Det finns över 200 miljoner aktiva hemsidor till drygt 5 miljarder användare. I takt med attinternet har växt och utvecklingen gått kraftigt framåt har användare blivit mer kräsna överinnehållet på en hemsida.Varje gång en användare besöker en hemsida gör användaren ensnabb utvärdering av hemsidan för att avgöra om denne vill stanna på sidan eller lämna denpå en gång.Målet med denna studie är att titta närmare på hur webbdesign påverkar användarens val attstanna kvar på en hemsida eller inte. Studien inriktar sig på hur valet av typsnitt, bild och färgpåverkar användarens vilja att stanna kvar på hemsidan utifrån trovärdighet ochanvändbarhet.För att besvara forskningsfrågorna så skickades en enkät ut där deltagarna fick besvara denuppleva trovärdigheten och användbarheten utifrån tre designaspekter, typsnitt, bild och färgsamt se hur detta påverkar deltagarnas vilja att stanna kvar på hemsidan.Resultatet visar att alla tre designprinciper påverkade användarens vilja att stanna kvar påhemsidan. De designelementen som undersöktes påverkade användarens vilja att stanna kvarpå hemsidan i följande ordning från minst till störst påverkan: typsnitt, färg, bild.Sammanfattningsvis visar studien på att en användares benägenhet att stanna kvar på enhemsida kan öka om designen är tillfredsställande och genomtänkt. / There are over 200 million active websites for just over 5 billion users. As the internet hasgrown and technological development has progressed greatly, users have become more pickyabout the content of a website. Each time a user visits a website, the user makes a quickevaluation of the website to decide if they want to stay on the page or leave it at once.The aim of this study is to take a closer look at how web design affects the user's choice tostay on a website or not. The study focuses on how the choice of font, image and color affectsthe user's desire to stay on the website based on trustworthiness and usability.To answer the research questions, a survey was sent out where the participants had to answerthe perceived trustworthiness and usability based on three design aspects, font, image andcolor and see how this affects the participants' willingness to stay on the website.The results show that all three design principles influenced the user's willingness to stay onthe website. The design elements that were examined influenced the user's willingness to stayon the website in the following order from least to greatest impact: font, color, image.In summary, the study shows that a user's tendency to stay on a website can increase if thedesign is satisfactory and well thought out.

Social trading platforms are an increasingly popular venue for sharing investment ideas. We investigate if followers on these platforms herd, i.e., copy leader traders’ trades, to the extent that it affects stock markets. To do so, we use an event study to detect abnormal returns and trading volume after a trade is conducted by a leader trader. Furthermore, we investigate if the leader trader’s sentiment on the stock and the leader trader’s perceived trustworthiness have explanatory power over the herding behavior. We find signs of herding in terms of abnormal returns within thirty-minutes after the leader trader’s trade. However, we do not find signs of herding for longer event windows, nor do we find that the sentiment of the leader trader can explain herding. Finally, we find some signs that affect-based signals of trustworthiness can explain herding. However, most signals of trustworthiness are non-significant.

Greenwashing is a frequent issue within the FMCG industry, such as vague ecological claims and misleading communication. When a brand is accused of greenwashing it is important to understand how to manage this transgression, since both forgiveness and loyalty are concepts that are affected. Therefore, the purpose of this thesis was to investigate how brand forgiveness influences brand loyalty, within the FMCG industry, with the aim of contributing to limited research in this area. Ten interviews were conducted with loyal customers to examine how they felt and experienced two brands after perceived transgressions. The critical incident technique was used during the interviews to remind the customers of the transgressions and obtain a comprehensive view of their narratives. The results indicated that the lack of repair efforts influenced the customers to perceive the brands as dishonest and untrustworthy, which affected their unwillingness to forgive and damaged their trust. Based on the customer's perceptions and feelings regarding the brands and the specific transgressions, brand forgiveness and loyalty were influenced differently. In addition, the customer places high importance on the product quality and their relationship with the brand and would rather stay with the brand instead of switching to another if the quality and/or the relationship is strong.

