241 |
Algorithmic Fidelity and the use of Large Language Models in Social Science ResearchRytting, Christopher Michael 23 May 2023 (has links) (PDF)
In this dissertation, we argue that large language models (LLMs) exhibit a considerable amount of \textit{algorithmic fidelity}, a property where they have modeled ideas, behaviors, and attitudes of the population who generated their training data. This has important implications for social science, as this fidelity theoretically allows for the use of LLMs as effective proxies for human beings in experiments and research. We demonstrate this empirically in various social science domains (political partisanship, demographic surveying, voting behavior, hot-button policy issues, news media, populism, congressional summaries), in various applications (replicating social science survey findings, assisting in coding of text datasets, inferring demographics, automating interventions to improve conversations about divisive topics), and at various levels of granularity (from findings about the entire U.S. population down to specific demographics and down to the individual level). It is intrinsically interesting that LLMs could learn such behaviors on the unsupervised objective whereby they are trained. It is also strategically useful to establish where and to what extent they have, so that we can study them in cheaper and formerly impossible ways. This work serves as a preliminary study on these phenomena and an early, demonstrative methodology for drawing out the algorithmic fidelity from large language models.
|
242 |
Pipelines for Computational Social Science Experiments and Model BuildingCedeno, Vanessa Ines 12 July 2019 (has links)
There has been significant growth in online social science experiments in order to understand behavior at-scale, with finer-grained data collection. Considerable work is required to perform data analytics for custom experiments. In this dissertation, we design and build composable and extensible automated software pipelines for evaluating social phenomena through iterative experiments and modeling. To reason about experiments and models, we design a formal data model. This combined approach of experiments and models has been done in some studies without automation, or purely conceptually.
We are motivated by a particular social behavior, namely collective identity (CI). Group or CI is an individual's cognitive, moral, and emotional connection with a broader community, category, practice, or institution. Extensive experimental research shows that CI influences human decision-making. Because of this, there is interest in modeling situations that promote the creation of CI in order to learn more from the process and to predict human behavior in real life situations.
One of our goals in this dissertation is to understand whether a cooperative anagram game can produce CI within a group. With all of the experimental work on anagram games, it is surprising that very little work has been done in modeling these games. Also, abduction is an inference approach that uses data and observations to identify plausibly (and preferably, best) explanations for phenomena. Abduction has broad application in robotics, genetics, automated systems, and image understanding, but have largely been devoid of human behavior. We use these pipelines to understand intra-group cooperation and its effect on fostering CI. We devise and execute an iterative abductive analysis process that is driven by the social sciences.
In a group anagrams web-based networked game setting, we formalize an abductive loop, implement it computationally, and exercise it; we build and evaluate three agent-based models (ABMs) through a set of composable and extensible pipelines; we also analyze experimental data and develop mechanistic and data-driven models of human reasoning to predict detailed game player action. The agreement between model predictions and experimental data indicate that our models can explain behavior and provide novel experimental insights into CI. / Doctor of Philosophy / To understand individual and collective behavior, there has been significant interest in using online systems to carry out social science experiments. Considerable work is required for analyzing the data and to uncover interesting insights. In this dissertation, we design and build automated software pipelines for evaluating social phenomena through iterative experiments and modeling. To reason about experiments and models, we design a formal data model. This combined approach of experiments and models has been done in some studies without automation, or purely conceptually.
We are motivated by a particular social behavior, namely collective identity (CI). Group or CI is an individual’s cognitive, moral, and emotional connection with a broader community, category, practice, or institution. Extensive experimental research shows that CI influences human decision-making, so there is interest in modeling situations that promote the creation of CI to learn more from the process and to predict human behavior in real life situations.
One of our goals in this dissertation is to understand whether a cooperative anagram game can produce CI within a group. With all of the experimental work on anagrams games, it is surprising that very little work has been done in modeling these games. In addition, to identify best explanations for phenomena we use abduction. Abduction is an inference approach that uses data and observations. Abduction has broad application in robotics, genetics, automated systems, and image understanding, but have largely been devoid of human behavior.
