In a questionnaire survey of 119 psychologists currently undertaking doctorate degrees, the influence of social support, professional supervision and academic self-concept on the tendency to quit and the wish to pursue an academic career was examined. The final model indicates differences in effects on men and women dependent on the variable in question. Thus, the quality of the supervision shows coherence to the tendency to quit in women whilst it is connected to the wish to gain a professorship in the men. Moreover, social support does not affect the tendency to quit of women whilst it clearly is connected for men. For men the academic self-concept shows no connection to the wish to pursue a further academic career, whilst it does for women.

Just as in commercial organisations, long established disparities between the career opportunities of men and women can be observed within the academic community. At European universities, the proportion of women decreases continuously with increasing academic status ('horizontal segregation'). Two ways of leaving the university are separated: Firstly, more women than men tend to turn their backs on the scientific career immediately after acquiring a qualification, and secondly, women seem to be more likely to dropout from their doctoral studies. Several reasons for this attitude are discussed (e.g. Lind 2004), especially the quality of the professional supervision (e.g. Briede et. al. 2004, Falk & Berning 2004) and the social support of family, friends and the partner (e.g. Dres 1996, Mummendey 1996). Also, the self-concept of the postgraduates (e.g. Sieverding 1990, Super et. al. 1963) seems to be important.

The aim of this study is to examine what underpins the under-representation of women in higher positions within the German scientific community. This study intends to explore factors that influence women to leave the university after receiving their doctorate degree. Normally, it is assumed that men and women are influenced by the same factors in the same way and hence that it is only the difference in quantity of those factors that constitutes the disparity between the sexes. The author, conversely, holds the view that not only those quantitative differences matter but also important is the way each sex responds to a given factor (qualitative differences). The differences between men and women are not only determined by the amount of support received and self-concept, but the factors should also be of different importance to them, which in turn influences the dropout thoughts and plans for a further career.

The first study question is to discover if this study reveals any connections between social support, professional supervision and self-concept on the one hand and the tendency to quit and career planning on the other hand. The second question concerns the differences between men and women i.e. are those differences visible in the connections between social support, professional supervision, academic self-concept and the tendency to quit and career planning (quality)?

A questionnaire survey with 119 German psychologists (43 men and 76 women) currently undertaking doctorate degrees is presented. Five scales are used for the measurement of the input and outcome variables, all of which are of high statistical reliability. For the input variables the scales taken are social support (Holzbecher et. al., 2002), professional supervision (Holzbecher et. al, 2002) and academic self-concept (Schwanzer & Köller, 2002). The outcome variables are the tendency to quit and the wish to pursue a further academic career (Holzbecher et. al, 2002). Plans for career development are divided into the wish to attain a professorship and the general interest in pursuing a scientific career.

No differences are found in the quantity of social support, professional supervision and academic self-concept reported by postgraduate men or women. Also, there appears to be no difference between men and women in the occurrence of the tendency to quit and the wish to work as a scientist. In this sample, men are more interested in gaining a professorship than women. This is the only difference in the quantitative factors between the sexes.

Path diagrams are then calculated to explore the connections between the factors in the overall group. For all subjects, there appear to be connections between the professional support and the tendency to quit on the one hand, and professional supervision and the will to gain a professorship on the other. Social support is connected to the tendency to quit. Academic self-concept shows connections to the will to gain a professorship and the wish to work as a scientist.

If evaluated separately for men and women, the results of the path diagrams change: for the women the connections between professional supervision and the tendency to quit are only almost significant (p<.10). Also, connections between social support and the tendency to quit and between professional supervision and the will to gain a professorship are not significant. There are connections between academic self-concept and the will to gain a professorship and the wish to work as a scientist. For men, the connections between professional supervision and the tendency to quit, and between social support and the tendency to quit are statistically significant. The results are also significant for the connection between professional supervision and the will to gain a professorship. For men, there are no significant connections between academic self-concept and the wish to gain a professorship and the wish to work as a scientist.

There are differences in the connections between these factors for men and women in this sample. The desire to gain a professorship and to work as a scientist seems to be a greater expectation for men than for women. Women psychologists require a proportionally higher opinion of their own academic abilities to express the desire to pursue a further career at a university. From the perspective of supervision of postgraduates at universities, these results show important evidence for possible weak spots. In the course of the "Bologna Process", certain directions have been given for the supervision of postgraduates but the final destination is not obvious in many cases.