roommates
"This summer, incoming freshmen at five universities can use a Facebook application to find their roommates. Students can use the application, RoomBug, to fill out forms about their preferences for living and qualities they'd like to see in a roommate. Students can then request a match, which the other incoming freshman must confirm.
RoomBug is hardly the first service to let students match themselves: Tulane University announced a partnership with online service RoommateClick two years ago.
But RoomBug, which the company U-Match LLC just rolled out at Emory University, the University of Florida, Temple University, Wichita State University, and William Paterson University of New Jersey, tries to go where the students are.
"Everyone is on Facebook," says Robert Castellucci, the service's co-founder and sales director.
Over a quarter of the University of Florida's incoming freshmen have added the Facebook application, he says, but numbers on how many matches the service helped make are not yet available."

Maureen Dowd regrets the trend, writing in her NY Times column
"The serendipity of ending up with roommates that you like, despite your differences, or can’t stand, despite your similarities, or grow to like, despite your reservations, is an experience that toughens you up and broadens you out for the rest of life.

"So I was dubious when I read in The Wall Street Journal last week that students are relying more on online roommate matching services to avoid getting paired with strangers or peers with different political views, study habits and messiness quotients."

Roommate matching

"Male or female? Smoker or nonsmoker?
Those two questions have long been the basis for the University of Arizona’s roommate-pairing formula. But a year ago the university decided to give incoming students seeking deeper compatibility another option: shopping for their roommates on the Web."


That's the first paragraph of an interesting story at Inside Higher Ed about using the web to improve roommate matching: Match at First Site .


"Arizona is one of a small but growing number of colleges that have contracted with RoommateClick, a service that lets students take the lead on a task that has historically fallen to campus housing officials by browsing and communicating with future classmates who have also signed up for the service. Students who have hit it off online can then request to bunk with each other."


For a different market, see my earlier post Housemate match .

Housemate match

A matching service run by the Atlanta JCC: Housemate Match

"Housemate Match (HMM) is a unique, nationally recognized, home sharing program that matches mature adult homeowners who have extra room in their homes with adults (tenants) seeking a roommate in a beautiful and safe place to live in the Atlanta area.
HMM provides rooms to rent for those who prefer to share a home rather than living alone and for those who choose to remain in their home and age in place.
International friends are welcome. "