Choice of Strategy in the Analysis of Complex Data

Williams, W. T.; Lance, G. N.

doi:10.2307/2987164

Cited by 32 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, although a number of attempts have been made to evaluate the relative performance of methods at various points over the past 25 years (e.g. Williams & Lance 1968;Gauch & Whittaker 1972b;Dale 1975;Austin 1976;Fasham 1977;Goodall 1978;Whittaker & Gauch 1978;Brown et al 1984;Minchin 1987a), a clear and consistent consensus over which method or methods should be recommended for general use has never emerged. Rather, as each technique or group of techniques has evolved, it has usually been assumed to represent the best available.…”

Section: The Range Of Choice Of Methodsmentioning

confidence: 99%

Trends and problems in the application of classification and ordination methods in plant ecology

Kent¹,

Ballard²

1988

Vegetatio

View full text Add to dashboard Cite

Classification and ordination methods represent the two primary groups of ordering techniques for the analysis of floristic data in plant ecology. Current problems in the use and application of the methods are introduced and through a review of 734 articles across 11 ecological journals from 1960 to 1986, the history and evolution of the methods are displayed, the extent to which different techniques have been applied in the past and up to the present is demonstrated and problems and trends are discussed. A clear and substantial increase in the application of ordination and classification methods over the survey period is shown but with this increased usage, problems have come in terms of choice and evaluation of methods, the emphasis on inductive rather than deductive approaches, possible over-emphasis on methodology and technique rather than ecological application, low levels of use in applied studies and an increased tendency to use complementary and multiple analyses.All of the problems are related to broader aspects of the present position of vegetation description and analysis within vegetation science in general and in particular the relationship between community ecology and individualistic plant ecology.

show abstract

Section: The Range Of Choice Of Methodsmentioning

confidence: 99%

Trends and problems in the application of classification and ordination methods in plant ecology

Kent¹,

Ballard²

1988

Vegetatio

View full text Add to dashboard Cite

show abstract

“…The predictive attribute analysis of Macnaughton-Smith (Williams and Lance, 1968) failed to show high-order interactions that were helpful in predicting neurosis. Application of the programme MULTIST (Lance and Williams, 1967) to 72 qualitative attributes, 11 numerical attributes, and 9 multi-state attributes revealed no high-level discontinuity and suggested that the system had little structure.…”

Section: Impairmentmentioning

confidence: 96%

A study of neurosis and occupation

Ferguson

1973

Occupational and Environmental Medicine

View full text Add to dashboard Cite

A study of neurosis and occupation. Claims that male telegraphists in an Australian communications undertaking were unduly subject to neurosis and certain psychosomatic disorders as a result of the stress of their work were investigated by sickness absence and environmental and prevalence studies. The absence records of all telegraphists in the mainland capital city offices of the undertaking were compared with those of random samples of clerks and mechanics and, because of excess absence among Sydney telegraphists, with those of mail sorters in that city. Subsequently, 516 telegraphists, 93 % of those available in Sydney, Melbourne, and Brisbane, and 155 Sydney mail sorters (79 % of a sample) were examined medically.Absence attributed to neurosis was much commoner in telegraphists than in the other occupations in each capital, and in Sydney telegraphists than in those of other capitals. Employees having such absence were more likely than others also to have uncertified and repeated absences, and absence attributed to bronchial and dyspeptic disorder and to injury. One-third (33%) of the 516 telegraphists examined were considered to have or to have had disabling neurosis, the prevalence being much greater in Sydney (44%) than in Melbourne (19 %) or Brisbane (26 %O). The onset, course, associations, and other characteristics of neurosis are described.There was some evidence that the neurotic employee had increased liability to some other disorders but also that he was more likely to report ill health than others. Interpretation of increased other ill health in neurosis is confounded by the effects of an excess indulgence in habits. An increase in indices of mental stress was noted but some disorders commonly attributed to stress were not unduly prevalent in neurotics. Loss of craft status, monotony, dissatisfaction with job, fear of displacement by machine, group size, and supervisory practices were all thought to predispose to the high prevalence of neurosis in Sydney telegraphists. However, personal and social maladjustment was particularly evident in telegraphists in that city, and the population from which telegraphists were drawn may have been less well adjusted in Sydney than in Melbourne or Brisbane.Though it was possible in general to characterize the employee liable to neurosis, the predictive power of the characterization would be poor. The disorder followed no one pattern. Rather it appeared to be a collection of clinical syndromes which present as a result of the complex interaction of the personality with multiple factors at work and elsewhere over most of a lifetime. In individual subjects the relationship of stress at work to symptoms was usually ill defined, even in cases in which the identified probable factors were mainly or solely occupational. Nevertheless, there seems much to be gained from the establishment of mental health programmes in industry.Claims by unions in an Australian communications 1969). Other aspects of the investigation, on undertaking that the health of their members was occ...

show abstract

“…Methods and purposes 1. Canopy trees -coenocline ISA (Hill et al, 1975); CANMAR Where shown in Tables and Figures Table 2 Results from 2 and 3 by GOWER followed by CANONGO (Williams & Lance, 1968;Williams, 1976) and principal coordinate analysis (PCA; Gower, 1967;Gittins, 2012) followed by GOWECOR (first 20 Spearman correlations).…”

Section: Stages Of Analysismentioning

confidence: 99%

Natural Limits to the Expansion of Subtropical Rainforest at Mt Nebo, Queensland

Hegarty¹

2020

The Proceedings of the Royal Society of Queensland

View full text Add to dashboard Cite

Small patches of rainforest are numerous in subtropical Queensland. Their expansion into adjacent eucalypt forests is limited by fires and other factors. This study was completed in 1980 within a compact site of 200ha in conserved forest at Mt Nebo, Queensland. It describes the separate contributions of many natural site factors to the stability of the boundaries between these forest types. The survey area included areas of eucalypt forest with a grassy understorey and adjoining ecotonal areas. These, in turn, merged into patches of old-growth complex notophyll vine forest (CNVF). Three separate datasets were recorded from 160 plots, clustered around 32 grid points: (1) canopy species; (ii) woody species in the understorey stratum; and (iii) site characteristics, as indicators of the potential for localised progression from eucalypt to CNVF at canopy level. CSIRO Division of Computing Research in Brisbane processed the data, using methods developed to define initial stages of recovery from clearing inside old-growth CNVF at nearby Mt Glorious. Apart from recent fires, the long-term expansion of CNVF across an ecological gradient at Mt Nebo was limited at various stages by factors such as the orientation, aspect and shape of slopes, soil derivation, the type and thickness of litter cover, soil moisture levels, and the presence and persistence of the exotic shrub lantana (Lantana camara). The compact survey area was free of some confounding influences seen in broader regional studies (e.g. variations in elevation, soil derivation, temperature and rainfall, and incomplete records of fire, grazing and clearing). In this study the ranges of separate sets of observations of canopy and understorey trees, and natural site characteristics within each plot, were unusually broad, allowing definition of how natural site factors combine to allow, and limit, succession from eucalypt forest to mature subtropical CNVF in the absence of fire.

show abstract

Choice of Strategy in the Analysis of Complex Data

Cited by 32 publications

References 21 publications

Trends and problems in the application of classification and ordination methods in plant ecology

Trends and problems in the application of classification and ordination methods in plant ecology

A study of neurosis and occupation

Natural Limits to the Expansion of Subtropical Rainforest at Mt Nebo, Queensland

Contact Info

Product

Resources

About