ABSTRACT Aim of the study: This study aims to contribute to the existing knowledge on similarities and differences of networking styles across Europe and support either network contextualists or universalists with new empirical knowledge. Method used: A multiple case study was used in this paper. Qualitative data was gathered gathering from 10 interviews collected amongst experienced Swedish, Polish and Dutch entrepreneurs. Findings: Empirical findings have revealed significant differences in the networking patterns in the Netherlands, Sweden and Poland. Main networking motivation for Dutch entrepreneurs is seeking new sales opportunities, for Swedish it is the exchange of knowledge and information. All the entrepreneurs admitted that trust is an important component of network relationship. However, in Sweden trust seem to play more significant role. The empirical findings questioned the theory that trust can be easily transferred based on recommendations. No entrepreneur directly suggested that asked about trust acquisition. In Sweden informal networks are strongly preferred, while in the Netherlands formal network are much more popular. In Poland informal networks are also more popular. In the Netherlands, commercial, paid organizations may be extremely popular given the fact that all of the Dutch respondents participate in them, on the other hand in Sweden governmental networking organizations are quite common, in Poland no respondent have mentioned neither of them. Findings indicate that networking is still not as developed in Poland as it is in two other researched countries. Practical Implications: Common networking patterns cannot be expected in different European countries and background studies should be conducted before entering foreign market. There is still market growth opportunity for commercial networking organizations in Poland. Polish government should engage in more effective promotion of networking amongst entrepreneurs. Contribution: Empirical data confirm the contextualists approach that emphasizes the importance of national differences on networking and warns managers that resemblance of certain patterns cannot be assumed for different countries. This approach was confirmed since the findings exposed significant differences in several areas of networking. Keywords: Business networking, internationalization of business network, networking motivations, trust building, network type preferences, network contextualism and universalism.

Conditional preference networks (CP-nets) exploit the power of ceteris paribus rules to represent preferences over combinatorial decision domains compactly. CP-nets have much appeal. However, their study has not yet advanced sufficiently for their widespread use in real-world applications. Known algorithms for deciding dominance---whether one outcome is better than another with respect to a CP-net---require exponential time. Data for CP-nets are difficult to obtain: human subjects data over combinatorial domains are not readily available, and earlier work on random generation is also problematic. Also, much of the research on CP-nets makes strong, often unrealistic assumptions, such as that decision variables must be binary or that only strict preferences are permitted. In this thesis, I address such limitations to make CP-nets more useful. I show how: to generate CP-nets uniformly randomly; to limit search depth in dominance testing given expectations about sets of CP-nets; and to use local search for learning restricted classes of CP-nets from choice data.

Thesis (MAgricAdmin)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The objective of this study was to determine the sensory characteristics of Cheddar cheese that drive consumer acceptance. In addition, it investigated whether specific sensory profiles would satisfy specific consumer segments to ultimately facilitate greater acceptance and consumption of Cheddar cheese. A trained panel applied descriptive sensory analysis (DSA) to characterise the sensory attributes of six commercially produced cheeses ranging in maturity from 1 to 32 months. The cheese samples were differentiated by appearance, aroma, flavour and textural attributes. A consumer panel comprising of 115 target consumers rated preference for the cheeses on a 9-point hedonic scale. Moreover consumers completed a screener designed to collect socio-demographic information and opinions on cheese and cheese products. Instrumental colour analysis was performed on the cheese samples and in-depth consumer opinions and attitudes towards cheese were examined using the focus group technique. Analysis of variance (ANOVA) was conducted on the sensory and instrumental data, as well as the consumer liking and perception data. Additionally, a t-test was performed at a 5% significance