I explore whether participation on Estimize.com, a crowdsourced earnings-forecasting platform aimed primarily at novices, improves professional analysts’ forecast accuracy and career outcomes. Estimize provides its contributors with frequent and timely feedback on their forecast performance and offers them a new channel for disseminating their forecasts to a wider public, features that could help analysts improve their forecast accuracy and raise their online visibility. Using proprietary data obtained from Estimize and a difference-in-differences research design, I find that IBES analysts who are active on Estimize improve their EPS forecast accuracy by 13% relative to the sample-mean forecast error, as well as reduce forecast bias. These improvements in performance vary predictably in ways consistent with learning through feedback. Additionally, I find increased market reaction to the positive earnings-forecasts revisions issued by analysts who are active on Estimize. I also find that analysts active on Estimize enjoy incremental positive career outcomes after controlling for forecast accuracy. My results suggest that professional analysts can learn to become better forecasters through online feedback and consequently garner more attention from the market. My results also suggest analysts can improve their career outcomes by gaining additional online visibility. / Business Administration/Accounting

The current study assessed an error reflection intervention on Algebra I students’ conceptual and procedural knowledge and sense of belonging to mathematics. Also of interest was whether perceptions of the functionality of errors mediated the effect of condition on learning and sense of belonging to mathematics. Middle school students (N = 207) were randomly assigned within classroom to one of four conditions: 1) a Problem-Solving Control group, 2) a Correct Examples Control group, 3) a Correct Examples Error Reflection condition that promoted reflection on hypothetical errors through self-explanation prompts, or 4) an Incorrect Examples Error Reflection condition that promoted reflection on displayed errors within the example through self-explanation prompts. Conceptual and procedural knowledge, sense of belonging to mathematics and perceived functionality of errors were measured pre- and post-intervention. After controlling for unanticipated clustering effects, results suggest that reflecting on and explaining errors within a worked examples intervention is just as effective at promoting learning as traditional problem solving alone or working with traditional correct worked examples and written self-explanation prompts. Students’ sense of belonging to mathematics or perceived functionality of errors for learning were high at the start of the study and remained so throughout the intervention. Perceptions of the functionality of errors were unrelated to learning and sense of belonging to mathematics. The limited size of the minority population in the sample did not allow for exploration of differential effects of condition for underrepresented minority (URM) students. However, these students reported lower feelings of belonging to mathematics than non-URM students. Implications for theory and practice are discussed. / Educational Psychology

Within dynamic 21st century knowledge economies, future-building knowledge, that bears capacities to transcend existing boundaries and create something new, is of particular importance. Within the first decade of the new century, new concepts such as "learning from the future" or "self-transcending knowledge" developed within knowledge management. So far, they lacked a theoretical grounding in relevant learning theory as well as a sound acknowledgement and consideration of such knowledge structures' emergence and social embeddedness. Thus, key principles and leverage factors for designing respective knowledge processes were difficult to derive. This dissertation investigates theoretical ground that can provide a basis to explain the creation of future-building knowledge in collective structures. It is guided by the following research question: "How can the emergence of self-transcending knowledge in collective organizational settings be rooted in theories of knowledge, learning and cognition?" Starting from the model of knowledge-based management, the model is expanded by exploring cognitive, creative and social systemic aspects of knowledge creation on a transdisciplinary basis. Research draws on constructivist learning theory, complexity-based approaches in knowledge management and organizational learning, recent accounts in cognitive science (enaction/embodiment) and a creative logic of emergence to derive an integrated model for collective self-transcending knowledge. The model contributes to the integration of knowledge management, organizational learning and cognitive science, expanding knowledge-based management towards attention-based management. The model's three dimensions and three domains form an integrated theoretical basis to derive key principles and leverage factors for steering future-building knowledge processes. Simultaneously, they reveal leverage factors' limited - i.e. enabling, not determining - impact on processes of "organizing future". (author's abstract)