In a group anagrams web-based networked game setting we do the following. We use these pipelines to understand intra-group cooperation and its effect on fostering CI. We devise and execute an iterative abductive analysis process that is driven by the social sciences. We build and evaluate three agent-based models (ABMs). We analyze experimental data and develop models of human reasoning to predict detailed game player action. We claim our models can explain behavior and provide novel experimental insights into CI, because there is agreement between the model predictions and the experimental data.
|
243 |
Expanding Audience Engagement with State Wildlife Agencies: Strategies for Supporting Diverse ParticipationJennings, Kelsey Kathleen 03 May 2024 (has links)
State fish and wildlife agencies are increasingly expected to meet accelerating wildlife conservation needs while addressing systemic problems that limit who has historically been able to access wildlife and conservation. To effectively address both of these needs, agencies need to better understand their constituencies, including both long-standing supporters and those who have been historically excluded. This thesis set out to achieve this through three chapters. (i) Using a nationwide panel survey, we sought to understand how patterns of wildlife viewing differ between Black, Indigenous, and people of color (BIPOC) and White viewers, as well as what most strongly supports this participation. We found that participation rates differ across ethnoracial groups, with most BIPOC viewers participating less than White viewers, especially around their homes. Additionally, BIPOC viewers indicated that social support was a key factor in encouraging participation. Finally, despite wildlife viewing being of similar or greater importance to BIPOC participants, they generally self-identified less as viewers, suggesting historic exclusion from wildlife viewing locales may impact contemporary participation. (ii) Through a questionnaire of Minnesotans engaged with the MN Department of Natural Resources and their Nongame Wildlife Program, we examined what most encourages donations and volunteer participation. The findings of this chapter suggest that donors had specific motivations, namely supporting biodiversity and conservation, whereas volunteer motivations were more varied, such as place-based preferences and access to conservation professionals. (iii) Using a series of focus groups with outdoor recreation and community science practitioners and participants, we aimed to understand what encouraged BIPOC Minnesotans to participate in nature-based community science programs. We found that participants rely on numerous forms of cultural capital to persist in community science, especially in inhospitable environments. Often, this included utilizing social relationships and continuing to participate despite negative interactions. Overall, this research contributes to a growing body of literature aiming to understand diverse engagement in wildlife viewing and nature-based community science, as well as general engagement with state agencies. By laying a foundation for sustainable engagement and support of diverse communities, particularly BIPOC groups, this thesis provides insights and recommendations for fostering equitable conservation practices in an era of environmental change and social evolution. / Master of Science / North America and the world are witnessing unprecedented environmental changes, including biodiversity loss, climate change, and urbanization, impacting both human and wildlife communities. Concurrently, shifts in public engagement with state wildlife agencies and outdoor recreation are evident, with historic constituencies of hunters and anglers declining, posing challenges to conservation efforts. To address these challenges, state agencies strive to understand and diversify their supporters. Against this backdrop, this thesis investigates wildlife viewing patterns among ethnoracial groups, constituent engagement with state wildlife agencies, and factors motivating participation in community science initiatives among Black, Indigenous, and people of color (BIPOC) communities. Using a nationwide survey of wildlife viewers, we found significant differences in participation, both around and away from home. We also found that social support from friends, family, mentors, and peers was associated with increased participation. Additionally, despite wildlife viewing being of similar or greater importance to BIPOC participants, they generally self-identified less as viewers, suggesting historic exclusion from wildlife viewing locales may impact contemporary participation. Through a questionnaire of Minnesotans engaged with the MN Department of Natural Resources and their Nongame Wildlife Program, we examined what most encourages donations and volunteer participation. The findings of this chapter suggested that donors had specific motivations, namely supporting biodiversity and conservation, whereas volunteer motivations were more varied, such as place-based preferences and access to conservation professionals. Finally, using a series of focus groups with outdoor recreation and community science practitioners and participants, we aimed to understand what encouraged BIPOC Minnesotans to participate in nature-based community science programs. Our results suggested that social support, community-specific and community-led programs, and programs co-produced with community groups may facilitate participation in community science. Overall, this thesis provides vital recommendations for state agencies aiming to increase BIPOC participation in programs and better engage their constituencies. Through guidance on developing sustainable engagement strategies and how to best support diverse communities, particularly BIPOC groups, this thesis provides insights and recommendations for fostering equitable conservation in an era of environmental change and social evolution.