level to determine the direction of the difference between the mean values. Multivariate analyses were performed on the sensory and consumer data to determine whether relationships existed between sample attributes and consumer liking. Consumer acceptability scores were segmented by agglomerative hierarchical clustering (AHC) using Ward’s test. ANOVA was performed on the consumer data per cluster for colour liking, texture liking, flavour liking and overall liking. Partial least squares regression (PLS) was performed in an attempt to relate consumer degree of liking data, the socialdemographic and attitudinal data, as well as the sensory and instrumental data. The younger cheeses; Gouda at one month, Edam at two months and Cheddar at 4 months were characterised as glossy, with a buttery and creamy aroma; creamy flavour, salty and slightly sour taste combined with lingering cheese flavour. Textures were described as creamy, springy and rubbery. Sensory attributes of visible white crystals, moisture seepage, brothy and Cheddar aromas; brothy, Cheddar and prickle flavours together with textural attributes of firm, hard, crumbly, grainy and teeth-coating characterised the aged Cheddar cheeses. Flavour development was not progressive as the cheeses aged and atypical flavour development was identified in the Cheddar cheese aged to 8 and 15 months, respectively. Flavour followed by price and convenience drive cheese purchase amongst this group of consumers. From a sensory perspective, glossy appearance, a red hue and yellow colour drive cheese colour liking. Texture liking is driven by springy and hard attributes. Flavour liking is driven by cooked milk, buttery and creamy aroma and creamy flavour. Finally overall cheese liking is driven by glossy appearance; cooked milk, buttery and creamy aroma; creamy flavour; springy and creamy texture; specific instrumental colour values, as well as fat, moisture and salt content. Three clusters per variable of colour, texture, flavour and overall liking of cheese were identified through segmentation performed on the preference data. Results indicate that consumers are not homogenous in their sensory requirements, attitudes and behaviour towards cheese. Based on the consumer preference for specific sensory characteristics of cheese, consumer segments can be identified and cheese can be optimised to satisfy these sensory requirements. / AFRIKAANSE OPSOMMING: Die doel van hierdie studie was eerstens om te bepaal watter spesifieke sensoriese eienskappe van Cheddarkaas verbruikersaanvaarbaarheid dryf, en tweedens om verbruikersdata te segmenteer ten einde vas te stel of verskillende verbruikersegmente verskillende voorkeure het wat betref Cheddarkaas. Kennis hiervan sal uiteindelik bydra tot verhoogde verbruikerstevredentheid, en dus verhoogde verkope van Cheddarkaas op die plaaslike mark. Beskrywende sensoriese analise en ’n opgeleide paneel is gebruik om die sensoriese eienskappe van ses kommersiële kaas variante te bepaal. Instrumentele kleuranalise is ook bepaal. Die monsters kaas het gewissel in ouderdom (van 1 tot 32 maande veroudering), asook in voorkoms, aroma, geur, smaak en teksturele eienskappe. ’n Verbruikerspaneel bestaande uit 115 individue het die voorkeur en aanvaarbaarheid van die ses variante kaas ge-evalueer deur gebruik te maak van die 9-punt hedoniese skaal. Sosio-demografiese inligting en verbruikeropinies betreffende kaas is ook versamel. Ten laaste is die fokusgroep tegniek gebruik om in-diepte inligting te verkry wat betref verbruikeropinies en –houdings ten opsigte van Cheddarkaas. Analise van variansie (ANOVA) is op sensoriese, instrumentele en verbruikersdata toegepas. T-toets is by ‘n betekenispeil van 5% uitgevoer om te bepaal of daar per eienskap betekenisvolle verskille tussen monstergemiddeldes was. Meerveranderlike analise is uitgevoer op die sensoriese en verbruikerdata ten einde te bepaal of spesifieke sensoriese produkeienskappe verbruikersvoorkeur dryf. Ward se statistiese groepering is gebruik om te bepaal of die verbruikers in verskillende voorkeursegmente gegroepeer kan word. Hierna is ANOVA op elke segment uitgevoer ten einde per segment verbruikersvoorkeur van die onderskeie produkte te bepaal wat betref voorkoms, geur en tekstuur. Meervoudige regressie analise is toegepas om die verwantskap tussen verbruikervoorkeur-, houding- en sosio-demografiese eienskappe van die verbruikers te bepaal, asook tussen die sensoriese en instrumentele produkeienskappe. Die kase wat vir ‘n kort tydperk verouder is, m.a.w. Gouda, Edam en Cheddar na 4 maande veroudering is beskryf as glansend, met ‘n ryk botter en romerige aroma, romerige geur, sout en effense suur smaak, asook ‘n standhoudende kaasgeur. Teksturele eienskappe van hierdie drie kaassoorte is beskryf as romerig, veerkragtig