Achieving robust and reliable autonomous operation even in complex unstructured environments is a central goal of field robotics. As the environments and scenarios to which robots are applied have continued to grow in complexity, so has the challenge of properly defining preferences and tradeoffs between various actions and the terrains they result in traversing. These definitions and parameters encode the desired behavior of the robot; therefore their correctness is of the utmost importance. Current manual approaches to creating and adjusting these preference models and cost functions have proven to be incredibly tedious and time-consuming, while typically not producing optimal results except in the simplest of circumstances. This thesis presents the development and application of machine learning techniques that automate the construction and tuning of preference models within complex mobile robotic systems. Utilizing the framework of inverse optimal control, expert examples of robot behavior can be used to construct models that generalize demonstrated preferences and reproduce similar behavior. Novel learning from demonstration approaches are developed that offer the possibility of significantly reducing the amount of human interaction necessary to tune a system, while also improving its final performance. Techniques to account for the inevitability of noisy and imperfect demonstration are presented, along with additional methods for improving the efficiency of expert demonstration and feedback. The effectiveness of these approaches is confirmed through application to several real world domains, such as the interpretation of static and dynamic perceptual data in unstructured environments and the learning of human driving styles and maneuver preferences. Extensive testing and experimentation both in simulation and in the field with multiple mobile robotic systems provides empirical confirmation of superior autonomous performance, with less expert interaction and no hand tuning. These experiments validate the potential applicability of the developed algorithms to a large variety of future mobile robotic systems.

台灣向來展現中小企業旺盛的生命力，近年來國際性評比的報告中指出台灣的經商環境與創業精神皆名列世界前茅，然而失敗是創業活動不可避免的一環，愛迪生曾說過：失敗也是我需要的，它與成功對我一樣有價值。既然人們說失敗為成功之母，創業家實際面對時是否能相信失敗具有價值？向失敗學習日趨重要，市面上亦增加許多向失敗學習以取代傳統成功典範的科普書籍，畢竟成功無法複製，卻能以失敗借鏡。從失敗的經驗累積未來進取成功的機會。本研究以曾有失敗或挫折經驗的創業家作為訪談對象，以深度訪談的方式，了解創業家失敗或挫折過後，針對前次經驗作內外部歸因的分析，進而讓創業家執行行為修正。 　　經過深度訪談與資料分析，本研究得出以下三點結論：(1) 面對原來的挫折時，所啟動的「敗因歸納－學習－反思與行為改變」歷程有下列影響：( i ) 內外部歸因前次的挫折，使創業家得到經驗學習的對象，進行反思。( ii ) 以前次的挫折為基礎，修正自我的行為，改變成不再犯同樣錯誤的。( iii ) 歷程過後，反思後得到內化後的能力將成為面對下次創業的依據之一。(2) 歷程中的行為改變可能引發控制錯覺，而成為下一次創業的失敗原因之一：當創業家在「敗因歸納－學習－反思與行為改變」歷程時，將針對前次挫折經驗視為學習與修正自我的對象，正因為創業家專注在解決與修正前一次的行為，使得歷程中的行為改變中，隱藏了下一次創業活動的行為盲點，而成為了當下檢討歷程的控制錯覺之一。(3)研究發現內外部的歸因與內外部資源有關聯：歷程中將內外部歸因以表格式的分類與分析，將敗因學習與行為修正所得到的回饋，轉而成為創業家實際的自我價值與資源。 / Small and medium enterprises in Taiwan have shown people their great vitality. Recently, the international business reports pointed that the entrepreneurial environment in Taiwan and Taiwanese entrepreneurship are almost on the top world. However, it is unavoidable to make mistakes when people stay in entrepreneurial process. There is a quotes spoken by Thomas A. Edison, “Negative results are just what I want. They're just as valuable to me as positive results.” People say that failure is the mother of success whether can entrepreneur really treat failure as a valuable experience? 　　 　　It has been more and more important to learn from failure. We can easily find lots of books talking about this topic. After all, success can’t be copied but people can be able to learn from failure. People can accumulate experiences from failure and make themselves to be success. This research examines how entrepreneurs do with reflection after being failure. Through interviewing 5 entrepreneurs who had been suffered from failure or frustrated experiences, researcher tried to understand the relationships between those experiences and entrepreneur’s behaviors by attribution theory.Through in-depth interviews and data analysis, there are three main findings. First, the process in “attribute frustration - learning from failure - reflection and behavior adjustment” has 3 significant impacts on the behavior of previous setbacks which help entrepreneur achieve. Second, the behavior adjustment in the process may trigger the illusion of control, which may become one of the failure reasons for the next entrepreneurship. Last, according to the research, the dimensions of attribution and the dimensions of entrepreneur’s resources have connection.