Despite more than $769 million in charitable gifts in 2013, U.S. nonprofit organizations lost $735 million in lapsed and reduced gifts. Donor attrition is a problem for most charitable organizations, and many are using the Internet to cultivate donors. Online communication has become an important part of fundraising for many charitable organizations. The online communication factors in the current study include trust, credibility, and satisfaction. These factors may affect donor loyalty. Donor loyalty may increase or decrease donor attrition. Reducing donor attrition is important to anyone who plays a role in the success of a nonprofit organization. The purpose of the current cross-sectional quantitative study was to examine the relationship between the communication factors and the loyalty among online donors. The theoretical foundation for this study includes Bandura's social cognitive theory and Luhmann's social systems theory. Data were collected online from a random sample of online donors aged 18 years or older in the United States. Spearman correlation was used to assess the correlation between the independent and dependent variables. The results indicated there is a correlation between communication factors and loyalty among online donors in the United States. This study may help organizations communicate better with donors in an online environment and reduce online donor attrition. Reducing attrition will increase funding to a charitable organization through repeat donations, thereby helping improve finances to support the organization's mission and positively influencing societal change.

Background: The evidence-based approach to policymaking has greatly facilitated policymakers' capacity to make scientifically informed policy decisions, especially in the medical and public health contexts. However, this approach is most beneficial for policy development when trustworthy research is used. Predatory journals and publishers pose a potential threat to evidence-based policy making, since they are more likely than traditional academic journals to publish unreliable evidence. Research Objectives: The purpose of this thesis is to advance knowledge on what factors contribute to the citation of predatory journal articles in policy documents by answering the following research questions: 1) How do people preparing public health documents consider the trustworthiness of research evidence? 2) How do they source and evaluate the research evidence they cite? Methods: I identified a cross-sectional sample of public health policy documents from Overton - the world's largest policy document database - that cited articles published by the OMICS group. OMICS is a well-established predatory publisher. I extracted meta-data (e.g., document source) and document characteristics such as whether they described their method of selection or quality assessment for cited sources. Authors of these documents with contact information listed, as well as a convenience sample of people who have prepared public health documents, were invited for a semi-structured interview. I thematically analyzed these interviews by organizing the codes (both deductive and inductive) into key overarching themes. Results: Two hundred forty-two public health policy documents were included. The World Health Organization was the most common source accounting for 45 documents (19%). A total of 283 articles were cited from 126 OMICS journals. Only 54 (22%) of the policy documents described their source-selection methodology, and 22 (9%) assessed the quality of cited sources. Five key overarching themes were generated from the thematic analysis of the interview data, highlighting that information cited in policy documents is sourced and evaluated in several ways, many of which are related to a series of factors which could be contributing to the predatory journal citations. Conclusion: Public health policy documents are prepared using a variety of methods for information selection and evaluation, but the exact approach for doing so is rarely reported within the document itself. This may contribute to the reliance on untrustworthy research to inform policy; and thus, may help amplify misinformation entering policy globally. Certain steps can be taken to help minimize any potential negative impact of relying on such sources, but a better understanding of policymakers' perspectives may be required to ensure successful implementation.

<p>An increasing number of AEC firms are positioning themselves to expand operations in the international construction market. To undertake such decisions, it is critical for companies to analyze the risks, benefits, and future market potential of the host country. Further, companies typically select appropriate entry modes and develop business strategies to navigate the complex formal and informal institutions of the host country. The Board of directors and upper-management employees are typically involved in making such decisions. Previous researchers have identified various risks such as the political, financial, supply-chain using tools like the International Construction Risk Assessment Model (ICRAM) and the International Project Risk Assessment (IPRA). Such comprehensive models encompass various risk factors at the country, market, and project levels. At each level, data are to be collected from various sources that could be susceptible to political agenda, bias, and inaccessibility. Research findings suggest that the results provided by such models heavily rely on the credibility of the information and the information source.</p><p>First, this research provides comprehensive definitions for Information and Source Credibility with respect to the international construction decision-making domain. Second, the research proposes a framework to assist decision-makers in evaluating Information and Source credibility to rank and prioritize their sources. This research has identified (a) 5 Criteria and 20 Metrics for Information Credibility and (b) 3 Criteria and 18 Metrics for Source Credibility. Third, a survey of construction industry professionals was carried out to determine the relevancy and relative importance of the identified criteria and metrics with respect to international construction decisions. The collected survey responses have a Cronbach’s Alpha value of 0.89 which indicates that the responses are reliable for further analysis. A one-way ANOVA test was conducted to investigate the influence of the prior credibility evaluation experience of the respondents on the relative weights for the framework. The results indicate that prior evaluation experience does not significantly affect the perception of the users towards selecting and weighing the criteria and metrics.</p><p>Finally, a multi-criteria decision-making (MCDM) model is proposed to incorporate the proposed credibility evaluation framework with the relative weights of the criteria and metrics. An Excel-based macro-enabled tool (with a user-friendly interface) is developed to accept user inputs associated with their information and sources. A well-defined scale of measurement is developed to reduce the subjectivity and biases involved in making such abstract assessments. Final scores, Cr_Information and Cr_Source, are calculated for every piece of information and source selected by the user. The MCDM results are expected to assist users in determining the undertaking of robust decisions by incorporating one additional layer of determining the credibility of their information and prioritizing their information sources. The suggested point of application for this research is before Financial Investment Decision (pre-FID) stage to ensure sufficient time for all the stakeholders to reevaluate their FIDs.</p>