|
244 |
'It Should've Never Been Broke Out': Understanding Participation in the Conservation Reserve Program in Southwest Kansas and Southeast ColoradoSteinmetz, Alexandra Corcoran Meyers 06 July 2018 (has links)
The Conservation Reserve Program (CRP) plays a vital role in restoring grasslands by removing highly erodible land from production; however, landscape-scale conservation success depends on participation. Fluctuating trends in participation suggest a need to better understand landowners' motivations for enrolling. Since participation hinges on agricultural producers' perceptions of programs, there is utility in understanding programs through their lens to ensure program design accounts for their needs. To understand what drives enrollment, I conducted immersive ethnographic fieldwork in farming and ranching communities of southwest Kansas and southeast Colorado. Through interviews and participant observation, I examined producers' reasons for participating, program perceptions, and the degree to which CRP fits with their lived experiences. I also explored challenges faced by field staff of the Farm Service Agency (FSA) and Natural Resources Conservation Service (NRCS) in working within the program structure. I used open coding to identify common themes and quotes to capture producers' and field staff's points of view. I identified several frames through which producers think about CRP and themes related to how CRP fit well or poorly with producers' and field staff's lives. Frames characterized producer perceptions of CRP as a financial savior, a way to maintain financial solvency, and to gain leverage for their operation. Additionally, CRP was framed as a retirement fund and a conservation program that provides a solution for erodible land. Lived experiences related to wind erosion and the Dust Bowl, perceived community impacts of CRP, and the cultural and economic history of the region, also influenced how producers make sense of and 'frame' the program. Guaranteed payments to maintain cover incentivize participation, especially for land which some producers believe should have never been farmed, or 'broke out', in the first place. Even so, the economic and cultural aspiration to farm may prompt program avoidance or re-cultivation of prior CRP land. In identifying program fit, many felt the program serves a noble purpose but is complicated by rules which lack 'common sense'. While producers valued the program's role in soil stabilization and increased wildlife habitat, CRP requirements during the grass establishment phase and mid-contract management do not always align with producer and field staff visions. Mixed opinions existed around suitable grass species and management practices such as disking, interseeding, and grazing. A dominant theme emerged from producers, echoed by field staff, in the benefits of grazing and need for CRP to increase flexibility to maximize grazing compatibility. Broader program concerns included a shifting program focus, inconsistent enforcement of rules, and one-size-fits-all management. Personal relationships between FSA, NRCS, and producers were generally regarded as positive, and staff members value their role in working with producers to harmonize program requirements with producer needs, within the bounds of the program. Juggling various programs with limited time and other procedural issues leave many field staff feeling overwhelmed and understaffed. Field staff expressed a desire for greater one-on-one time with producers to better communicate program requirements or amend management plans. Both producers and field staff felt CRP could be enhanced to achieve a greater conservation benefit, alleviate staff burdens, and improve overall satisfaction if program rules had both greater flexibility and regional tailoring to correlate with the variable climate and local conditions. In exploring CRP 'frames' and 'fit', this case study provides a window into the interplay of producers' lived experiences in the shadow of the Dust Bowl, and a ubiquitous conservation program's impact on the way land is used. / Master of Science
|
245 |
Är elever förberedda för ett aktivt medborgarskap? : En studie i politiskt engagemang hos sistaårselever på studie- och yrkesförberedande programEdgren, Amanda January 2024 (has links)
