The quality of the requirement affects the work done in the later phases of the software development life cycle and, consequently, in the product. The requirements with poor quality increase the cost and the schedule of a project. Therefore, in recent years, different proposals have emerged on the elicitation of requirements. However, these investigations do not analyse how the activities contribute the elicitation process to obtain a "good requirement" and, consequently, in the quality of the product of software. Because getting a quality requirement depends on all the activities of the elicitation process as a whole, 54 relationships between the activities and the qualities of a good requirement were identified in this paper, to know which qualities must meet each activity. The relationships were obtained by analysing the qualities that each activity requires to obtain a "good requirement". To demonstrate these relationships, an empirical study was conducted on 128 respondents, whereby the analysis of Simple and Multiple Correspondence were analysed, showing that all relationships have a rating between "High" and "Very high". Thus, according to the test of the T-Student hypothesis, all proposed relationships are valid with 95% confidence.