en effe rubberig. Die Cheddar kase wat langer verouder is, het tekens getoon van sigbare wit kristalle en sinerese op die kaasoppervlak, ‘n prikgevoel op die tong, asook sopagtige (“brothy”) en Cheddaragtige aromas en geure. Die teksturele eienskappe van die verouderde Cheddar kase is beskryf as ferm, hard, krummelagtig en greinerig, asook met’n geneigdheid van die kaas om aan die tande vas te kleef. Geurontwikkeling het nie progressief plaasgevind soos die kase verouder het nie. Die Cheddar kase wat onderskeidelik vir 8 en 15 maande verouder is, het atipiese geurontwikkeling getoon. Belangrike faktore wat tydens die aankoop van kaas’n rol speel, en dus voorkeur dryf is eerstens geur, tweedens prys en derdens produk-gemak. Sensoriese aspekte wat voorkeur dryf, is ‘’n glansende voorkoms en effe rooierige-gelerige kleur. Tekstuur voorkeure sluit in veerkragtiheid en hardheid, terwyl geur voorkeure die volgende insluit: romerige, botterige en melk aromas, asook ‘n romerige geur. In die geheel gesien, dryf die volgende eienskappe dus verbruikersvoorkeur: kaas met ‘n glansende voorkoms, geure soos botteragtig, roomagtig en melkerig en teksture soos veerkragtig en romerig. Voedinstofinhoud beȉnvloed ook verbruikersvoorkeur, nl. soutinhoud, vetinhoud en voginhoud. Segmentasie van verbruikersdata het getoon dat die onderskeie verbruikersegmente nie ooreenstem wat betref hul verbruikersvoorkeure, houdings en persepsies van kaassorte nie. Die resultate het egter spesifike sensoriese rigtingwysers geidentifiseer wat die bedryf kan gebruik om kaas te produseer vir elk van die onderskeie marksegmente.

Thesis (MComm)--Stellenbosch University, 2000. / ENGLISH ABSTRACT: It is generally accepted that profitable relationships with customers are increasingly important in determining the long-term success of the modem enterprise. Customer relationship management (CRM) is widely considered as an appropriate methodology for building strong and mutually beneficial relationships with customers and, therefore, the study established the definition and scope of CRM. Customer knowledge could be considered a cornerstone of any CRM strategy and the ability of the enterprise to create and leverage customer knowledge is viewed as a source of competitive advantage. The study aimed at establishing clarity on the methodologies and approaches to the organizational knowledge creation process. Finally, the integration of knowledge creation and leverage processes into the CRM strategy was considered. Scrutiny of the available literature revealed a comprehensive description of the concept of CRM. The various models of CRM presented in the study clearly illustrated the different focus areas and components of CRM. The most prominent characteristics of a CRM strategy were found to be an emphasis on strategy, the networking of key stakeholders, customer intimacy and the use of appropriate technology. The approaches to the creation of organizational knowledge were analysed and it was found that the conversion between tacit and explicit knowledge is key to the creation of organizational knowledge. The generally accepted principles for the leverage of knowledge were adapted for possible incorporation into a CRM strategy. The salient conclusion of this study was that a knowledge creation process could be used to build the enterprise's stock of customer knowledge. It was found that the principles of knowledge leverage could be applied to the customer knowledge base in order to build lasting and profitable customer relationships. Recommendations were offered for the improved leverage of a CRM strategy through the integration of knowledge creation, knowledge leverage and CRM methodologies and principles. Finally, suggestions were made for further academic research on the anatomy and implementation of CRM. / AFRIKAANSE OPSOMMING: Dit word algemeen aanvaar dat kliente verhoudingsbestuur 'n toenemende rol sal speel in die lang termyn sukses van die moderne onderneming. Kliente verhoudingsbestuur word wyd erken as 'n gepasde metodologie om sterk en wedersyds voordelige verhoudinge met kliente te bewerkstellig. Die studie het ten doel gehad om die definisie en omvang van kliente verhoudingsbestuur te bepaal. Kliente kennis is 'n hoeksteen van enige kliente verhoudingsbestuur strategie en die onderneming se vermoeë om kliente kennis te skep en te benut word gesien as 'n bron van kompeterende voordeel. Die studie het gepoog om duidelikheid te verkry oor die metodologieë en benaderings wat gevolg kan word in die ondernemingskennis skeppingsproses. Ten slotte is die oorvleueling tussen die kennis skeppingsproses en die kliente verhoudingsbestuur strategie ondersoek. Gebasseer op die studie van die relevante literatuur is 'n omvattende definisie van kliente verhoudingsbetuur blootgelê. Die onderskeie modelle van kliente verhoudingsbestuur wat ondersoek is in die studie, het die fokus en verskeie komponente van kliente verhoudingsbestuur uitgelig. Die mees prominente karakter eienskappe van die kliente verhoudingsbestuur strategie is aangedui as die fokus op ondernemings strategie, netwerke van belangegroepe, kliente intimiteit en die gebruik van toepaslike inligtingstegnologie. 