The curriculum emphasizes the importance of social science teachers to affect students to be active, democratic and participating citizens. At the same time the interest in politics among students has been on a steady decline since the beginning of the 21th century. Apart from the diminishing interest in politics, studies have shown that students are given different civic skills and competencies depending on their choice of educational path. This study takes an interest to understand what affects students' engagement in democracy and politics, and how it differs between last-year students from study preparatory and vocational education. This study also aims to understand how the programs differ in the experience of social science teaching and how it has affected their civic engagement. This study utilized qualitative group interviews with students from both educational paths, the results of this study illustrate that differences in political engagement, depending on educational path, do exist. The differences in education of social science has given different prerequisites to be politically engaged for study preparatory- and vocational students. / <p>Godkänd 2024-04-05</p>
|
246 |
"Kultur är allt som berör och formar människor" : Lärares förståelse och användning av kultur i de samhällsorienterade ämnena / ”Culture is Anything that Affect and Shapes Human Beings” : Teachers Understanding and use of Culture in Social Science EducationNilsson, Elina, Frej, Jessica January 2024 (has links)
Abstract Culture is a significant part of the Swedish curriculum for compulsory school, despite this there is no definition of the concept. Therefore, it was found interesting to explore what understanding teachers who teach social science in primary education possess, and how they integrate culture in their teaching methods. As culture is a wide concept, the current study focused on two definitions of culture, the aesthetics and the anthropological. The study enlightened different views of culture and how students consume and construct culture. We chose to do an interview study where the reflexive interview technique was used. Open questions that were relevant were utilized which made it possible to ask follow up questions. Furthemore, the study contained interviews with six different teachers. To analyze the material from the interviews content analysis was used. The result of the analysis showed that the aesthetic concept of culture is more common in both the teachers' understandings and teachings. Some of the teachers emphasized the anthropological parts of culture. The participants often connected the concept with other subjects than social science. The study showed that the teachers integrated culture in social science through different out of schedule activities. Moreover, this work is often connected to the school's cultural mission and specific days. It was further recognized that culture is not a part of the day to day education in social science and has a development potential. Furthermore, there seems to be a lack of concrete teaching methods when working with culture. If there had been concrete teaching methods, it would favor the teachers in their everyday work. The study contributed with an understanding of how teachers integrate culture in their day to day education, and enlightened their understanding of the concept of culture. The results from the study can support future teachers in their profession. By widening their knowledge about the concept and the benefits of including it in their lesson plan, they can elaborate their teaching.
|
247 |
A Machine for ImaginationRafati, Tofan 12 January 2012 (has links)
It began with the question, "What if the Modern Man was successful in his dominion over nature?"
By means of Architecture this thesis became a speculation and commentary on the human condition. But, more than that, this is a story that tells the evolution and outcome of a series of questions and inquiries into the relationship between Architecture, art and the mythopoetic-narrative realm. / Master of Architecture
|
248 |
A study of the role of local supervisors of social studies education in VirginiaElliott, Thomas A. January 1982 (has links)
The purpose of this study was to assess the extent to which relationships exist between the perceptions of local supervisors of social studies, social studies teachers, and building principals relative to the role of the local supervisor of social studies education. Data were gathered to determine the perceptions of supervisors, teachers, and principals regarding actual and ideal performance on forty-two selected supervisory activities.
The descriptive-survey method was utilized in the study. Population sample included twenty local social studies supervisors, one hundred and ninety-seven building principals, and four hundred and sixty social studies teachers. The four hundred and sixty social studies teachers (twenty-five percent) and one hundred and ninety-seven building principals represented employees in each school division in Virginia during the 1980-81 school year where local supervisors of social studies education were employed full-time.
Mail questionnaires were developed, piloted, and forwarded to the selected sample population. Usable responses were received from three hundred and twenty-five teachers (seventy-one percent), one hundred and fifty-one building principals (seventy-seven percent), and twenty local supervisors of social studies (one hundred percent).
