S D H T Speed-dating as a methodological...

Angeles named Yaacov Deyo provided the answer: speed-dating. In speed-dating,romantically eligible individuals attend an event where they have a chance tomeet all the attendees of the sex that theyromantically prefer. Each date lasts just a few minutes, and the attendees use theirquickly generated impressions to decidewhether or not they would (‘yes’) orwould not (‘no’) be interested in seeingeach of their speed-dates again. Typically,mutual yesses (‘matches’) are then giventhe ability to contact each other after theevent to further explore any romanticpossibilities.

At first glance, it might seem thatindividuals would only be able to learnshallow or trivial information about apotential romantic partner in just a fewshort minutes (e.g. how attractive he is).However, this notion is contradicted byan avalanche of evidence demonstratingthat individuals can make remarkablysophisticated social judgements based on‘thin slices’ of social behaviour lasting fiveminutes or less (Ambady et al., 2000).Perhaps unknowingly, the rabbi invented

speed-dating by applying this socialpsychological gem to a romanticcontext. If participants are able toaccurately judge teachingeffectiveness (Ambady & Rosenthal,1993) or personality (Kenny, 1994)within minutes, they might just asquickly glean accurate informationregarding romantic compatibility.Thus, speed-dating essentially lies atthe intersection of person perceptionresearch and popular culture (Finkel& Eastwick, in press). Researcherscan harness the power of speed-dating to do high-quality, high-impact research while at the sametime providing a rewarding andenjoyable experience for participants.

The basic structure of a speed-dating studyA speed-dating study typicallyconsists of three parts. First, as

individuals sign up to participate,

E very night, researchers whoinvestigate relationships and personperception miss out on great

opportunities. Millions of parties andsocial gatherings take place throughoutthe world, and no one is there to measurethe interpersonal dynamics taking place in these real-world environments. Asresearchers of romantic attraction, wethink about all the single people meetingeach other for the first time – chatting fora few moments, deciding whether or notthey’d like to get to know each otherbetter – and we areheartbroken not tobe observingunobtrusively.

We find somesolace by remindingourselves that partiesand bars are notexactly perfectresearchenvironments. It istrue that people atparties can oftenform realrelationships withreal futures, and thisexternal validitymakes such socialgatherings idealsources of data onreal-life matingbehaviour (Eastwick& Finkel, in press). But at a party,there are many

uncontrolled factors that weaken internalvalidity. For example, not all romanticallyeligible people have a chance to meet oneanother, and some people can getcornered for half the evening stuck in a dreadful conversation. How terrificwould it be if there existed a type of socialgathering with just a bit more structure;something that romantically eligibleindividuals would want to attend, but that would also permit data collection and experimental control?

About a decade ago, a rabbi in Los

402 vol 21 no 5 may 2008

Ambady, N., Bernieri, F.J. & Richeson,J.A. (2000). Toward a histology ofsocial behavior: Judgmental accuracyfrom thin slices of the behavioralstream. In M.P. Zanna (Ed.) Advancesin experimental social psychology (Vol.32, pp.201–271). San Diego, CA:Academic Press.

Ambady, N. & Rosenthal, R. (1993). Half aminute: Predicting teacherevaluations from thin slices of

nonverbal behavior and physicalattractiveness. Journal of Personalityand Social Psychology, 64, 431–441.

Bolger, N., Davis, A. & Rafaeli, E. (2003)Diary methods. Annual Review ofPsychology, 54, 579–616.

Eastwick, P.W. & Finkel, E.J. (2008). Sexdifferences in mate preferencesrevisited: Do people know what theyinitially desire in a romantic partner?Journal of Personality and Social

Psychology, 94, 245–264.Eastwick, P.W. & Finkel, E.J. (in press).

Speed-dating: A powerful and flexibleparadigm for studying romanticrelationship initiation. In S. Sprecher,A. Wenzel & J. Harvey (Eds.)Handbook of relationship initiation.New York: Guilford.

Eastwick, P.W., Finkel, E.J. Mochon, D. &Ariely, D. (2007). Selective versusunselective romantic desire.

Psychological Science, 18, 317–319.Finkel, E.J. & Eastwick, P.W. (in press).

Speed-dating. Current Directions inPsychological Science.

Finkel, E.J., Eastwick, P.W. & Matthews,J. (2007). Speed-dating as aninvaluable tool for studying romanticattraction: A methodological primer.Personal Relationships, 14, 149–166.

Fisman, R., Iyengar, S.S., Kamenica, E. &Simonson, I. (2008). Racial

Speed-dating as amethodological innovation Paul W. Eastwick and Eli J. Finkel argue that this unusual data collection technique is high in external and internal validity

refe

renc

esM

ETHO

DS

GU

YBELL/P

HOTOFU

SION

the researcher will want to assessbackground information about each ofthem. Using a paper-and-pencil or onlinequestionnaire, researchers can assessdemographic, personality and attitudemeasures. Time permitting, researchersmight also assess baseline levels ofbiomarkers such as cortisol and alphaamylase, and, in women, contraceptionuse and menstrual cycle phase.Researchers can readily customise thebackground information they collectdepending on their research interests.