'n Analise van die benaderings tot kenniskepping in die onderneming het uitgelig dat die oorgang van ontasbare kennis na tasbare kennis die kern uitmaak van ondernemingskennis skepping. Die algemeen aanvaarde beginsels tot die benutting van kennis is daarna aangepas vir die moontlike gebruik in die kliente verhoudingsbestuur strategie. Die belangrikste gevolgtrekking van hierdie studie was dat 'n kennis skeppingsproses gebruik kan word om die onderneming se kliente kennis uit te brei. Daar is ook gevind dat die beginsels van kennis benutting toegepas kan word op bestaande kliente kennis ten einde langtermyn winsgewende verhoudings met kliente te bewerkstellig. Aanbevelings is gemaak vir die verbeterde benutting van die kliente verhoudingsbestuur strategie deur middel van die integrasie van kennis skepping, kennis benutting en kliente verhoudingsbestuur metodologieë en -beginsels. Ten slotte is voorstelle gemaak vir verdere akademiese studies aangaande die struktuur en implementering van kliente vehoudingsbestuur.

Thesis (M Sport Sc (Sport Science))--University of Stellenbosch, 2009. / ENGLISH ABSTRACT: The study investigated the exercise preferences of female university students participating in exercise modes presented at the local gymnasium and their expectations of the outcomes of such participation. Secondly, the study aimed to determine the most important reason for their participation and whether this was satisfied by their choice of exercise mode. A third aim was to determine other reasons that contributed to the selection of exercise environment and mode. The size of the research group of the pilot study was 210 (n=210). For the final longitudinal study, over a period of three years, it was 985 (N=985). The study population was selected on a basis of convenient sampling, availability and interest among young female gymnasium members, (aged 18 to 27 years). Their participation was voluntary. Original questionnaires were constructed for the purpose of the study to provide general demographic and physical characteristic information of the participants, their exercise preference and choice of exercise mode, reasons for participation in particular exercise modes, time spent on physical activity, frequency of attendance of exercise sessions, exercise motivators, barriers to exercise, medication and supplementation prevalence, health problems and smoking. The questionnaire was completed in a five to 10 minute time slot before the commencement of exercise classes at the gymnasium. Information required on the questionnaire was verbally explained to the participants during the initial few minutes of data capturing. Guidance was given for each section of the questionnaire during the five to 10 minute period allocated for completion. Data was captured on Microsoft Excel spreadsheets and the analysis was performed using Statistica for Windows (Statsoft SA-2008). Descriptive statistics were used to analyse and present the data. The results of the study indicated that most (34.4%) young female students who regularly attended group exercise sessions at the gymnasium preferred participating in the exercise modality punchline (a boxing aerobic modality), with the aim of losing weight (45%) and improving their general fitness (24%). / AFRIKAANSE OPSOMMING: Die doel van die studie was om die oefenvoorkeure en oefenverwagtinge van jong damestudente wat gereeld by die plaaslike universiteitgimnasium oefen, vas te stel. Tweedens wou die studie die belangrikste redes vir deelname bepaal, en of hierdie verwagting gerealiseer het met die keuse van oefenmodaliteite in hierdie oefenomgewing. 'n Derde doel was om vas te stel watter ander redes 'n bydrae sou maak tot die keuse van 'n oefenomgewing en oefenmodaliteite. Die navorsingspopulasiegroep was jong dames in 'n universiteitsomgewing. Tydens die verkennende studie was die getal kandidate 210 (n=210). Vir die finale longitudinale studie, oor ’n tydperk van 3 jaar, was dit 985 (N=985). Die jong dames by die spesifieke universiteitgimnasium is geselekteer op grond van beskikbaarheid en belangstelling in die navorsing en was tussen die ouderdom van 18 en 27 jaar. Hul deelname was vrywillig. 