This paper carry out a qualitative meta-analysis of 112 “proposals for action” identified in 30 of the approximately 630 accident investigations that are published on the Swedish Civil Contingencies Agency (MSB) website. Accident investigations have been carried out at the discretion of each municipality and then sent to the MSB, which in turn, after a secrecy review and an ethical review, publishes most of the investigations on the web page Kommunala olycksundersökningar. The accident investigations are split into a number of different categories by type of event and the categories analysed in this paper are “Automatic alarm - not fire” and “Fire - not in building”.The process of learning from accidents can be illustrated using the CHAIN model (Reporting - selection - Investigation - spread - implementation). This model attempts to show how the process of learning from accidents step by step and point out that every step must be followed and implemented for a lesson to be learned - from event to implemented lesson learned. Previous studies however have found that the steps in the CHAIN model are not followed from the beginning to the end in terms of learning from municipal accident investigations. A first problem is that some proposals listed can be unclearly formulated in terms of who is supposed to carry out the proposal, but above all, there are weaknesses in distribution and thus also the implementation of the proposals.Aim and method This paper seeks to compile, analyse and present the proposals for action contained in the selected accident investigations. The method used is a qualitative meta-analysis understood here as an "analysis of analyses" performed with the qualitative method content analysis, which may also contain quantitative elements.Findings Almost all of the proposals set out are alone in its kind. There is just two proposals contained more than once. Proposals for measures can be further divided into categories by thought, or stated, receivers and for similarities between the proposals. More than half of all proposals are targeted to Rescue services own work, either regarding the intervention itself, organizational or other planning or collaboration with other agencies or stakeholders. One group is aimed towards the operator and propose improvements in various parts of their systematic fire prevention (SBA). Two smaller groups of proposals suggests improvements for handling errors and to do changes in the products and that this information should be communicated to operators/retailers or producer. Some proposals do not fall within any of the other categories.It noted that, in principle, all suggestions are workable and implementable and that these proposals can be generalized beyond its original context into a larger one. And that this kind of qualitative meta-analysis can be a part of the CHAIN model's implementation. / I denna uppsats görs en kvalitativ metaanalys av 112 förslag på åtgärder vilka anges i 30 av de runt 630 kommunala olycksundersökningar som finns publicerade på Myndigheten för samhällsskydd och beredskaps (MSB) hemsida. Olycksundersökningarna har genomförts efter beslut i respektive kommun och därefter skickats in till MSB, vilka i sin tur, efter en sekretessprövning och en etisk prövning, publicerar de flesta på webbsidan Kommunala olycksundersökningar. Olycksundersökningarna delas upp på ett antal olika kategorier efter typ av händelse och de kategorier vilka analyseras i denna uppsats är Automatlarm – ej brand samt Brand – ej i byggnad.Processen att lära från olyckor kan åskådliggöras med hjälp av CHAIN-modellen (rapportering – urval – utredning – spridning – genomförande). Denna modell vill visa på hur processen med lärandet från olyckor går till steg för steg samt poängtera att varje steg måste följas och genomföras för att ett lärande ska komma i mål – från händelse till implementerad lärdom. Tidigare studier har dock funnit att CHAIN-modellens steg inte följs från början till slut vad gäller lärandet från kommunala olycksundersökningar. Ett första problem är att vissa förslag som anges kan vara otydligt formulerade vad gäller vem som ska genomföra förslaget, men framförallt finns det brister i spridningen och således också genomförandet av förslagen.Syfte och metod Denna uppsats syfte är att sammanställa, analysera och presentera de förslag på åtgärder som återfinns i de utvalda olycksundersökningarna, samt att värdera dessa utifrån generaliserbarhet. Metoden som använts är en kvalitativ metaanalys med vilket här avses en ”analys