A semantic modelling approach to knowledge based statistical software

The topic of this thesis is the development of knowledge based statistical software. The shortcomings of conventional statistical packages are discussed to illustrate the need to develop software which is able to exhibit a greater degree of statistical expertise, thereby reducing the misuse of statistical methods by those not well versed in the art of statistical analysis. Some of the issues involved in the development of knowledge based software are presented and a review is given of some of the systems that have been developed so far. The majority of these have moved away from conventional architectures by adopting what can be termed an expert systems approach. The thesis then proposes an approach which is based upon the concept of semantic modelling. By representing some of the semantic meaning of data, it is conceived that a system could examine a request to apply a statistical technique and check if the use of the chosen technique was semantically sound, i.e. will the results obtained be meaningful. Current systems, in contrast, can only perform what can be considered as syntactic checks. The prototype system that has been implemented to explore the feasibility of such an approach is presented, the system has been designed as an enhanced variant of a conventional style statistical package. This involved developing a semantic data model to represent some of the statistically relevant knowledge about data and identifying sets of requirements that should be met for the application of the statistical techniques to be valid. Those areas of statistics covered in the prototype are measures of association and tests of location.

Statistical software usage by student and consultant statisticians in the UK - A comparative study

Abstract not available

Statistical tools for analyzing water quality data

Water quality data are often collected at different sites over time to improve water quality management. Water quality data usually exhibit the following characteristics: non-normal distribution, presence of outliers, missing values, values below detection limits (censored), and serial dependence. It is essential to apply appropriate statistical methodology when analyzing water quality data to draw valid conclusions and hence provide useful advice in water management. In this chapter, we will provide and demonstrate various statistical tools for analyzing such water quality data, and will also introduce how to use a statistical software R to analyze water quality data by various statistical methods. A dataset collected from the Susquehanna River Basin will be used to demonstrate various statistical methods provided in this chapter. The dataset can be downloaded from website http://www.srbc.net/programs/CBP/nutrientprogram.htm.

An overview of mixed-effects statistical models for second language researchers

As in any field of scientific inquiry, advancements in the field of second language acquisition (SLA) rely in part on the interpretation and generalizability of study findings using quantitative data analysis and inferential statistics. While statistical techniques such as ANOVA and t-tests are widely used in second language research, this review article provides a review of a class of newer statistical models that have not yet been widely adopted in the field, but have garnered interest in other fields of language research. The class of statistical models called mixed-effects models are introduced, and the potential benefits of these models for the second language researcher are discussed. A simple example of mixed-effects data analysis using the statistical software package R (R Development Core Team, 2011) is provided as an introduction to the use of these statistical techniques, and to exemplify how such analyses can be reported in research articles. It is concluded that mixed-effects models provide the second language researcher with a powerful tool for the analysis of a variety of types of second language acquisition data.

Statnote 8: statistical power and sample size

Statistical software is now commonly available to calculate Power (P') and sample size (N) for most experimental designs. In many circumstances, however, sample size is constrained by lack of time, cost, and in research involving human subjects, the problems of recruiting suitable individuals. In addition, the calculation of N is often based on erroneous assumptions about variability and therefore such estimates are often inaccurate. At best, we would suggest that such calculations provide only a very rough guide of how to proceed in an experiment. Nevertheless, calculation of P' is very useful especially in experiments that have failed to detect a difference which the experimenter thought was present. We would recommend that P' should always be calculated in these circumstances to determine whether the experiment was actually too small to test null hypotheses adequately.

A Computational Approach for the Statistical Estimation of Discrete Time Branching Processes with Immigration

2010 Mathematics Subject Classification: 60J80.

The Implication of Using NVivo Software in Qualitative Data Analysis: Evidence-Based Reflections

For a long time, electronic data analysis has been associated with quantitative methods. However, Computer Assisted Qualitative Data Analysis Software (CAQDAS) are increasingly being developed. Although the CAQDAS has been there for decades, very few qualitative health researchers report using it. This may be due to the difficulties that one has to go through to master the software and the misconceptions that are associated with using CAQDAS. While the issue of mastering CAQDAS has received ample attention, little has been done to address the misconceptions associated with CAQDAS. In this paper, the author reflects on his experience of interacting with one of the popular CAQDAS (NVivo) in order to provide evidence-based implications of using the software. The key message is that unlike statistical software, the main function of CAQDAS is not to analyse data but rather to aid the analysis process, which the researcher must always remain in control of. In other words, researchers must equally know that no software can analyse qualitative data. CAQDAS are basically data management packages, which support the researcher during analysis.