<p>  </p> <p>Federated learning (FL) has emerged as a promising approach for training machine learning models on distributed data while ensuring privacy preservation and data locality. However, one key challenge in FL optimization is the lack of high probability guarantees, which can undermine the trustworthiness of FL solutions. To address this critical issue, we introduce Federated Averaging with post-optimization (FedAvg-PO) method, a modification to the Federated Averaging (FedAvg) algorithm. The proposed algorithm applies a post-optimization phase to evaluate a short list of solutions generated by several independent runs of the FedAvg method. These modifications allow to significantly improve the large-deviation properties of FedAvg which improve the reliability and robustness of the optimization process. The novel complexity analysis shows that FedAvg-PO can compute accurate and statistically guaranteed solutions in the federated learning context. Our result further relaxes the restrictive assumptions in FL theory by developing new technical tools which may be of independent interest. The insights provided by the computational requirements analysis contribute to the understanding of the scalability and efficiency of the algorithm, guiding its practical implementation.</p>

Research indicates that students perform better academically in schools with higher levels of trust than in schools with lower levels of trust. School leaders are primarily responsible for building cultures of trust but are often at a loss as to how to do so effectively. With the assumption that as perceptions of school leader trustworthiness improve, teachers will be more likely to place their trust in that school leader, this research seeks to clarify how Ugandan headteachers improve teachers' perceptions of headteacher trustworthiness. In particular, we examined how specific types of headteacher visibility related to teachers' perceptions of headteacher relational and competence trustworthiness. This qualitative research used grounded theory methodology to interpret and analyze the interview responses of 28 Ugandan secondary school teachers in eight schools in Mukono District, Uganda. Findings from this research suggest that teachers' perceptions of headteacher relational trustworthiness were strongly related to both the level of risk and formality of headteacher visibility. Additional findings suggest that perceptions of both headteacher relational and competence trustworthiness were influenced by differences in teacher and headteacher personal characteristics. The final finding indicates that certain types of headteacher visibility moderated the influence that teacher characteristics have on perceptions of trustworthiness. These findings can inform school leaders about how to more effectively improve teachers' perceptions of school leader trustworthiness. The grounded theory model presented will provide opportunities for further theory building and testing with respect to the relationship between school leader visibility and teachers' perceptions of school leader trustworthiness.

<p>The Stereotype Content Model was used to challenge the untested assertion of the social identity/self-categorization perspective that different others will be seen as untrustworthy. Results from a sample of 29 student teams showed that trustworthiness perceptions were positively related to cognitive and bio-demographic diversity. Member satisfaction was also positively related to cognitive diversity. Trustworthiness perceptions mediated the relationship between team diversity and members’ satisfaction. First round satisfaction predicted second round team performance. The results suggest that outgroup bias is not automatic; teams can be successful if members see each other as trustworthy, regardless of cognitive or bio-demographic diversity.</p> / Doctor of Philosophy (PhD)

Due to the rapid digitalisation of mobility, and the emergence of cross-disciplinary businesses within the transport sector, transcending it from a traditionally product focused industry, into a service one. Therefore, the contextual understanding of mobility has become of even greater importance. With future mobility bringing possible solutions to a range of social problems it’s vital for researchers to understand how the adoption of these services can become even more substantial. By doing so, hopefully curtailing the negative effects that urbanisation has on society. This paper puts forward a method of how to involve young Swedish people in the development of future digitalised mobility (FDM). From the synthesis of insights from literature on adoption of future mobility and new intelligent technology a participatory design approach was used to explore how coming user perspective can be counted for in the development of FDM.