When interviewing school students for what constitutes a good mathematics teacher, the first characteristic usually listed is the ability to explain well. Besides well-founded content knowledge most important for classroom episodes of teacher explanations is knowledge about how to present mathematical concepts in a comprehensible way to students. This encompasses competencies in the area of verbal communication as well as the conscious use of means for illustrating and visualising mathematical ideas. We report about an analysis of explanatory processes in math lessons and about an analysis of prospective teachers\' explanatory competencies. As a result we identify improvements in teacher education at university.

This study examined people’s attitudes towards killing bugs and their bug-killing behaviour in the context of nature vs. nurture explanations of bug killing. Previous research shows that exposure to genetic (i.e., nature) explanations could have undesirable effects on people’s attitudes and behaviour, compared to the exposure to experiential(i.e., nurture) explanations. Genetic explanations for killing may affect attitudes towards killing and killing behaviour, because they suggest that killing behaviour is predetermined or programmed by nature. Such explanations may also be used by individuals to overcome guilt and dissonance from prior killing or killing in which they are about to participate. This study tested the idea that exposure to genetic explanations for bug killing would lead people to view killing bugs as more morally acceptable, as well as lead them to kill more bugs. A sample of university students was randomly assigned into three conditions, in which they read either genetic or experiential explanations for why people kill bugs or read a neutral passage. The study utilised a procedure in which participants were led to believe that they were killing bugs (although in actuality no bugs were killed), to observe their killing behaviour in a self-paced killing task. Half of the participants were also asked to kill a bug prior to the self-paced killing task. Results showed that participants who read genetic explanations viewed bug killing as more morally acceptable, compared to participants who read experiential explanations, and this occurred particularly among those who engaged in the prior killing task. However, no similar effects emerged for the number of bugs killed, though there was a positive correlation between the moral acceptance of bug killing and the number of bugs killed. Implications of genetic explanations with respect to aggression and killing are discussed.

abstract: Writing scientific explanations is increasingly important, and today's students must have the ability to navigate the writing process to create a persuasive scientific explanation. One aspect of the writing process is receiving feedback before submitting a final draft. This study examined whether middle school students benefit more in the writing process from receiving peer feedback or teacher feedback on rough drafts of scientific explanations. The study also looked at whether males and females reacted differently to the treatment groups. And it examined if content knowledge and the written scientific explanations were correlated. The study looked at 38 sixth and seventh-grade students throughout a 7-week earth science unit on earth systems. The unit had six lessons. One lesson introduced the students to writing scientific explanations, and the other five were inquiry-based content lessons. They wrote four scientific explanations throughout the unit of study and received feedback on all four rough drafts. The sixth-graders received teacher feedback on each explanation and the seventh-graders received peer-feedback after learning how to give constructive feedback. The students also took a multiple-choice pretest/posttest to evaluate content knowledge. The analyses showed that there was no significant difference between the group receiving peer feedback and the group receiving teacher feedback on the final drafts of the scientific explanations. There was, however, a significant effect of practice on the scores of the scientific explanations. Students wrote significantly better with each subsequent scientific explanation. There was no significant difference between males and females based on the treatment they received. There was a significant correlation between the gain in pretest to posttest scores and the scientific explanations and a significant correlation between the posttest scores and the scientific explanations. Content knowledge and written scientific explanations are related. Students who wrote scientific explanations had significant gains in content knowledge. / Dissertation/Thesis / M.A. Curriculum and Instruction 2011

Magister Artium - MA (Psychology) / Antisocial offending behaviour relate to instances of criminal and societal norm violating behaviour. In South Africa, statistics on antisocial offending behaviour reflect high incidence rates among South African young people who engage in societal norm violations that are largely aggressive and criminal in nature. In the country's prisons, the average inmate is a young substance abuser who is unemployed, has dropped out of school before high school, is functionally illiterate and, more often than not, homeless. This is particularly true for emerging adults (those between the ages of 18 and 25 years) who constitute the poorest economic age group, the average house burglar as well as make up more than a quarter of the prison population in South Africa. In South Africa as well as internationally, the antisocial offending behaviour of emerging adults remain grossly understudied especially through qualitative self-explanations. The present study utilised a qualitative research approach to explore experiences and self-explanations regarding antisocial offending behaviours of a group of 10 South African emerging adults. The study used an exploratory research design. Participants were drawn from a number of non- government organisations, and two townships in Cape Town which is located in the Western Cape Province of the Republic of South Africa.

The aim of this study is to examine how teachers explain mathematical tasks in the mathematical education in grade 2, and also which perception they have of explanations and their own teaching. The survey is supported by theories that focuses on explanations in the education. These are Vygotsky´s theory of the development of the scientific concepts, general and literary concepts, proximal zone and scaffolding. To achieve the aim, the following questions has been formulated: How do teachers explain mathematical tasks in the mathematical education in grade 2? How do teachers speak about explanations and their own teaching? In summary, the results of this study are that teachers explain mathematical tasks in the mathematical education in grade 2 by using so called mathematical, daily and visual explanations when explaining these tasks in their education. Beyond this, the results of this study are also that teachers speak about explanations as something clear and concrete. They also define explanations as something that promote for student development and learning, and this is something the teachers in this study want to achieve in their own education when explaining mathematical tasks.

