Mathematical models applied to subjective appraisals in environmental engineering
Stainsby, A. G. (1975). Mathematical models applied to subjective appraisals in environmental engineering. (Unpublished Doctoral thesis, The City University)
Abstract
Mathematical models appropriate to different types of psychophysical measurement are considered. It is stressed that the model should arise out of what is believed to be the type of observation made by the observer, This is in contrast to the type of model which is introduced simply because its equations can be made to fit (rather than explain) the data. In considering the type of model required, consideration must be given to the type of scale likely to arise. This consideration lends to a subdivision of S.S. Stevens' ratio scale into two classes, here called the geometric and arithmetic classes. It is shown that models based on these two scales will explain some of the differences noted by Stevens between his prothetic and metathetic
continua.
The model used by the Author in analysing categorical observations of differences in a published paper, is discussed in greater detail than in the original paper and the method of analysis is extended to include variability of the observer in two dimensions i.e. on the colour surface, It is shown that average values of observations almost identical to those obtained for variability along only one axis are predicted, but the distribution of scores between categories is different, Comparison of the predictions with actual observations indicate that the bivoriate distribution is an model for some colours.
The model used for analysis of colour differences is put forward as one appropriate for analysis of observations near to the limit of observation, arid a theory is proposed. Five published papers are included in which models have been used in the solution of environmental engineering problems. One relates to street lighting, the others relate to colour rendering of lamps.
Download (15MB) | Preview
Export
Downloads
Downloads per month over past year
Metadata
Metadata