[Girls are more successful than boys at the university. Gender group differences in models integrating motivational and aggressive components correlated with Test-Anxiety].
It is surprising to note the evolution of success rates in Belgian universities especially in the first Year. Men are less successful than women and the differences are escalating in an alarming way. Dropouts take the same direction and women now represent a majority of the students at the university. In a previous study, we assessed 616 students in the first Year at the university of Liège with Vasev, the English name of which was TASTE, a self report questionnaire constituted of 4 factors: anxiety, self confidence, procrastination and performance value; anxiety particularly concerned somatic expression of students before and during test evaluations; self confidence was a cognitive component close to self efficacy; procrastination was the behavioral component characterizing avoidance when students are confronted with the risk of failure; performance value referred to intrinsic and extrinsic motivations. French validation of TASTE led to an abbreviated version of 50 items (THEE) consisting of 5 factors, the four of TASTE and an additional one, very consistent, at first called depression because of its correlations with this dimension, then called sense of competence on account of its semantic content. Self-competence has been described in the literature of Achievement Motivation and corresponded to expectancy and ability beliefs in performance process which was also relevant to self-efficacy except the particularity of comparison with others, which was not included in the last construct. Self-competence has been considered as an important part of the Worry component of test anxiety. Some Authors didn't hesitate to view causality flowing from self-competence to test anxiety and have conceptualized the latter as a failure of the self where one's sense of competence has been undermined as a result of experienced failure. In our study, only that factor was equally scored in women and men whereas it was scored higher in failed students. In other respects anxiety and performance value were scored higher in women, self-confidence and procrastination higher in men. Because TASTE didn't discriminate the different components of motivation (performance value referred to intrinsic and extrinsic motivations without precise distinction) we decided to use the MPS (Multidimensional Perfectionism Scale) which gave the opportunity to distinguish SOP (Self Oriented Perfectionism) ie, the self-imposed unrealistic standards with inability to accept faults in order to know and master a subject, that corresponded to intrinsic motivation; SPP (Socially Prescribed Perfectionism) ie, the exaggerated expectancies of others which are subjectively believed as imposed and uncontrollable leading to anxiety, feelings of failure or helplessness, that corresponded to extrinsic motivation; POO (Perfectionism Oriented to Others) ie, the unrealistic demands expected from significant others, which especially characterized males. We assumed that women attached more importance to succeed and submitted more to society exigencies. That way extrinsic and intrinsic motivations were probably more combined unlike men who, dreading a loss of self esteem, tried to avoid failure responsibility in using self handicapping or aggressive behaviours, so separating motivation in an extrinsic part turned to performance value and an intrinsic one more concerned by self confidence and sense of competence with the result that the motivational balance was surely disrupted in case of high competition leading to failure or avoidance. In another previous study we established a structural model illustrating, according to gender, correlations between anxiety, sense of incompetence, self-oriented perfectionism and socially prescribed perfectionism. Self-oriented perfectionism was less correlated to socially prescribed perfectionism in boys than in girls; furthermore especially by those who had never failed, it was negatively correlated to sense of incompetence, thus leading to lower scores of anxiety while in girls, by contrast, such a correlation didn't exist, thus involving higher anxiety. That way, on the one hand, intrinsic and extrinsic motivations by female students complementarily operated on the sense of incompetence and consequently on anxiety, the emotional component of test anxiety; on the other hand, by male students, intrinsic motivation had a negative correlation with the sense of incompetence and a lower correlation with extrinsic motivation, thereby shedding some light on the problem of anxiety level differences according to gender. More, that observation corresponded well to the model of self-worth where test anxiety was understood as a manifestation of perceived incompetence and as a defensive way to ward off negative self-evaluation; that model suited particularly well to boys and explained their attempts to maintain self-worth when risking academic failure. The present research assumes that independence or combination of motivation components is also correlated to different expressions of aggressiveness: hostility corresponding to threat and characterizing more girls while physical aggression is corresponding to personal challenge, a more masculine attribution. If fighting against the sense of incompetence actually characterizes men and consequently shows too the competitive aspects of performance strong enough to mobilize intrinsic motivation, what would be expected regarding the notion of threat suspected to be predominant in girls? The idea of using a questionnaire discriminating the specific dimensions of aggressiveness in fact the Aggression Questionnaire should meet the following purposes: At first establish a French version of that aggression questionnaire, perform the factorial analyses and internal consistency, compare them with other previous samples, then differentiate gender in general and in failure versus success situations. Finally include the different components of aggressiveness in the first described model and build a new one liable to define in boys the explicit pathways between test anxiety, perfectionism and aggressiveness. Statistical analyses have confirmed, in a 3 factor solution, the presence of emotional (anger), cognitive (hostility) and behavioural (physical aggression) components. Internal consistency is satisfactory. It is demonstrated that physical aggression characterizes boys (F=12.04, p=0.0001) while hostility (F=5.22, p=0.0015) and anger (F=0.49, p=0.0001) characterizes girls; furthermore it is noted that physical aggression characterizes more failed students (F=13.43, p=0.0003). Four models (see figures 2, 3, 4, 5) have been established, at first focused on the distinction of correlations between motivation and cognitive and emotional components on the samples of boys (n=268) and girls (348), then developed on the samples of successful students, male (n=193) and female (n=271). They describe the differentiated action of intrinsic and extrinsic motivations on the different components of aggressiveness and test-anxiety according to gender and without experience of failure. The dynamic process of the organizational factors is different according to gender and psychopathology resulting from the combinations of behaviors, cognitions and emotions would be assumed, prioritizing physical aggression and psychopathy by boys, anxiety and depression by girls. Anyway more explanation about the evolution of success rates of boys and girls in Belgian universities is proposed.