av analyser” utförd med den kvalitativa metoden innehållsanalys, vilken också kan innehålla kvantitativa inslag.Resultat Nästan alla de förslag på åtgärder som anges i olycksundersökningarna är ensamma i sitt slag, det är bara 2 förslag som återfinns mer än en gång. Detta beror troligen på att en utökad olycksundersökning görs först om utredningen bedöms kunna tillföra ny information. Förslagen på åtgärder kan vidare delas in i kategorier efter tänkt, eller angiven, mottagare samt efter likheter mellan förslagen. Men än hälften av alla förslag är riktade mot Räddningstjänstens eget arbete, antingen gällande själva insatsen, organisatoriskt eller annat planeringsarbete eller gällande samverkan med andra myndigheter eller aktörer. En grupp förslag riktar sig mot verksamhetsutövaren och föreslår förbättringar i olika delar av dessas Systematiska brandskyddsarbete. Två mindre grupper förslag ger förslag på förbättringar efter handhavandefel respektive på förändringar av produkter samt att denna information ska delges verksamhetsutövare/återförsäljare eller producent. Några förslag faller inte inom någon utav de övriga kategorierna.Konstateras görs också att i princip alla förslag är konkreta och genomförbara samt att dessa förslag kan generaliseras utanför sitt ursprungliga sammanhang till en större kontext. Samt att denna typ av kvalitativ metaanalys kan vara ett led i CHAIN-modellens genomförande.

The overall aim of this study is divided into two parts, firstly to analyze the view of four swedish upper secondary school students on what the meaning with religious education in Sweden should be. Secondly, to analyze how these four students experience their education in religion and compare it with their teachers view on the meaning of religious education and how she conducts the teaching. It is an empirical study and the results are based on semistructural interviews made individually with each participant. The analytical approach is a hermeneutic analysis with interpretation as the main focus. The findings and the result of the study indicates that the swedish students view the religious education as an important subject to understand and explore their own worldviews, other peoples views, traditions and differences. The students and the teacher have similar thoughts on what the meaning with religious education should be, but in practise the students lack an inside perspective of religions, a meeting with people who identify themselves as religious. The teacher however does not include this in her education, nor does she speak of this during the interviews. The results also show that the swedish students thoughts about religious education is very similiar to what previous research has shown is lacking in the swedish education in religion. In summary, both the students and previous research points out that the inside perspective in religious education is very important to be able to reach the goals of the curriculom in the subject, which include a development of understanding different religions and people.

Med utgångspunkt i min bristande notläsningsförmåga är syftet med föreliggande studie att undersöka vilka resurser utöver själva noterna som jag använder vid inlärningen av ett noterat musikstycke. Detta sker via videoobservation, dikterad loggbok, samt insamling av tidigare data. Tre olika övningspass á 20 minuter ligger till grund för resultatet. Dessa analyserades och bearbetades utifrån ett designteoretiskt perspektiv. I resultatet beskrivs vilka resurser jag använde, hur de användes, samt varför de användes. I diskussionsdelen belyses resultatet utifrån ett designteoretiskt perspektiv, jag resonerar kring behovet av att använda dessa resurser, samt diskuterar orsaken till min problematik och vägen framåt. / Based on my lacking ability to read sheet music, the purpose of this study is to examine what resources beyond the music notation I use in the process of learning a piano piece from sheet music. This is done through video observation, a dictated journal, and collection of previously available data. Three separate practice sessions of 20 minutes each is the basis for the result. These were analyzed and processed based on a design theoretical perspective. The result describes which resources I used, how they were used, as well as why they were used. In the concluding discussion section, I illuminate the result from the perspective of design theory, I argue regarding the need to use these resources, as well as discuss the cause of my problems and a possible way forward.