Master of Arts / Department of Economics / James F. Ragan Jr / This paper examines some common explanations for the earnings gap between males and females. Over recent decades, the average pay of women has increased faster than the average pay of men; however, a substantial earnings gap remains. As of 2006, the U.S. Census estimated that for year-round full-time workers the earnings ratio of women to men was 77%; in other words, for every one dollar a man earns, a woman earns $0.77. The wage gap likely consists of both non-discriminatory and discriminatory aspects, and concern remains over how much of the gender wage gap is caused by discrimination against women. However, the part of the wage gap due to discrimination cannot be measured directly, so it is typically interpreted as the portion of the gap that is "unexplained" by other factors. Numerous economists and sociologists have studied this issue, but their conclusions differ vastly. This paper discusses various economic explanations for the gender pay gap, both discriminatory and non-discriminatory. It also briefly summarizes some sociological responses to economic arguments, as well as some policy recommendations and their possible implications.

Explainable AI (XAI) is a research field dedicated to formulating avenues of breaching the black box nature of many of today’s machine learning models. As society finds new ways of applying these models in everyday life, certain risk thresholds are crossed when society replaces human decision making with autonomous systems. How can we trust the algorithms to make sound judgement when all we provide is input and all they provide is an output? XAI methods examine different data points in the machine learning process to determine what factors influenced the decision making. While these methods of post-hoc explanation may provide certain insights, previous studies into XAI have found the designs to often be biased towards the designers and do not incorporate necessary interdisciplinary fields to improve user understanding. In this thesis, we look at animal classification and what features in animal images were found to be important by humans. We use a novel approach of first letting the participants create their own post-hoc explanations, before asking them to evaluate real XAI explanations as well as a pre-made human explanation generated from a test group. The results show strong cohesion in the participants' answers and can provide guidelines for designing XAI explanations more closely related to human reasoning. The data also indicates a preference for human-like explanations within the context of this study. Additionally, a potential bias was identified as participants preferred explanations marking large portions of an image as important, even if many of the important areas coincided with what the participants themselves considered to be unimportant. While the sample pool and data gathering tools are limiting, the results points toward a need for additional research into comparisons of human reasoning and XAI explanations and how it may affect the evaluation of, and bias towards, explanation methods.

In this paper, the author examines the concept of explorable explanations. It has emerged as a genre of educational software within the last 7 years, yet descriptions of it are vague at best. The author works with the genre through a generic design approach that consists of an analysis of existing explorables and the design of three iterations of the author's own explorable explanation on the topic of neural networks. 22 examples, of which 9 are presented in-depth, are analyzed with educational theory and games research theory as tools. It is found that explorable explanations tend to be digital experiences with a high degree of interactivity that attempt to teach facts, concepts and procedures to the user. Furthermore, the author embarks on a design process of creating explorable explanations of their own to understand what can be relevant when designing and evaluating an explorable explanation. The paper is concluded with reflections on the employed method in the project. Future work is also briefly outlined about what impact the analysis and design work can have on the practice of other designers seeking to work with the genre, as well as to other researchers.

Students will very often write incomplete explanations when asked to do so. This may be due to various reasons, including but not limited to a lack of understanding of the content or a lack of motivation. Developing an understanding of how students form and orally express scientific explanations (i.e., participating in scientific discourse) and comparing that to written explanations will give insight into how students communicate the construction of their scientific explanations. This dissertation used sociocultural theory as the framework to study investigate the transcribed conversations of student work groups as they complete Model-Evidence Link (MEL) diagram activities and then write evaluative explanations of the evidence to model relationships (Lombardi, Bailey et al., 2018).Transcripts coded with the sociocultural discourse analysis (SDA) framework and the Lombardi, Brandt et al. (2016) evaluation rubric provided the primary data for this project, as well as data previously collected as part of the MEL Project. Structural equation modeling investigated the impact of the Quality of Conversation (QoC) construct on final MEL Project structural equation model (tSEM). Multivariate analysis was used to determine the trends of change over time for the evaluation scores. Finally, an adapted text analysis framework was used to make comparisons between transcripts and the written explanation tasks. Analysis found that the pathway from QoC to post-plausibility through evaluation in the experimental SEM was significant with a good model fitness (Tenenhaus GoF= 0.349, medium > 0.25). Analysis of the evaluation scores of the students showed no particular growth trends and qualitative analysis supported previous findings that the students use descriptive levels of evaluation when talking about and writing about the relationships between explanatory models and lines of evidence. This project found that the use of the MEL diagram may enhance student levels of evaluation when investigating the links between lines of evidence and explanatory models. However, a qualitative analysis of the student transcripts finds that students tend to use the basest level of evaluation (i.e., descriptive) in their discourse. This is consistent with the level of evaluation found in the written explanation tasks. This led me to suggest to teachers that they continue to support their students’ group conversations with discourse scaffolds. / Teaching & Learning

Videos manipulated by machine learning have rapidly increased online in the past years. So called deepfakes can depict people who never participated in a video recording by transposing their faces onto others in it. This raises the concern of authenticity of media, which demand for higher performing detection methods in forensics. Introduction of AI detectors have been of interest, but is held back today by their lack of interpretability. The objective of this thesis was therefore to examine what the explainable AI method local interpretable model-agnostic explanations (LIME) could contribute to forensic investigations of deepfake video.  An evaluation was conducted where 3 multimedia forensics evaluated the contribution of visual explanations of classifications when investigating deepfake video frames. The estimated contribution was not significant yet answers showed that LIME may be used to indicate areas to start examine. LIME was however not considered to provide sufficient proof to why a frame was classified as `fake', and would if introduced be used as one of several methods in the process. Issues were apparent regarding the interpretability of the explanations, as well as LIME's ability to indicate features of manipulation with superpixels.