'n Vraelys is spesiaal vir hierdie navorsing opgestel. Die vraelys het inligting ingewin oor algemene demografiese aspekte en fisieke kenmerke van die deelnemers, oefenvoorkeure en verwagtings van oefenmodaliteite, redes vir die seleksie en deelname in oefenmodaliteite, tyd bestee aan oefening, frekwensie van deelname, motiveerders vir oefening, beperkende faktore ten opsigte van gereelde deelname aan fisieke aktiwiteit, medikasie en supplementasie gebruik, gesondheidsprobleme en rook. Die vraelyste is voltooi in die eerste vyf tot 10 minute van oefenklasse, voor die aanvang van die gereelde gimnasiumprogram. Inligting op die vraelys is verbaal aan die teikengroep verduidelik in die eerste paar minute van elke klas. Tydens hierdie tydperk is gereeld inligting oor elke vraag verskaf vir kontrole en akkuraatheid van voltooiing van die vraelys. Data oor die veranderlikes is in die rekenaarprogram Microsoft Excel gekodeer en die statistiese ontleding is deur middel van Statistica vir Windows (Statsoft SA 2008) gedoen. Beskrywende data is gebruik om die resultate te ontleed en aan te bied.

Thesis (MBA)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: Brands, compnsing goods and services, serve to differentiate them from those of competitors. These have now proliferated to such an extent that it is difficult for brand owners to reach consumers and make their brands stand out. Brand loyalty has become one of the most frequently studied aspects of the purchasing process. It is thus defined through its measurements as the repeated purchase behaviour of one brand vis-á-vis other brands in the same category. The definition assumes a choice between two or more brands on the shelf. The objective of this project was to establish the presence of a direct relationship between brand loyalty and brand choice (or the lack thereof), in the emerging black markets of South Africa. The issue at hand is what happens in the absence of choice, i.e. the situation where there is only one brand per category. The second aspect to this is how do you measure brand loyalty in the absence of competition. The, outcome of the study clearly states that the choice that consumers are faced with in a particular market, lacks the dictionary meaning, In the absence of other brands in the same stable, loyalty has been found to exist because of the following: Brand being the first to market. Brand's superiority in terms of delivering on its promise, The presence of a strong advertising and promotions team to demonstrate its quality. Disposable incomes that are low and as such consumers' unwillingness to try other unknown brands, The lack of choice was described as a hostage situation, As long as the brand continues to deliver on quality, consumer choice will be deliberate and biased towards one brand. One very interesting other aspect to loyalty was that once it exists; it gets passed from one generation to the next. It becomes a family tradition, / AFRIKAANSE OPSOMMING: Handelsmerke onderskei produkte en dienste van hul mededingers. Die getal handelsmerke het nou tot so 'n mate toegeneem dat handelsmerkeienaars dit moeilik vind om verbruikers te bereik en hul handelsmerke duidelik van ander te onderskei. Handelsmerklojaliteit is een van die aspekte van die koopproses wat die meeste bestudeer word. Op grond van metingsprosedures word dit gedefinieer as 'herhaalde koopgedrag wat een handelsmerk bo ander handelsmerke in dieselfde kategorie bevoordeel'. Hierdie definisie veronderstel 'n keuse tussen twee of meer handelsmerke op dieselfde rak. Die doel van hierdie projek is om die direkte verwantskap tussen handelsmerklojaliteit en - keuse (of die gebrek daaraan) in die opkomende swart markte van Suid-Afrika te bepaal. Die eerste aspek van die ondersoek handel oor wat gebeur as verbruikers geen keuse het nie, m.a.w. as daar net een handelsmerk per kategorie op die rak is. Die tweede aspek handel oor die meting; van handelsmerklojaliteit as daar geen kompetisie van ander handelsmerke is nie. Die gevolgtrekkings van die studie illustreer duidelik dat die keuse wat verbruikers in 'n bepaalde mark het, nie aan 'n woordeboekdefinisie voldoen nie. Daar is ook bevind dat wanneer verbruikers nie 'n keuse van meer as een handelsmerk het nie, lojaliteit wel ontwikkel, om die volgende redes: Die handelsmerk was die eerste een wat in die mark beskikbaar was. Die betrokke handelsmerk doen sy beloftes beter gestand as ander handelsmerke. 'n Sterk advertensie- en promosiespan is teenwoordig om die kwaliteit van die handelsmerk te demonstreer. Die lae besteebare inkomste van verbruikers maak hulle onwillig om ander, onbekende handelsmerke te beproef Die gebrek aan keuse word beskryf as 'n "gyselaar"-situasie. Solank die handelsmerk aan die gehaltevereistes voldoen, sal verbruikers se keuse doelgerig daardie handelsmerk bevoordeel. Een van die interessantste aspekte van lojaliteit wat vorendag gekom het, is die volgende: as lojaliteit eers gevestig is, word dit van een generasie na die volgende oorgedra. So word dit dan 'n familietradisie.

