Browsing by Author "Johnsson, Tommy"

Now showing 1 - 4 of 4

A stepwise regression procedure applied to an international production study - a multiple inference solution
(University of Gothenburg, 1997-04-01) Johnsson, Tommy; Ivarsson, Inge
The aim of this paper is to apply a certain procedure for stepwise regression analysis to a problem included in a study of integrated international production. The question, which factors are important for the transnational company's trade behaviour in certain respects, have been analysed by means of traditional methods. Hence the more interesting findings on the subject are likely to be reported already and the results here are only to be regarded as supplementary ones. Comparisons between old and new outcomes, and to some extent the interpretations of them, is however one point here.
HOUSEHOLD MARKET AND NON-MARKET ACTIVITIES - DESIGN ISSUES FOR A PILOT STUDY
(University of Gothenburg, 1982-02-01) Johnsson, Tommy
The design of a pilot survey of households is the subject of this paper. A few methods to collect data about consumption expenditures and time-use are to be compared. Among the design issues are the allocation of the sample on experimental groups, on strata and on days of the week.
MULTIPLE COMPARISON TESTS BASED ON THE BOOTSTRAP
(University of Gothenburg, 1986-01-01) Johnsson, Tommy
A multiple test procedure for pairwise comparisons based on the bootstrap is presented. It is a stagewise test without any distributional assumptions. It is also very general according to the number and types of hypotheses to be tested. The procedure is evaluated and to some extent compared to existing procedures. A FORTRAN computer program is available for the practical performance of the procedure suggested.
On statistics and scientific thinking
(University of Gothenburg, 1999-08-01) Johnsson, Tommy
Scientific thinking could in a wider scope be regarded as the process resulting in scientific knowledge. Such a definition is however at the same time to narrow and to wide since you could easily find projects ruled by the maxims of science not producing any knowledge what so ever, as well as established scientific facts emerging from activities far from the aims and scoops of science. A better attempt to define the scientific way of thinking is perhaps to look at it as a mode of operation or the using of certain tools. In both cases with a certain aim or goal before your eyes. Doing so you could include activities not yielding any new knowledge, even if they were supposed to, as well as activities and operations performed in certain ways in order to reach specified goals others than creating knowledge. A scientific approach, in acting and thinking, could payoff even in a non-scientific context and thus the process of scientific thinking could be regarded as a kind of useful kit of tools or universal technique. The aim of this paper is to discuss the role of statistics in the process of scientific thinking. Especially when it comes to the drawing of conclusions, mainly in a scientific way, and the creation of new knowledge. Such a topic could be regarded as the interface between statistics and the theory of knowledge, or epistemology, but even if the purpose here is to put forth some questions, without necessarily answering them, which deal with that borderline one should note that the total coverage of that topic is far beyond the scope of this paper. The objective is merely to form the basis for a discussion of some issues where statistics meet, or possibly do not meet, the processes of scientific thinking and creation of knowledge. Parts of the background are here formed by the concepts used by Karl Popper in his model of the research process. As the paper deals with, at least parts of, the interface between the theory of knowledge and statistics, the disposition is quite obvious. Hence the next section, the second that is, gives a brief overview of some traditional ways to knowledge, mainly kept at the operational level without any metaphysical discussions, while section three deals with some basic concepts of statistics, especially the ones connected with statistical inference or the drawing of conclusions, and eventually the scene is set for the meeting which is discussed in section four as the epistemology of statistics. The outcome of that meeting and some related issues are commented upon in the fifth and last section.