The alpha level selected for determination of statistical significance and rejection of the seven null hypotheses was p < .05. Descriptive statistics, correlated t tests, uncorrelated t tests, and the Spearman rho (ρ) were used in analyzing the data. Data were reported by means, frequencies, ranks, standard deviations, and percentages. To analyze data within groups the correlated t test was used. The uncorrelated t test was used to analyze the data between groups. To determine the relationship between the rank differences reported by the sample groups, Spearman rho (ρ) correlations were computed using the formula:
ρ = l - (6Σd²)/[N(N² - 1l)]
In a comparison of perceptions of local supervisors of social studies education versus perceptions of teachers regarding actual performance on forty-two selected supervisory activities, it was determined that there was a high agreement between teachers and supervisors as to the order of priority placed on activities by supervisors. The Spearman rho computation based on rank differences of actual time spent as perceived by supervisors and the teacher's perceptions of actual time spent yielded a high positive correlation. Also, principals and supervisors were in close agreement as to the priority of supervisors' actual use of time. A high positive correlation was found to exist based upon rankings of the two groups relative to actual use of supervisors' time.
Significant differences were noted between the perceptions of local supervisors of social studies education and those of social studies teachers and building principals relative to ideal supervisory performance. Teachers and supervisors were in moderate agreement relative to the setting of priorities for time utilization by supervisors across the forty-two activities. Principals and supervisors were in high agreement relative to the priorities based on ideal rankings for supervisors.
Major findings of the study were as follows:
1. Supervisors, teachers, and principals agreed that supervisors should spend more time on nearly all of the identified activities than they currently do.
2. Each group surveyed agreed that the ideal performance of supervisors should be closely related to that group's perceptions of what supervisors currently do.
3. There was a high positive correlation between the supervisors, teachers, and principals as to the allocation of priorities by supervisors.
4. There was a moderate to high positive correlation across groups as to how supervisors should allocate their time. / Ed. D.
|
249 |
Developing and Evaluating a Community Social Science Program for the Seventh GradeThomas, Lewis C. 08 1900 (has links)
The purpose of this study is to develop and to evaluate a community social science program for the seventh grade. The control group was taught in the traditional method, while the experimental group was taught with the community as the basis of all activities.
|
250 |
Sderot : an analysis of the marginalization of an Israeli border town populationDansky, Ariel 01 January 2010 (has links)
This research focuses on the Israeli town of Sderot and the rocket attacks it experienced since 2001. Sderot is a unique case study because it represents a group of individuals in a democratic country that lived with terrorism for almost a decade before the state took major defensive action. The situation in Sderot is one which has lacked attention in the media and in Political Science research. By analyzing the level of attention by multiple actors to the crisis in Sderot, the reasons for the perpetual insecurity of its population are discussed.
The crisis is analyzed on four main levels: the experience of individuals in Sderot, the response of the Israeli government, Israeli-Palestinian negotiations, and the role of the United Nations. The preliminary chapter examines the impact of living with rocket fire while exploring methods by which Sderot residents have engaged in activism to improve their quality of life. The following chapter discusses Israeli national defense policy and examines where Sderot has ranked on the State's list of priorities. The third chapter consists of two main sections: an analysis of Israeli-Palestinian peace negotiations, an explanation of Barnas' rise to power in Gaza. The latter section consists of an exploration of the politics surrounding the United Nation's level of attention to the crisis in Sderot.
As one transitions from the individual level of analysis to the state level, the voices of Sderot residents become much quieter, and the realities of a state that is constantly attacked from beyond its borders can be understood. By analyzing the past failings of peace negotiations between Israeli and Palestinian leaders, lessons for future attempts at negotiations are discussed, and the ever-present link between peace and security is emphasized. Overall, the realities of daily life in a state which pursues a policy of security over diplomacy are illuminated.
|
Page generated in 0.1028 seconds