Title from title page of PDF (University of Missouri--St. Louis, viewed February 15, 2010). Includes bibliographical references (p. 89-96).

The vast number of different study variables or population characteristics and the different domains of interest in a survey, make it impractical and almost impossible to calculate and publish standard errors for each estimated value of a population variable or characteristic and each domain individually. Since estimated values are subject to statistical variation (resulting from the probability sampling), standard errors may not be omitted from the survey report. Estimated values can be evaluated only if their precision is known. The purpose of this research project is to study the feasibility of mathematical modeling to estimate the standard errors of estimated values of population parameters or characteristics in survey data sets and to investigate effective and user-friendly presentation methods of these models in reports. The following data sets were used in the investigation: • October Household Survey (OHS) 1995 - Workers and Household data set • OHS 1996 - Workers and Household data set • OHS 1997 - Workers and Household data set • Victims of Crime Survey (VOC) 1998 The basic methodology consists of the estimation of standard errors of the statistics considered in the survey for a variety of domains (such as the whole country, provinces, urban/rural areas, population groups, gender and age groups as well as combinations of these). This is done by means of a computer program that takes into consideration the complexity of the different sample designs. The direct calculated standard errors were obtained in this way. Different models are then fitted to the data by means of regression modeling in the search for a suitable standard error model. A function of the direct calculated standard error value served as the dependent variable and a function of the size of the statistic served as the independent variable. A linear model, equating the natural logarithm of the coefficient of relative variation of a statistic to a linear function of the natural logarithm of the size of the statistic, gave an adequate fit in most of the cases. Well-known tests for the occurrence of outliers were applied in the model fitting procedure. For each observation indicated as an outlier, it was established whether the observation could be deleted legitimately (e.g. when the domain sample size was too small, or the estimate biased). Afterwards the fitting procedure was repeated. The Australian Bureau of Statistics also uses the above model in similar surveys. They derived this model especially for variables that count people in a specific category. It was found that this model performs equally well when the variable of interest counts households or incidents as in the case of the VOC. The set of domains considered in the fitting procedure included segregated classes, mixed classes and cross-classes. Thus, the model can be used irrespective of the type of subclass domain. This result makes it possible to approximate standard errors for any type of domain with the same model. The fitted model, as a mathematical formula, is not a user-friendly presentation method of the precision of estimates. Consequently, user-friendly and effective presentation methods of standard errors are summarized in this report. The suitability of a specific presentation method, however, depends on the extent of the survey and the number of study variables involved. / Dissertation (MSc (Mathematical Statistics))--University of Pretoria, 2007. / Mathematics and Applied Mathematics / unrestricted