Second, thereis the speed-dating event itself.Commercialspeed-datingcompanies will tryto recruit manyindividuals toattend each event,thus maximising revenue and,consequently, the total number of speed-dates. However, there are several reasonswhy researchers might want to restrictevent attendance, perhaps limitingparticipants to a dozen dates or so. For one, some evidence suggests thatparticipants have a better speed-datingexperience when they go on a moderatenumber of dates rather than a largenumber (Fisman et al., 2006). Secondly,with the time freed up by including fewerdates, researchers can administer aquestionnaire at the end of each date toassess participants’ impressions of eachspeed-dating partner. Thesequestionnaires provide an invaluablesource of data that go beyond the simpleyes/no response and allow for dataanalysis using Kenny’s (1994; Kenny &La Voie, 1984) Social Relations Model(e.g. Eastwick et al., 2007) – see box. Ifresources are available, researchers mightalso wish to take photographs ofparticipants or to audio- and video-recordthe speed-dates themselves.

Third and finally, researchers canfollow up with their participants in thewake of the speed-dating event. In somecases, researchers will be content to assess

who has said yes to whomand which matches havesubsequently contacted oneanother. However, very littleresearch has explored thespan of time between aninitial encounter and theformation of a romanticrelationship. Therefore, weencourage researchers toconsider administering oneor perhaps severalquestionnaires in the wake

of the speed-datingevent. Using adiary-type format(Bolger et al.,2003), researcherscan assessinformation about

participants’ lives ingeneral as well as

specific details about theirrelationships with eachspeed-dating match.

This discussion isintended to provide a generaloverview of speed-datingprocedures. Elsewhere, wehave provided an extensivehow-to manual forresearchers planning toconduct their own speed-dating research,reviewing issues such as recruitmentstrategies, institutional review boardconcerns, payment considerations, anduse of the internet (Finkel et al., 2007).

Recent research Several different research teams have usedspeed-dating in recent years to explore awide variety of topics. For example, wehave used speed-dating to explore men’sand women’s preferences for a romanticpartner’s physical attractiveness andearning prospects (Eastwick & Finkel,2008). In contrast to the large corpus offindings in which participants report theirstated preferences for these twocharacteristics in a romantic partner, ourresults revealed no sex differences in the

importance that participants placed onphysical attractiveness and on earningprospects at and following a speed-datingevent. Other researchers have also notedthat stated mate preferences don’t predictactual choices of dating partners, but theydo find that women are more selectivethan men at speed-dating events (Todd et al., 2007; see also Kurzban & Weeden,2005). Speed-dating also offers anopportunity to study interracial datingdynamics: for example, individuals aremore likely to prefer same-race overinterracial speed-dating partners if theygrew up in a location characterised bystrong opposition to interracial marriage(Fisman et al., 2008).

In short, speed-dating presents anexcellent opportunity for researchers tostudy a variety of topics related tointerpersonal relationships. Furthermore,straightforward extensions of speed-dating (e.g. speed-networking) also helpto broaden the scope of this method. Weanticipate that many scholars will findspeed-dating to be a usefulmethodological tool.

read discuss contribute at www.thepsychologist.org.uk 403

methods

I Paul Eastwick and Eli Finkel are in theDepartment of Psychology, NorthwesternUniversity, Evanston, Illinois, USAp-eastwick@northwestern.edufinkel@northwestern.edu

“speed-dating lies at theintersection of personperception research andpopular culture”

preferences in dating. Review ofEconomic Studies, 75, 117–132.

Kenny, D.A. (1994). Interpersonalperception: A social relationsanalysis. New York: Guilford.

Kenny, D.A. & La Voie, L. (1984). Thesocial relations model. In L.Berkowitz (Ed.) Advances inexperimental social psychology (Vol.18, pp.139–180). San Diego, CA:Academic Press.

Kurzban, R. & Weeden, J. (2005).Hurrydate: Mate preferences inaction. Evolution and HumanBehavior, 26, 227–244.

Todd, P.M., Penke, L., Fasolo, B. &Lenton, A.P. (2007). Differentcognitive processes underliehuman mate choices and matepreferences. Proceedings of theNational Academy of Sciences, 104,15011–15016.

The Social RelationsModel in actionIn speed-dating, participants meet multiple individuals and are met by multiple individuals. This design permitsresearchers to make use of the Social Relations Model(SRM), should they choose to administer a questionnaireabout each speed-date.

Imagine that such a questionnaire included a measureof romantic desire. SRM distinguishes between threeindependent reasons why Maria might desire Trey:I Compared to the other female speed-daters on

average, Maria might desire all of her speed-dates. In this case, Maria would have a strong actor effect.

I Compared to the other male speed-daters on average,Trey might be desired by all of his speed-dates. In thiscase, Trey would have a strong partner effect.

I Maria might desire Trey above and beyond her actoreffect and his partner effect. In this case, Maria wouldhave a strong relationship effect with Trey.

The ability to calculate relationship effects is one of theexceptional features of SRM. In the above example, therelationship effect assesses the unique romantic desirespeed-daters experienced on each date; this unique desireis similar to the colloquial concept of ‘romantic chemistry’.