EITM Summer Training Institute(Political Science Program: EITM Competition IIIa)

对政治的科学研究需要对理论模型进行实证检验，但理论往往是在没有充分检验的情况下产生的，实证工作过于频繁地使用粗略和过于简单化的理论。理论和经验方法之间出现了差距，这些差距损害了科学进步。要改善这一点，学者们需要培养能够更好地将理论与实证工作联系起来的新一代学者。首席调查员将促进这种智力发展，为年轻学者提供机会，通过与在推进理论和实证工作方面处于领先地位的老年学者一起观看和做事情来学习，重点放在已经存在整合理论和方法的可观研究的实质性领域。一个成功的理论模型的实证影响(EITM)暑期项目，考虑到目前这种社会科学方法的相对初级阶段，必须与传播实质一样多地发展。因此，关于培训的关键是，每个学院都包括重要的教学和研究部分，为学生提供与EITM导师的高度个性化互动，这些导师的范围和深度远远超过任何单个机构。首席调查员的活动超越了被动接受授课材料的范围，还包括利用新获得的关于理论和方法整合的知识来促进参与者自己的研究的重要机会。研究人员在2002-2005年的四年中，每年在四个地点(分别是哈佛大学、密歇根大学、杜克大学和加州大学伯克利分校)提供为期四周的暑期项目，以促进综合培训。每所学院最多可容纳25名高级研究生和初级教员。该项目是有选择性的，录取在很大程度上是基于提出的研究的质量和潜力。一个由多达15名研究人员组成的团队在下面详细说明的框架内管理每个研究所。研究人员对用于评估和测试含义的理论模型和经验方法进行了广泛的研究。教学将建立的形式模型包括博弈论、微分方程动态模型、简单决策理论和更复杂的行为决策模型。经验工具包不仅应涵盖统计推断，而且应侧重于以分析为基础的案例研究和实验方法。计算模型在程序中既有理论上的相关性，也有经验上的相关性。本文概述了六个领域，这些领域在将面向微观的、制度上丰富的理论与严格的结构估计相结合方面已经足够成熟，为综合教学提供了富有成效的可能性。这六个领域以综合研究和教学大纲为例，涵盖空间模型、制度分析、宏观政治经济学、谈判和联盟、国际安全和国际政治经济学。研究人员每年包括三个这样的区域，并保证每年至少有一个重复的积累和开发以及至少一个新的模块。使用多个主机可以利用任何单一地点都无法获得的多样化、地区性的深厚人才库。它在可访问性方面提供了区域平衡，并促进了与几个地点的其他部门的联系。这四个国家在计算模型、实验以及宏观和国际政治经济方面具有互补优势。

The scientific study of politics requires empirical testing of theoretical models, but theories are often produced without adequate testing and empirical work too frequently uses sketchy and oversimplified theory. Gaps have appeared between theory and empirical method, and these gaps impair scientific progress. To ameliorate this scholars need to train a new generation of scholars who can better link theory and empirical work. The Principal Investigators will to foster this intellectual development by offering younger scholars an opportunity to learn by seeing and doing in conjunction with older scholars who have been leaders in advancing theory and empirical work, focusing on substantive areas where appreciable research integrating theory and methods already exists. A successful Empirical Implications of Theoretical Models (EITM) summer program, given the current relative infancy of this approach to social science, must as much develop as disseminate substance. Thus the key regarding training is that each institute include significant teaching and research components, providing students a high degree of individualized interaction with a far wider and deeper array of EITM mentors than are available at any individual institution. The Principal Investigators have activities transcending passive receipt of lecture material, involving also significant opportunities to advance participants' own research by employing their newly gained knowledge about the integration of theory and method. The investigators foster integrated training by providing, in each of the four years from 2002-2005, in four locations (respectively Harvard, Michigan, Duke, and UC-Berkeley), a four-week summer program. Each of these institutes accommodates up to 25 advanced graduate students and junior faculty. The program is selective and admission is based significantly on the quality and potential of research presented. A team of up to 15 research faculty conducts each institute in a framework detailed below. The investigators take a broad view of the theoretical models and empirical methods used to evaluate and test implications. Formal models on which instruction will build include game theory, differential equation dynamic models, and simple decision theory and more complicated behavioral decision-making models. The empirical toolkit should cover not only statistical inference but also focused analytically-based case studies, and experimental methods. Computational models have both theoretical and empirical relevance in the program. Six areas are outlined that have sufficiently matured in combining micro-oriented, institutionally rich theory with rigorous structural estimation to offer fruitful possibilities for integrated instruction. The six areas, with examples of integrated research and teaching outlines, cover spatial models, institutional analysis, macropolitical economy, bargaining and coalitions, international security, and international political economy. The researchers include three such areas each year, and guarantee that there will be at least one repeat for cumulation and development and at least one new module each year. Using multiple hosts capitalizes on diverse, regionally available deep pools of talent unavailable in any single location. It offers regional balance in accessibility and fosters outreach to other departments in several locales. The four have complementary strengths in computational models, experimentation, and macro- and international political economy.