My thesis is entitled "Contributions to Simulation-based High-dimensional Sequential Decision Making". The context of the thesis is about games, planning and Markov Decision Processes. An agent interacts with its environment by successively making decisions. The agent starts from an initial state until a final state in which the agent can not make decision anymore. At each timestep, the agent receives an observation of the state of the environment. From this observation and its knowledge, the agent makes a decision which modifies the state of the environment. Then, the agent receives a reward and a new observation. The goal is to maximize the sum of rewards obtained during a simulation from an initial state to a final state. The policy of the agent is the function which, from the history of observations, returns a decision. We work in a context where (i) the number of states is huge, (ii) reward carries little information, (iii) the probability to reach quickly a good final state is weak and (iv) prior knowledge is either nonexistent or hardly exploitable. Both applications described in this thesis present these constraints : the game of Go and a 3D simulator of the european project MASH (Massive Sets of Heuristics). In order to take a satisfying decision in this context, several solutions are brought : 1. Simulating with the compromise exploration/exploitation (MCTS) 2. Reducing the complexity by local solving (GoldenEye) 3. Building a policy which improves itself (RBGP) 4. Learning prior knowledge (CluVo+GMCTS) Monte-Carlo Tree Search (MCTS) is the state of the art for the game of Go. From a model of the environment, MCTS builds incrementally and asymetrically a tree of possible futures by performing Monte-Carlo simulations. The tree starts from the current observation of the agent. The agent switches between the exploration of the model and the exploitation of decisions which statistically give a good cumulative reward. We discuss 2 ways for improving MCTS : the parallelization and the addition of prior knowledge. The parallelization does not solve some weaknesses of MCTS; in particular some local problems remain challenges. We propose an algorithm (GoldenEye) which is composed of 2 parts : detection of a local problem and then its resolution. The algorithm of resolution reuses some concepts of MCTS and it solves difficult problems of a classical database. The addition of prior knowledge by hand is laborious and boring. We propose a method called Racing-based Genetic Programming (RBGP) in order to add automatically prior knowledge. The strong point is that RBGP rigorously validates the addition of a prior knowledge and RBGP can be used for building a policy (instead of only optimizing an algorithm). In some applications such as MASH, simulations are too expensive in time and there is no prior knowledge and no model of the environment; therefore Monte-Carlo Tree Search can not be used. So that MCTS becomes usable in this context, we propose a method for learning prior knowledge (CluVo). Then we use pieces of prior knowledge for improving the rapidity of learning of the agent and for building a model, too. We use from this model an adapted version of Monte-Carlo Tree Search (GMCTS). This method solves difficult problems of MASH and gives good results in an application to a word game.

Le retour d'expérience (REX), sous la forme d'analyses d'accident le plus souvent, constitue un instrument privilégié de management de la sécurité dans les organisations industrielles à hauts risques. Cette recherche s'inscrit dans le domaine de la sécurité dans les tunnels routiers et vise à proposer une approche intégrée du REX, qui repose sur le travail quotidien de gestion de la sécurité par les opérateurs en s'appuyant sur les espaces collectifs de construction de l'expérience. La thèse défendue est que la sécurité dans les environnements dynamiques tels que les tunnels routiers repose sur la capacité des collectifs de travail transverses à faire face aux imprévus, les perturbations quotidiennes comme les évènements. Cette capacité se construit notamment à travers les discussions entre opérateurs et avec leurs managers sur les difficultés de leur travail. Les dispositifs de REX doivent favoriser l'existence de débats sur l'activité pour améliorer la gestion de la sécurité et favoriser ainsi le développement de l'individu et de l'organisation. Cependant cela implique des conditions d'organisation pour que le débat soit possible et pérenne.Une intervention-recherche visant la conception d'un dispositif de REX fondé sur l'activité individuelle et collective a été réalisée chez un exploitant de tunnels routiers pour éclairer cette problématique. Les analyses menées en collaboration avec les opérateurs de l'exploitation avaient pour objectifs d'aider les opérateurs à mettre en relation la manière dont ils gèrent la sécurité sur le terrain avec la manière dont le REX traite les évènements. L'analyse du REX existant montre dans un premier temps qu'il est centré sur les évènements de trafic majeurs, faisant l'objet de procédures et de règles formelles, alors que les incidents critiques pour les opérateurs sont majoritairement des incidents