Order effects on children's moral and achievement judgments.

Abstract

JACQUELYNNE E. PARSONSSmith CollegeDIANE N. RUBLE, ELLEN CHERESKIN KLOSSON,NINA S. FELDMAN, AND WILLIAM S. RHOLESPrinceton UniversityThroughout the literature on the developmentof moral judgments and in a more recent exten-sion of this literature to achievement judgments(Weiner & Peter, 1973), it has been reported thatpreschoolers base their judgments more on out-come than on intent. However, outcome has gen-erally been the second, and thus the most recent,cue in the stimulus statements. Studies using bothfree-recall and information-processing paradigmshave reported that younger children recall andweight the most recent cues more than earlieroccurring cues (Cole, Frankel, & Sharp, 1971;Kun, Parsons, & Ruble, 1974). Consequently,the results reported in studies of moral andachievement judgments may reflect the order ofstimulus presentation rather than the youngerchildren's propensity to focus on concrete ratherthan subjective cues.Twenty-four white children, 12 male and 12female, from a Long Beach City school kinder-garten class served as subjects. Each child waspresented with 24 stories varying on the actor'sintent, the outcome of the actor's behavior, andthe competitiveness of the situation (half thestories depicting noncompetitive or moral di-lemmas and half depicting competitive orachievement dilemmas). Each story was accom-panied by a pictorial representation of the actor'sintent and the outcome. Half the children re-ceived a story booklet in which the outcome in-formation followed the intent information (Order1), and the remaining 12 subjects received book-lets in which the outcome cue preceded the intentcue (Order 2). The children were asked toevaluate the actor by administering rewards orpunishments in each of the 24 stories on an 11-point graphic scale: -6 = maximum punishment,0 = no reward or punishment, +6 = maximumreward. In summary, there were three within-subject variables: intent (three levels), outcome(two levels), competitiveness (two levels), andone between-subject variable: order of cues (twolevels).A mixed-design analysis of variance revealedthree significant interactions involving theorder-of-cues factor: Order of Cues x Outcome,F(l, 18) = 10.945, p < .01; Order of Cues x Com-petitiveness x Intent, F(2, 36) = 5.84, p<.01;Order of Cues x Intent x Outcome, F(2, 36) =3.74,p<.05.With regard to the Order of Cues x Outcomeinteraction, the rewards for positive outcomesdecreased when outcome was the first (M = .79)rather than the second (M = 2.67) cue. Thepunishment for negative outcomes was not af-fected by the order-of-stimulus-cue presentation(Order 1 M = -2.42, Order 2 M = -2.74). Thus,order had an effect, but only on the evaluativejudgments associated with positive outcome.Possibly, this Order of Cues x Outcome interac-tion reflects a differential developmental timeta-ble for the decreasing importance of the outcomecues. Costanzo, Coie, Grumet, and Farnill(1973) and Parsons (1974) found that the relianceon positive outcome cues for allocating rewardsdecreased earlier, developmentally, than the re-liance on negative outcome cues for allocatingpunishment. If the evaluative importance of posi-tive outcomes declines at an earlier age than theevaluative importance of negative outcome, thenone would expect the effects of situational cues toemerge first for positive outcomes.The means associated with the significantOrder of Cues x Competitiveness x Intent in-teraction are shown in Table 1. The meansTable 1: Mean Evaluations as a Function of CueOrder, Competitiveness, and Actor's IntentRequests for reprints should be sent to JacquelynneE. Parsons, Clark Science Center, Smith College,Northampton, Massachusetts 01060.IntentOrder Positive Neutral Negative