Thesis (MBA)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: The purpose of the research was to investigate the decision-making styles of urban Zambian Generation X females shopping for apparel products. The research made use of a Consumer Styles Inventory (CSI) scale developed by Sproles and Kendall (1986) to measure the characteristics of various shopping styles. Out of 300 self-administered questionnaires distributed, 180 were used for data analysis. The Cronbach Alpha coefficients confirmed the reliability of the CSI scale on 7 out of 8 decision-making styles that could be associated with the consumers under review. The study then used Analysis of Variance (ANOVA) to establish the variation between the different decision-making styles. The findings revealed that the decision-making styles of quality consciousness and a recreational shopping orientation are highly correlated. The research findings have policy implications and recommendations for the development of marketing strategies and further research have been made. / AFRIKAANSE OPSOMMING: Die doel van die navorsing was om ondersoek te doen na die besluitnemingstyle van stedelike Zambiese vroue van Generasie X wanneer hulle klereprodukte koop. Die navorsing het gebruik gemaak van die inventaris van verbruikerstyle wat deur Sproles en Kendall (1986) ontwikkel is, ’n skaal wat in Engels as die Consumer Styles Inventory (CSI) bekendstaan, om die eienskappe van verskillende inkopiestyle te meet. Uit die 300 self-geadministreerde vraelyste wat versprei is, is 180 vir die ontleding van die data gebruik. The Cronbach Alpha-koëffisiënte kon die betroubaarheid van die CSI-skaal bevestig op 7 uit die 8 besluitnemingstyle wat geassosieer kon word met die verbruikers wat ondersoek is. Die navorsingstudie het daarna van variansieanalise gebruik gemaak om die variasie tussen die verskillende besluitnemingstyle te bepaal. Die bevindings het onthul dat die besluitnemingstyle van gehaltebewustheid en inkopies wat as rekreasie beskou word, hoogs gekorreleerd is. Die navorsingsbevindinge het implikasies vir beleid en aanbevelings is gedoen vir die ontwikkeling van bemarkingstrategieë en vir verdere navorsing.