d'exploitation, peu cadrés. Dans un second temps, l'analyse de l'activité montre que les opérateurs développent de façon informelle des instances d'échange collectif pour anticiper les aléas, s'organiser et analyser les évènements. Une démarche de simulation organisationnelle avec les opérateurs, les managers et les partenaires de la recherche a été engagée pour transformer le dispositif de REX. Elle débouche sur un REX élargi en termes de définition des évènements et d'acteurs impliqués. Le nouveau dispositif s'appuie sur les processus-métiers existants, tout en développant les instances collectives d'analyse inter-métiers et inter-organisations. Il structure les pratiques informelles d'organisation et de partage d'expérience des opérateurs. Des espaces de discussion inter-métiers sur les pratiques réelles ont pu s'élaborer à partir de simulations d'évènement, animés par des managers de proximité. L'expérimentation de cette méthodologie montre que ces espaces favorisent la confrontation des représentations, des pratiques et la transmission des connaissances. Ils permettent aussi l'élaboration de règles opérationnelles transverses et font émerger les besoins de règles formelles à relayer aux managers. Ils sont constructifs car ils participent au développement de l'organisation formelle autant que de l'activité des opérateurs. Cette méthodologie a favorisé la construction de l'expérience à partir de l'analyse des évènements.L'intervention-recherche a permis de revenir sur les classifications officielles des évènements dans les tunnels routiers. Les conditions et modalités de mise en place d'un REX intégré sont définies et discutées, ainsi que celles des espaces de discussion par simulation. L'ingénierie de l'intervention ergonomique de conception organisationnelle et la place de l'intervenant sont aussi revisitées à l'aune de nos résultats. / Operating experience feedback, mainly through accident analysis is a method of choice for safety management in high risk organizations. The present research takes place in the field of road tunnel safety, proposing an integrated approach of experience feedback from the daily management of safety by tunnel monitoring staff through collective spaces for discussion and experience building.Our hypothesis is that safety in such dynamic environments as road tunnels depends on the capacity of operating teams to face the unexpected events, disturbances and accidents. This capacity is built through discussions between operating staff and with the management about the difficulties encountered during work. The return on operating experience should allow the debate on work activity in order to reach a better safety management, foster individual development and a better organization. However, there are organizational conditions for the debate to be possible and sustainable.A research and intervention protocol was set up with a road tunnel operator to test and improve a scheme for the return on operating experience based on individual and collective activity. Analyses were conducted to help operational staff link the way events are treated in day to day security management with the processing in the operational experience feedback scheme. An analysis of the current scheme showed that it is focussed on major traffic events, comprising of a set of procedures and formal rules, while the operational events, felt as critical by operators do not have a framework for discussion. Further, work activity analysis shows that operational staff develops informal spaces of collective exchanges to organize work, anticipate hazards and analyse events.Organizational simulations including operating staff, managers and the research partners were set up to transform the feedback scheme. The simulation led to an enlarged framework for feedback, defining the events to be analysed and participants to be included. The new scheme roots in the existing professional processes, develops inter-professional as well as inter-organization discussion spaces. The new scheme facilitates the sharing of experience among operating staff and structures the informal organizational practices we observed.Inter-professional discussion spaces, conducted by the team managers, were set up based on the analysis of simulated events. These structured discussion spaces proved to be effective in the confrontation of work practices, events and roles representations as well as conducive for knowledge transmission. The discussion space allows setting transverse operational rules and highlights the need for formal rules to be set by the management. The inter-professional discussion spaces are operant in improving the formal organisation as well as developing the activity of operational staff. This methodology favoured the building of experience from event analysis.This research-intervention was an opportunity to review the official classification of road tunnel events. The conditions for an integrated operating experience feedback scheme are set and discussed, as well as the implementation of simulation-based discussion spaces.From our results, we re-consider the design of interventions in organizational design, as well as the role of the intervening ergonomist.