Individual goals can be highly relevant in the process of helping parents to bring up their children, for instance in the field of youth welfare. On the one hand, they are able to translate the will of the clients into a desirable status, which is standard practice in the so-called concept of „social space focussed welfare“. On the other hand, they are predestined as a basis for evaluating the helping process by reviewing the attainment of such goals. An important requirement for both is a certain quality of the goals which is generally described by the acronym SMART (specific, measurable, assignable, realistic and time-related). However, up to now there is a shortage of empirical studies dealing with the systematic analysis of the above mentioned quality and accordingly of feasible methods for evaluating the effectiveness of the welfare on the basis of the attainment of goals. Hence, in this dissertation a study is presented to bridge that gap. Based on an elaborate description of both the „social space focussed concept“ and the essential principals of academic evaluation the quality of goals was measured by comparing ratings of different experts (for example inter-rater reliability). The most important result of the study is the fact that the quality of goals depends on the number of goals as well as the number of ratings. The more of them are involved in the process, the more accurate the measurement is. Therefore, the quality also seems to be more uncertain, the less goals and ratings of different experts are analysed („principle of uncertainty“). Thus, it is impossible to determine the quality of only one goal by only one rater. Adapted from these empirical findings a method for three-dimensional evaluation is finally drafted and reviewed regarding its potentials for practice and for the development of promising research projects, especially in the field of „social space focussed youth welfare“.