Thesis (PhD)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: Limited information is available on the apple preferences of the South African consumer market, which is characterised by diverse consumers from different age and ethnic groups with different food preferences. White, coloured and black consumers from different age groups were selected from the Stellenbosch area, Western Cape, South Africa. Consumer preference analysis for apple eating quality and appearance, and descriptive sensory analysis (DSA) were performed on nine commercial apple cultivars. Analysis of variance (ANOVA) conducted on mean preference scores for each age and ethnic group showed that preference generally differed between these groups. However, Ward’s statistical clustering that was applied to the same data set showed that the sociodemographic composition of consumer groups with similar apple preferences is not homogenous. Three consumer clusters were identified with similar preferences for apple eating quality (E1-3) and appearance (A1-3): E1 liked firmness and therefore tolerated sour taste and disliked mealiness. Although E1 liked sweet fruit, they indicated lower preference for sweet fruit compared to E2 and E3. E2 liked sour taste and apple flavour more compared to the other clusters while E3 disliked sour taste and had the highest preference for sweetness. Although coloured and black consumers generally disliked sour taste and E3 constituted a larger proportion of these consumers, the coloured and black consumers who liked or tolerated sour taste constituted approximately 41% of the total consumer population in the Western Cape. White and younger (<26 years) consumers were mostly in cluster E1 liking firm fruit. Peel colour preferred by the appearance preference clusters were: Green and pink bi-colour (A1), green/yellow and red-striped (A2); and red peel colour (A3). Consumers preferred the appearance of cultivars that associated with the eating quality attributes that they liked. When consumers’ preference for the eating quality of five cultivars were analysed during presentation with different levels of visual pictorial information (no, correct and incorrect photograph), mismatches between expected and actual eating quality preference resulted in lower preference scores. Apple breeding is time-consuming and expensive. Comprehensive knowledge of fruit quality parameters that drive consumer preference is required to streamline the breeding process. Eating quality and appearance attributes of four apple breeding families were subjected to instrumental and individual assessment by a trained assessor and DSA by a trained panel. Instrumental measurements could not predict the sensory attributes analysed by the individual assessor. Sensory textural attributes, apple flavour and sweet taste as quantified by DSA and instrumental measurement of titratable acidity (TA) and total soluble solids (TSS)/TA, but not TSS, could predict consumer preference. The assessor responsible for individual assessment could not predict the preference of the total consumer group. A quantitative genetic analysis of the data was carried out to quantify within- and between-family variation using ANOVA, variance components and heritability estimates. Variation between families was shown for attributes relating to colour and acidity, but not for sweet taste, TSS and apple flavour. Strong genetic control that was generally shown for colour attributes predicts a rapid selection response. Most attributes were inherited quantitatively, but TA showed complicated inheritance mechanisms. / AFRIKAANSE OPSOMMING: Min inligting is beskikbaar oor die appelvoorkeure van die diverse Suid-Afrikaanse verbruikersmark wat bestaan uit verbruikers van verskillende ouderdomme en etnisiteite met verskillende voedselvoorkeure. Wit, bruin en swart verbruikers van verskillende ouderdomsgroepe is geselekteer in Stellenbosch in die Wes-Kaap, Suid-Afrika. Verbruikersvoorkeuranalise vir die eetkwaliteit en voorkoms van appels en beskrywende sensoriese analise (BSA) is uitgevoer op nege kommersiële appelkultivars. Analise van variansie (ANOVA), uitgevoer op gemiddelde voorkeurdata per ouderdom en etniese groep, het getoon dat die voorkeure van hierdie verskillende groepe oor die algemeen verskil het. Volgens Ward statistiese groepering op dieselfde datastel was die sosiodemografiese samestelling van verbruikersgroepe met soortgelyke voorkeure egter nie homogeen nie. Drie verbruikersgroepe is geïdentifiseer met soortgelyke voorkeure vir appel eetkwaliteit (E1-3) en voorkoms (V1-3). E1 het ‘n voorkeur vir fermheid, ‘n afkeur vir melerigheid en verdra suurheid. Alhoewel E1 van soet vrugte gehou het, het hulle ‘n laer voorkeur vir soetheid as E2 en E3. E2 het ‘n voorkeur vir suurheid en appelgeur terwyl E3 ‘n afkeur vir suur smaak en die hoogste voorkeur vir soetheid getoon het. Alhoewel swart en bruin verbruikers meestal ‘n renons in suur smaak getoon het en meer van hierdie verbruikers tot E3 behoort, maak swart en bruin verbruikers wat suur smaak aanvaar ongeveer 41% van die totale verbruikersgroep in die Wes-Kaap uit. Wit en jonger (<26 jaar) verbruikers was meestal in E1 en het ‘n voorkeur vir fermheid getoon. Vrugkleurvoorkeure was vir groen en pienk (tweekleurig) (V1), groen/geel en rooi gestreep (V2) en rooi (V3). Verbruikers het ‘n voorkeur gehad