probably accomplishes the test’s goal ofnsurprising them with a correct answer.nHowever, at least as many (in mynexperience) are aware that a relativelynsmall percentage of welfare recipientsnare malingerers or cheaters. In fairnessnit should be said that this may not havenbeen true when the question was originallynwritten and, if we again wish to bengenerous, we might give this question anpassing grade.nQ. The income gap between blacksnand whites has narrowed in recentnyears?nA. Despite civil rights and othernlegislation, the income gap betweennblacks and whites has actually widenednin recent years; black workers are generallynless skilled than white workersnand less-skilled workers suffer more inntimes of depression.nComment: While the flaw here isnsomewhat less deep than that in thennext question (to which it is similar),nthe belief that seems to be under attacknis not the belief that this answer challenges.nThe strong belief of the studentnis that a black and a white in the samenposition earn more nearly equal incomesnthan they did formerly. Thisnbelief is correct. That blacks are thenfirst to be laid off in times of “depression”n(by which I assume is meantnrecession; there has not been a depressionnin fifty years) is a point worthnmaking, but hardly one that will surprisenany student who is sentient.nQ. The income gap between malenand female workers has narrowed innrecent years?nA. The income gap between malenand female workers has widened rathernthan narrowed; women hold few highpayingnpositions and the average workingnwhite woman earns less than thenaverage working black man.nComment: This is the test’s quintessentialnquestion. It elicits the “incorrect”nanswer from — and therefore surprises—nthe student by introducing anconstruct (“income gap”) with whichnthe student is unfamiliar. It therebynseems to capture and refute the strongnbelief on which the student bases hisn”incorrect” answer. However, thenstrong belief is that, after twenty yearsnof the women’s movement, a man andna woman occupying the same positionn(and having equal credentials) receivenmore-nearly-equal incomes than theyndid formedy. This belief is entirelyn58/CHRONICLESncorrect. In other words, to the student,nthe term “income gap” refers to a mannand a woman in the same position.nThe fact that the “income gap” referrednto by the question — the averagenincomes of all full-time male and femalenworkers — has increased does notnchallenge any deeply-held belief of thenstudent and is not likely to surprisenhim. It is particularly unlikely to surprisenhim when he realizes that then”income gap” referred to in the questionnhas increased as a result of thensame forces that have led to a decreasenin the difference in incomes of a mannand a woman in the same position:nwomen have joined the work force innlarge numbers and, since one joiningnthe work force earns an entry-levelnincome, these new female workersnbring down the average income ofnworking women and increase then”income gap” between the sexes.nQ. Human beings have a naturalninstinct to mate with the opposite sex?nA. Human beings do not have anninstinct to mate with the opposite sex..nOur sexual preferences are entirelynlearned; in fact, if an instinct is definednas an inherited complex behavior pattern,nhuman beings do not have anyninstincts at all.nComment: This answer commitsnthree errors in two sentences: 1) Whilenthe causation of heterosexuality andnhomosexuality is far from understood,nit has for a decade become increasinglynapparent that there is a physiologicalncomponent that interacts with environmentalnfactors to generate sexual direction.nI know of no researcher who anynlonger holds that the causation of sexualndirection is entirely environmental.n2) Even if one rejects this and goes sonfar as to see the issue as entirelynundecided, the question is a poor one,nfor the student feels — correctly — thatnhis answer is as likely to be correct as isnthat of the test. 3) As is so common innsociology, there is an attempt to instillnin the student’s mind a rejection of thenpossibility that physiological factorsnplay an important role in determiningnhuman behavior. The first-day studentnis hardly likely to know that, if “instinct”nis defined as requiring “complexnbehavior patterns,” there are stillnphysiologically rooted tendencies andnpredispositions other than “instincts”nthat cleady affect human behavior {i.e.,na predisposition is clearly involved innnnsexual arousal, even though this predisposesnone to actions whose specificncharacteristics are socially determined).nA severe critic might add a fourthnerror: it is questionable whether thenterm “learned” means much at thisnlevel of interaction of physiology andnenvironment; just as the term “causation”ntends to lose its meaning, or, atnthe very least, its fertility, at the quantumnlevel, so does “learning” lose itsnmeaning at deep levels.nQ. For religious reasons, mostnAmerican Catholics oppose birth controlnand are less likely than Protestantsnto enter interfaith marriages or to bendivorced.nA. More than 80 percent of AmericannCatholics favor birth control; ,(nCatholics are more likely than Protestantsnto enter interfaith marriages andnCatholics have a higher divorce ratenthan Protestants.nComment: A scratch single, a longnfly out, and a strikeout on three callednchange-ups.nIt is unclear from the Roper poll onnbirth control whether the 80 percentnfigure refers to Catholic approval ofnbirth control for everyone or just fornnon-Catholics. More important, it isnlikely that the beginning studentnequates “favors” with “uses,” and it is /’nonly Catholic usage that would surprisenthe student. (It is axiomatic innsociology that actual behavior oftennfails to reflect stated opinion.) Nonetheless,nit is probable that most beginningnstudents underrate Catholic usenof birth control.nThere is somewhat less than meetsnthe eye in the fact that a Catholic isnmore likely to enter an interfaith marriagenthan a Protestant. Most Catholicninterfaith marriages are to Protestantsnand most Protestant interfaith marriagesnare to Catholics. Since there arentwice as many American Protestants asnCatholics, it is hardly surprising that anhigher percentage of Catholics intermarryn{i.e., that a Catholic is “morenlikely” to intermarry).nThe “fact” about Catholics having anhigher divorce rate than Protestants isnas astonishing as it is untrue. Protestantsnhave a much higher divorce rate.nSteven Goldberg is the chairmannof the sociology departmentnat City College, City Universitynof New York.n