The present investigation concerns the prediction of nsktaking in games of skill and chance from direct and mdirect measures of achievement-related motives Risk-taking is assessed under both relaxed and achievement-oriented conditions in order to study the eflFect of motivation arousal on the relationships between risk-talong and Need for Achievement and Test Anxiety.Accordmg to Atkinson's theory of the motivational determinants of nsk-takmg behaviorThe strength of motivation to perform some act is assumed to be a multiphcative function of the strength of the motive, the expectancy (subjective probability) that the act will have as a consequence the attamment of an mcentive, and the value of the mcentive When both motivation to approach and motivation to avoid are simultaneously aroused, the resultant motivation is the algebraic summation of approach and avoidance The act which is performed among a set of alternatives is the act for which the resultant motivation is most positive (Atkmson, 1957, pp 360-361) Atkinson assumes that as the subjective probabihty of success (Ps) of a task becomes smaller, the mcentive value of success (Is) becomes greater (specifically. Is = 1 -Ps) The definition of Is as "the relative amount of satisfaction to be expenenced m any personal accomplishment" (Atkinson, 1958b, p 298) mdicates that the model is mtended to apply to tasks m which sucAduevement motives and nsk-taking 177 cess results from the exercise of skill or eflEort rather than purely from chance.An important consequence of the model is that for an alternative with Ps -50, resultant motivation is maximally positive for persons m whom the motive for success is stronger than the motive to avoid failure (Ms > Mf), and resultant motivation is maximally negative for persons m whom the motive to avoid failure is stronger than the mobve for success (Mf>Ms) Hence, it is expected that m a game or task m which Ss are free to choose among alternatives of varymg difficulty there vnH be greater preference for mtermediate nsk by those m whom Ms > Mf than by those m whom Mf > Ms This expectation has been borne out m studies by Atkmson, Bastian, Earl, and Litwm (i960), Atkinson and Litwm (i960), Litwm (1958), McClelland (1958), and Smith (1963 An attempt to check the hypothesis m Germany, Brazil, and India was only moderately successful (McClelland, 1961, pp 212-213) In the present experiment, as m previous studies, the projective measure of Need for Achievement (n Ach) is used as a measure of the motive for success and the Test Anxiety Questionnaire (Mandler & Sarason, 1952) is used as a measure of the motive to avoid failure In the studies applymg Atkinson's model to nsk-takmg m games of chance, the results have not been consistent Atkinson et al (i960) found support for the mtermediate risk hypothesis m both a shuffleboard game (skiU) and a task mvolvmg preference between lmagmary bets (chance) Specifically, for bets with a small expected value (30 cents) Ss with high n Ach showed a significantly greater preference for bets wit...