Empirical Software Engineering Experts on the Use of Students and Professionals in Experiments

Abstract

Controlled experiments are an important empirical method to generate theories. Many software engineering experiments are conducted with students. It is oaftnedn vclaaliidmaetde twhhaitl et hues iunsge p orfo fsetsusdioenntasls a dso peasr ntiocti.p [aOnbtjse cinti veex]p Wereim beenlitesv eco am deese paet rt huen dceorsstt aonf dloinwg eisx tneerendael dv oalnid tihtye epxrotefrenssailo nvaalsli.d Witye aiomf tsoo gftawina irnes igehntg ainbeoeurti nthge perxopse rainmde cnotns s ocof nudsiuncgt esdtu dwenitths asntdu dperonftess soiorn awlsi tihn seuxprveeryim. eFnirtsst. , [Mduertihnogd ]a Wsees spieornf oarmt eISdE aRnN u2n0c1on4v, e6n5t ioenmapl,i rfoiccaul s rgersoeuarpc haeprpsr, oiancchl uadnidng a tfhoell oswev-uepn awuet hroervsi,s iatregdu ethde a ntodp dici sacunsds eedli ctihteed u seex poef rsttsu’ doepnitnsi oinn se xtop efroimsteern tds iswciuthss aionn os.p eTnh emni nwde. Adfeterirvweadr d1s4, satgarteeemmeenntts wanitdh athskee dst attheem eISnEtsR. NF inatatlelyn,d ewees aenxacllyuzdeindg ththe er easuetahrocrhse, rsto’ opprionvioidnes tahnedi r ulseevde l thoef friensdpionngds etno tfsu drtishaegrr deiesdc uwsist hth ues s atabtoeumt ethnets d. [rRawesbualtcsk]s O ouf rp sruorfevsesyi orneasulsl.t sW seh, oowne tdh eth caotn, tinra greyn, setrraoln, tghlye ibne laiebvseo ltuhtaet tneor mpos.p u[Claotniocnlu s(siotund] eUnstisn, gp rsotfuedsesinotns aalss, opra rottichieprasn) tcsa rne bmea dinese ma evda lbide ttseimr tphliafnic aatnioonth eorf trheeaolirtyie sn eaendde dt eicnh nlaobloograietso ryb ucto, nltiekxet sa. nIyt iost haenr eaffsepcetcivt e owf asytu tdoy asdevtatinncges , soshftowualdre been gcianreeefruilnlyg considered during the design, execution, interpretation, and reporting of an experiment. The key is to understand which developer population portion is being represented by the experiment. Thus, a proposal for describing experimental participants is put forward