Abstract: | Statistical methods have the potential of being effectively used by industrial practitioners if they
satisfied two criteria: functionality and usability.
Statistical methods are usually the product of statistical research activities of universities and other
research organizations. Some already satisfy these criteria; however, many do not. The effect is that
potentially relevant methods are not used in practice as often as they could be. In this paper we will
present an approach regarding ‘statistics development,’ in which the end-user is given a central
position, so that the results from statistical research aim to meet the needs and requirements of the practitioner.
Examples of known and new methods will be presented, and we will discuss issues such as education in statistics,
the link with statistical consultancy and publication of methods through various channels. |