Improving quality management system implementation in Indonesian construction firms : a research project

This paper has two main sections, the first of which presents a summarized review of the literature concerning previous studies on the implementation of ISO 9000 quality management systems (QMSs) both in global construction companies as well as in Indonesian construction firms, and the perceived correlation between organisational culture and QMS practices in the construction sector. The first section of the paper contributes to the development of the second section, which presents details of the research project being undertaken. Based on the fundamental questions that led to the development of the main research objectives, suitable research methods have been developed in order to meet these objectives. Primary data will be collected by use of a mixed methods approach, i.e., questionnaire surveys and focus group discussions/interviews in order to obtain opinions from respondents drawn from targeted ISO construction firms. Most of the data expected to be obtained will be in future be analyzed using statistical software then the findings will be discussed in order to ultimately develop a culture-based QMS framework.

PySSM : a Python module for Bayesian inference of linear Gaussian state space models

PySSM is a Python package that has been developed for the analysis of time series using linear Gaussian state space models (SSM). PySSM is easy to use; models can be set up quickly and efficiently and a variety of different settings are available to the user. It also takes advantage of scientific libraries Numpy and Scipy and other high level features of the Python language. PySSM is also used as a platform for interfacing between optimised and parallelised Fortran routines. These Fortran routines heavily utilise Basic Linear Algebra (BLAS) and Linear Algebra Package (LAPACK) functions for maximum performance. PySSM contains classes for filtering, classical smoothing as well as simulation smoothing.

Association between postmenopausal osteoporosis and experimental periodontitis

To investigate the correlation between postmenopausal osteoporosis (PMO) and the pathogenesis of periodontitis, ovariectomized rats were generated and the experimental periodontitis was induced using a silk ligature. The inflammatory factors and bone metabolic markers were measured in the serum and periodontal tissues of ovariectomized rats using an automatic chemistry analyzer, enzyme-linked immunosorbent assays, and immunohistochemistry. The bone mineral density of whole body, pelvis, and spine was analyzed using dual-energy X-ray absorptiometry and image analysis. All data were analyzed using SPSS 13.0 statistical software. It was found that ovariectomy could upregulate the expression of interleukin- (IL-)6, the receptor activator of nuclear factor-κB ligand (RANKL), and osteoprotegerin (OPG) and downregulate IL-10 expression in periodontal tissues, which resulted in progressive alveolar bone loss in experimental periodontitis. This study indicates that changes of cytokines and bone turnover markers in the periodontal tissues of ovariectomized rats contribute to the damage of periodontal tissues.