Zugl.: Jena, Univ., Diss., 2006

Thesis (MPhil)--Stellenbosch University, 2004 / ENGLISH ABSTRACT: The overarching aim of this study is to analyse what researchers in South African higher education institutes and science councils mean by the term 'utilisation'. In specific terms, this aim is interpreted as determining what researchers mean when they indicate that their research findings are being utilised. The data used to conduct this analysis is taken from a national survey conducted to establish the extent to which research findings in South Africa are utilised. The Centre for Interdisciplinary Studies - now called the Centre for Research on Science and Technology - conducted the survey for the National Advisory Committee on Innovation (NACI). This study for NACI involved a questionnaire survey of research being conducted with universities, technikons and science councils, and a telephone survey of a sample of research and development managers in 116 companies. As a researcher in the Centre, I played an active role in this two-fold survey component of the research process. Besides coding the questionnaire for the NACI survey, for the purposes of my study I did additional coding of the questionnaire data. This involved coding 1052 responses to an openended question using a software package. After exporting this data to Microsoft Excel, I further coded the data into one of three categories: Mode 1, Mode 2 and a combination of Mode 1 and Mode 2. This 'new variable' created, formed part of the quantitative analysis since it was correlated with the following variables: the 'trigger' or 'driver' behind the research; the expected outcome of the project or study; the scientific field of the project; the project's or the study's science culture; the source of funding of the study, the modes of dissemination of the results and the intended beneficiary(ies) of the research. Results of the qualitative analyses reveal a profile for projects exhibiting features of Mode 1, Mode 2 or a combination of Mode 1 and Mode 2 knowledge utilisation. With Mode 1 we see that research was likely to be utilised within the discipline, was conducted for the benefit of the researcher's peers and it expanded on existing knowledge - all indicating that Mode 1 knowledge utilisation is predominantly associated with fundamental research. With Mode 2 we see that research was produced primarily for its use value and included the user's interests - indicating that Mode 2 knowledge utilisation is predominantly applied, commissioned and/or strategic. Projects that exhibited both features of Mode 1 and Mode 2 were both indicative of fundamental and applied research. Besides the quantitative analyses also showing the same result mentioned above, it was also discovered that the majority of the projects exhibited features of Mode 2, with one third of the projects exhibiting features of Mode 1 and a small proportion of projects as comprising both modes of utilisation. This was a very interesting finding because it is no longer just speculation that a shift is occurring to more applied, strategic research. The analysis revealed that this shift is a reality. The correlations of mode of utility with the seven other variables produced a range of results that expanded on the features of Mode 1 and Mode 2 type of knowledge production and verified that each mode had qualities unique to itself. My recommendations to researchers in South African research institutes concerns conducting the type of research which will be more relevant to the needs of South African society at large. To funding bodies and programmes of South Africa, the suggestion is to become more informed about the dissemination and intended utilisation strategies that they fund. / AFRIKAANSE OPSOMMING: Die doel van hierdie studie is om 'n analise te doen van die begrip "navorsingsbenutting", soos verstaan deur navorsers binne die publieke Suid Afrikaanse navorsing- en ontwikkelingsektor. Data wat verkry is van 'n landswye ondersoek na die aanwending van navorsingsuitsette binne die publieke sektor, soos onderneem deur die Sentrum vir Interdissiplinêre Studies, is geanaliseer. As 'n lid van die navorsingspan wat die studie onderneem het, was ek aktief betrokke by die proses wat onder meer 'n vraelys-ondersoek ingesluit het na navorsingsprojekte wat in swang is aan SA universiteite, tegnikons en wetenskapsrade. Vir die doel van die studie het ek 1052 response wat die resultaat was van 'n oop vraag in die vraelys in drie kategorieë gekodeer naamlik Modus 1, Modus 2 asook Modus 3, 'n kombinasiekategorie met die benaming "modus van aanwending". Die kwalitatiewe analise het getoon dat in die geval van modus 1, die navorsing mees waarskynlik binne die dissipline en tot die voordeel van die navorser se eweknieë aangewend sal word en dat dit in die meeste gevalle 'n uitbreiding van bestaande kennis behels - 'n aanduiding dat modus 1 navorsing hoofsaakilk verbind kan word met fundamentele navorsing. In die geval van modus 2 blyk dit dat navorsing primêr berus op die utiliteit- of aanwendingswaarde daarvan. Dit vertoon 'n noue verband met die gebruiker se behoefte of belangstelling en is hoofsaaklik toegepaste, kontrak of strategiese navorsing. Projekte wat aspekte van beide modi 1 en 2 vertoon, het tegelykertyd gedui op fundamentele en toegepaste navorsing. Die kwantitatiewe analise het aangedui dat die meerderheid projekte eienskappe van modus 2 vertoon het, met 'n derde van die projekte wat eienskappe van modus 1 vertoon het. 'n Klein hoeveelheid projekte het aspekte van beide modi 1 en 2 vertoon. Hierdie analise het aangetoon dat 'n verkuiwing inderdaad plaasgevind het vanaf fundamentele na meer toegepaste en strategiese navorsing. Die korrelasie van die modus van aanwending met sewe ander veranderlikes, het 'n reeks resultate tot gevolg gehad wat die eienskappe van modi 1 en 2 verder uitgebrei het. Dit het ook bygedra tot die bevestiging van die spesifieke eienskap van elke modus en unieke aspekte van elke modus aangetoon. Aanbevelings word ten slotte gedoen met die oog op 'n keuse vir meer relevante navorsing binne die publieke sektor gemeet aan die kriteria van die navorsingsbehoeftes van die breër Suid Afrikaanse gemeenskap. Aanbevelings m.b.t. disseminasie en navorsingsaanwendingstrategieë van navorsing wat deur befondsingsliggame befonds word, word ook gemaak.