vir die voorkoms van kultivars waarvan hulle die eetkwaliteit verkies het. Die effek van gevestigde kleur en smaak assosiasies is getoets deur verbruikers te versoek om die eetkwaliteit van vyf kultivars te evalueer tydens aanbieding daarvan met drie vlakke van visuele inligting, naamlik geen, korrekte en verkeerde foto. Verwarring tussen verwagte en werklike eetkwaliteit het gelei tot ‘n laer voorkeur. Teling van appels is tydrowend en duur. Uitgebreide kennis van die vrugkwaliteit parameters wat verbruikersvoorkeur dryf, is noodsaaklik vir effektiewe teling. Eienskappe wat verband hou met die eetkwaliteit en voorkoms van saailinge is in vier appelfamilies geanaliseer. Die eienskappe is geassesseer deur middel van instrumentele en individuele evaluasie deur ‘n opgeleide assessor asook deur BSA deur ‘n opgeleide paneel. Instrumentele analise kon nie die vlakke van sensoriese eienskappe voorspel soos waargeneem deur die individuele assessor nie. BSA van sensoriese tekstuureienskappe, appelgeur en soetheid, en instrumentele meting van titreerbare suur (TS) en totale oplosbare vastestowwe (TOV)/TS, maar nie TOV nie, kon verbruikersvoorkeur voorspel. Die assessor wat individuele evaluasies uitgevoer het, kon nie die voorkeur van ‘n groot verbruikersgroep akkuraat voorspel nie. Kwantitatiewe genetiese analise van die data is uitgevoer en binne- en tussen-familie variasie is gekwantifiseer deur middel van ANOVA, variansie komponente en oorerflikheidsskattings. Variasie tussen families is gevind vir kleureienskappe en suurheid, maar nie vir soet smaak, TOV en appelgeur nie. Resultate het getoon dat kleureienskappe meestal aan sterk genetiese beheer onderworpe is en dit dui op vinnige vordering met seleksie vir vrugkleur. Vrugeienskappe is meestal kwantitatief oorgeërf. Oorerwing van TS blyk ingewikkeld te wees.

At present, the process of parental modelling of eating behaviours and attitudes has received limited research interest. While often mentioned as a possible factor in relation to child feeding and the development of eating behaviours, only a few studies have explicitly researched parental modelling. The main aims of this thesis were to develop a new measure to assess modelling multidimensionally and to explore the relationships between parental modelling of eating behaviours with a variety of parent and child factors. Initially, a parental self-report measure (the Parental Modelling of Eating Behaviours Scale; PARM) was developed, validated and piloted as part of a series of studies exploring the associations between modelling and a range of self-reported parental and child factors. To provide further validation for the PARM, an observational coding scheme was developed, based on the newly developed modelling measure, and this was utilised in two further studies which looked at self-reported and observed parent and child factors. The key findings from this thesis suggest a number of beneficial relationships. For example, maternal modelling was positively correlated with healthy food intake in both mothers and their children. In addition, both maternal and paternal modelling were associated with children s increased enjoyment of food and lower levels of food fussiness. Observations of maternal modelling were also found to be positively related to other observed adaptive, non-directive feeding practices, such as encouragement to eat. However, less positive relationships were also identified, with modelling being related to parents mental health symptoms and to unhealthy food intake in both mothers and their children. In conclusion, this thesis has identified three distinct facets of modelling and highlighted factors which might be linked to parental role modelling around eating behaviours. While much of the research within this thesis is exploratory, and the findings require replication, they would suggest that parental modelling has the potential to positively influence children s eating behaviours. However, parents should also be made aware of the potential detrimental effect that modelling less adaptive eating behaviours may have on their children s food intake, particularly those eating behaviours that parents may be unaware of modelling.

ITC/USA 2011 Conference Proceedings / The Forty-Seventh Annual International Telemetering Conference and Technical Exhibition / October 24-27, 2011 / Bally's Las Vegas, Las Vegas, Nevada / This paper presents a method for the optimization of mixed networks that incorporates a mixed layer optimization of performance features. The expanded integrated Network Enhanced Telemetry (iNET) system envisioned telemetering for large and complex networks which will require core telemetry networks with ad hoc extensions for coverage. Organizing such a network has been successfully accomplished in simulations using a K-mean clustering algorithm. This paper shows how the features of these network elements will be captured and disseminated in a real system. This management of network elements across multiple layers is characterized as cross-layer optimization. This paper will also show how such cross layer features can be combined for a globally optimum solution. It shows by example how the iNET system comprising multiple ground stations, gateways, frequency, nodes, and three performance measures can be optimized to achieve overall optimal system performance.