The epidemiology of basal cell carcinoma

Basal cell carcinoma (BCC) is a skin cancer of particular importance to the Australian community. Its rate of occurrence is highest in Queensland, where 1% to 2% of people are newly affected annually. This is an order of magnitude higher than corresponding incidence estimates in European and North American populations. Individuals with a sun-sensitive complexion are particularly susceptible because sun exposure is the single most important causative agent, as shown by the anatomic distribution of BCC which is in general consistent with the levels of sun exposure across body sites. A distinguishing feature of BCC is the occurrence of multiple primary tumours within individuals, synchronously or over time, and their diagnosis and treatment costs contribute substantially to the major public health burden caused by BCC. A primary knowledge gap about BCC pathogenesis however was an understanding of the true frequency of multiple BCC occurrences and their body distribution, and why a proportion of people do develop more than one BCC in their life. This research project sought to address this gap under an overarching research aim to better understand the detailed epidemiology of BCC with the ultimate goal of reducing the burden of this skin cancer through prevention. The particular aim was to document prospectively the rate of BCC occurrence and its associations with constitutional and environmental (solar) factors, all the while paying special attention to persons affected by more than one BCC. The study built on previous findings and recent developments in the field but set out to confirm and extend these and propose more adequate theories about the complex epidemiology of this cancer. Addressing these goals required a new approach to researching basal cell carcinoma, due to the need to account for the phenomenon of multiple incident BCCs per person. This was enabled by a 20 year community-based study of skin cancer in Australians that provided the methodological foundation for this thesis. Study participants were originally randomly selected in 1986 from the electoral register of all adult residents of the subtropical township of Nambour in Queensland, Australia. On various occasions during the study, participants were fully examined by dermatologists who documented cumulative photodamage as well as skin cancers. Participants completed standard questionnaires about skin cancer-related factors, and consented to have any diagnosed skin cancers notified to the investigators by regional pathology laboratories in Queensland. These methods allowed 100% ascertainment of histologically confirmed BCCs in this study population. 1339 participants had complete follow-up to the end of 2007. Statistical analyses in this thesis were carried out using SAS and SUDAAN statistical software packages. Modelling methods, including multivariate logistic regressions, allowed for repeated measures in terms of multiple BCCs per person. This innovative approach gave new findings on two levels, presented in five chapters as scientific papers: 1. Incidence of basal cell carcinoma multiplicity and detailed anatomic distribution: longitudinal study of an Australian population The incidence of people affected multiple times by BCC was 705 per 100,000 person years compared to an incidence rate of people singly affected of 935 per 100,000 person years. Among multiply and singly affected persons alike, site-specific BCC incidence rates were far highest on facial subsites, followed by upper limbs, trunk, and then lower limbs 2. Melanocytic nevi and basal cell carcinoma: is there an association? BCC risk was significantly increased in those with forearm nevi (Odds Ratios (OR) 1.43, 95% Confidence Intervals (CI) 1.09-1.89) compared to people without forearm nevi, especially among those who spent their time mainly outdoors (OR 1.6, 95%CI 1.1-2.3) compared to those who spent their time mainly indoors. Nevi on the back were not associated with BCC. 3. Clinical signs of photodamage are associated with basal cell carcinoma multiplicity and site: a 16-year longitudinal study Over a 16-year follow-up period, 58% of people affected by BCC developed more than one BCC. Among these people 60% developed BCCs across different anatomic sites. Participants with high numbers of solar keratoses, compared to people without solar keratoses, were most likely to experience the highest BCC counts overall (OR 3.3, 95%CI 1.4-13.5). Occurrences of BCC on the trunk (OR 3.3, 95%CI 1.4-7.6) and on the limbs (OR 3.7, 95%CI 2.0-7.0) were strongly associated with high numbers of solar keratoses on these sites. 4. Occurrence and determinants of basal cell carcinoma by histological subtype in an Australian community Among 1202 BCCs, 77% had a single growth pattern and 23% were of mixed histological composition. Among all BCCs the nodular followed by the superficial growth patterns were commonest. Risk of nodular and superficial BCCs on the head was raised if 5 or more solar keratoses were present on the face (OR 1.8, 95%CI 1.2-2.7 and OR 4.5, 95%CI 2.1-9.7 respectively) and similarly on the trunk in the presence of multiple solar keratoses on the trunk (OR 4.2, 95%CI 1.5-11.9 and OR 2.2, 95%CI 1.1-4.4 respectively). 5. Basal cell carcinoma and measures of cumulative sun exposure: an Australian longitudinal community-based study Dermal elastosis was more likely to be seen adjacent to head and neck BCCs than trunk BCCs (p=0.01). Severity of dermal elastosis increased on each site with increasing clinical signs of cutaneous sun damage on that site. BCCs that occurred without perilesional elastosis per se, were always found in an anatomic region with signs of photodamage. This thesis thus has identified the magnitude of the burden of multiple BCCs. It does not support the view that people affected by more than one BCC represent a distinct group of people who are prone to BCCs on certain body sites. The results also demonstrate that BCCs regardless of site, histology or order of occurrence are strongly associated with cumulative sun exposure causing photodamage to the skin, and hence challenge the view that BCCs occurring on body sites with typically low opportunities for sun exposure or of the superficial growth pattern are different in their association with the sun from those on typically sun-exposed sites, or nodular BCCs, respectively. Through dissemination in the scientific and medical literature, and to the community at large, these findings can ultimately assist in the primary and secondary prevention of BCC, perhaps especially in high-risk populations.

Using decision trees to understand structure in missing data

Objectives Demonstrate the application of decision trees – classification and regression trees (CARTs), and their cousins, boosted regression trees (BRTs) – to understand structure in missing data. Setting Data taken from employees at three different industry sites in Australia. Participants 7915 observations were included. Materials and Methods The approach was evaluated using an occupational health dataset comprising results of questionnaires, medical tests, and environmental monitoring. Statistical methods included standard statistical tests and the ‘rpart’ and ‘gbm’ packages for CART and BRT analyses, respectively, from the statistical software ‘R’. A simulation study was conducted to explore the capability of decision tree models in describing data with missingness artificially introduced. Results CART and BRT models were effective in highlighting a missingness structure in the data, related to the Type of data (medical or environmental), the site in which it was collected, the number of visits and the presence of extreme values. The simulation study revealed that CART models were able to identify variables and values responsible for inducing missingness. There was greater variation in variable importance for unstructured compared to structured missingness. Discussion Both CART and BRT models were effective in describing structural missingness in data. CART models may be preferred over BRT models for exploratory analysis of missing data, and selecting variables important for predicting missingness. BRT models can show how values of other variables influence missingness, which may prove useful for researchers. Conclusion Researchers are encouraged to use CART and BRT models to explore and understand missing data.