A field experiment examines within a financial services firm the impact of a customer satisfaction survey-based intervention that enables front-line employees to identify and contact less-than satisfied customers (less than 9 on a 10-point scale) to proactively prevent potential customer defections. The impact is measured using operational data from 28,000 new customers and their associated defection behavior over a period of eight years. The experiment applies binomial Z-tests of proportions to assess the difference in defection rates of targeted and non-targeted customers before and after the intervention. The research finds that the use of closed loop satisfaction measurement reduces customer defections (by 40%, p>.001). Further, the research finds that the primary reduction is for non-targeted customers rather than for targeted customers, contrary to expectations. The research also provides additional support for the 'service recovery paradox' wherein customers who are less-than satisfied who are satisfactorily resolved have reduced defection rates compared to customers that are satisfied (by 47%, p=.016). The primary limitation of the research is its reliance on data from a single company. Another limitation is the potentially confounding impact of the Great Recession on defections during the study period, which could threaten the validity of the analysis. Consequently, additional tests were performed to control for this and other potentially confounding factors. These tests show that the Great Recession and the host company's cost cutting reactions did impact defections and therefore data from these periods were eliminated from the analyses. The primary theoretical contributions are the use of actual customer defections to measure the impacts and the use of a proactive rather than a reactive intervention. The contribution from a practitioner perspective is the relatively low cost of this intervention for improving customer retention.

Depuis de nombreuses années, les évaluations de la faune circulante au sein de l’estuaire de la Gironde s’appuient sur plusieurs dispositifs de suivis ichtyologiques, chroniques de pêches scientifiques et professionnelles. Cependant, ces approches conventionnelles nécessitent souvent un échantillonnage conséquent, restent ponctuelles dans le temps, spatialement limitées et destructrices. Tester la mise en œuvre d’un suivi acoustique prend ainsi toute son sens et son intérêt. Plus largement appliqué en milieux marin et lacustre, cette approche reste encore rarement utilisée en estuaires. En raison de l’extrême variabilité intrinsèque de ce type d’écosystème, la mise en œuvre de cet outil est un réel défi. En s’appuyant sur la connaissance préalablement acquise, une réflexion méthodologique approfondie a été menée pour définir des stratégies d’échantillonnage acoustiques afin d’optimiser la récolte de ce type de données. Plusieurs stratégies acoustiques ont d’abord été testées en tenant compte de la complexité et de la taille importante de l’écosystème étudié. La pertinence de ces données recueillies a ensuite été confrontée aux données des pêches scientifiques. Enfin, l’hydroacoustique a été mis à profit pour répondre à des questions sur la dynamique ichtyologique à des échelles spatiales et temporelles plus fines notamment pour étudier les influences tidales et nycthémérales. Ce travail démontre qu’il est possible de mettre en place un suivi acoustique des peuplements ichtyologiques pour un estuaire comme la Gironde. Comme outil complémentaire aux approches traditionnelles, l’hydroacoustique ouvre un large champ de possibilités pour appréhender la biocénose piscicole pour les milieux estuariens. / For many years, assessments of fish densities within the Gironde estuary have been based on a number of different fish monitoring techniques, as well as professional and scientific surveys. However, these approaches are often destructive. They require substantial sampling, and are limited from the points of view of both time and space. With this in mind, the benefits of non-invasive acoustic monitoring are clear. While such methods are often used in lakes and other marine environments, they have yet to be widely applied to shallow-water areas, in particular estuaries. Applying an acoustic approach to this kind of intrinsically changeable environment represents a real challenge. In this research project, an in-depth methodological study was carried out to determine the optimum way of applying acoustic sampling. Several acoustic monitoring strategies were then tested, keeping in mind the size and complexity of the ecosystem under investigation. The results of these tests were then checked for relevance against data from scientific fisheries. Finally, hydroacoustics was used to examine fish dynamics at finer spatial and temporal scales, notably in the study of diurnal and tidal influences. The results of this study show that it is possible to use acoustic monitoring to complement traditional methods of estimating fish populations in estuarine areas, and that the use of such a combination opens up a broad range of new possibilities.