Pub Date : 2019-04-08DOI: 10.1080/2330443x.2020.1806763
M. Chikina, A. Frieze, J. Mattingly, W. Pegden
Abstract We give qualitative and quantitative improvements to theorems which enable significance testing in Markov chains, with a particular eye toward the goal of enabling strong, interpretable, and statistically rigorous claims of political gerrymandering. Our results can be used to demonstrate at a desired significance level that a given Markov chain state (e.g., a districting) is extremely unusual (rather than just atypical) with respect to the fragility of its characteristics in the chain. We also provide theorems specialized to leverage quantitative improvements when there is a product structure in the underlying probability space, as can occur due to geographical constraints on districtings.
{"title":"Separating Effect From Significance in Markov Chain Tests","authors":"M. Chikina, A. Frieze, J. Mattingly, W. Pegden","doi":"10.1080/2330443x.2020.1806763","DOIUrl":"https://doi.org/10.1080/2330443x.2020.1806763","url":null,"abstract":"Abstract We give qualitative and quantitative improvements to theorems which enable significance testing in Markov chains, with a particular eye toward the goal of enabling strong, interpretable, and statistically rigorous claims of political gerrymandering. Our results can be used to demonstrate at a desired significance level that a given Markov chain state (e.g., a districting) is extremely unusual (rather than just atypical) with respect to the fragility of its characteristics in the chain. We also provide theorems specialized to leverage quantitative improvements when there is a product structure in the underlying probability space, as can occur due to geographical constraints on districtings.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"7 1","pages":"101 - 114"},"PeriodicalIF":1.6,"publicationDate":"2019-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443x.2020.1806763","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46921817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-03-26DOI: 10.1080/2330443X.2022.2086191
David Puelz, R. Puelz
Abstract We explore the relationship between financial literacy and self-reported, reflective economic outcomes from respondents using survey data from the United States. Our dataset includes a large number of covariates from the National Financial Capability Study (NFCS), widely used by literacy researchers, and we use a new econometric technique developed by Hahn et al., designed specifically for causal inference from observational data, to test whether changes in financial literacy infer meaningful changes in self-perceived economic outcomes. We find a negative treatment parameter on financial literacy consistent with the recent work of Netemeyer et al. and contrary to the presumption in many empirical studies that associate standard financial outcome measures with financial literacy. We conclude with a discussion of heterogeneity of the financial literacy treatment effect on household income, gender, and education level sub-populations. Our findings on the relationship between financial literacy and reflective economic outcomes also raise questions about its importance to an individual’s financial well-being.
{"title":"Financial Literacy and Perceived Economic Outcomes","authors":"David Puelz, R. Puelz","doi":"10.1080/2330443X.2022.2086191","DOIUrl":"https://doi.org/10.1080/2330443X.2022.2086191","url":null,"abstract":"Abstract We explore the relationship between financial literacy and self-reported, reflective economic outcomes from respondents using survey data from the United States. Our dataset includes a large number of covariates from the National Financial Capability Study (NFCS), widely used by literacy researchers, and we use a new econometric technique developed by Hahn et al., designed specifically for causal inference from observational data, to test whether changes in financial literacy infer meaningful changes in self-perceived economic outcomes. We find a negative treatment parameter on financial literacy consistent with the recent work of Netemeyer et al. and contrary to the presumption in many empirical studies that associate standard financial outcome measures with financial literacy. We conclude with a discussion of heterogeneity of the financial literacy treatment effect on household income, gender, and education level sub-populations. Our findings on the relationship between financial literacy and reflective economic outcomes also raise questions about its importance to an individual’s financial well-being.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"9 1","pages":"122 - 135"},"PeriodicalIF":1.6,"publicationDate":"2019-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49584551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443X.2019.1574687
Wendy K. Tam Cho, Simon Rubinstein-Salzedo
ABSTRACT Recently, Chikina, Frieze, and Pegden proposed a way to assess significance in a Markov chain without requiring that Markov chain to mix. They presented their theorem as a rigorous test for partisan gerrymandering. We clarify that their ε-outlier test is distinct from a traditional global outlier test and does not indicate, as they imply, that a particular electoral map is associated with an extreme level of “partisan unfairness.” In fact, a map could simultaneously be an ε-outlier and have a typical partisan fairness value. That is, their test identifies local outliers but has no power for assessing whether that local outlier is a global outlier. How their specific definition of local outlier is related to a legal gerrymandering claim is unclear given Supreme Court precedent.
{"title":"Understanding Significance Tests From a Non-Mixing Markov Chain for Partisan Gerrymandering Claims","authors":"Wendy K. Tam Cho, Simon Rubinstein-Salzedo","doi":"10.1080/2330443X.2019.1574687","DOIUrl":"https://doi.org/10.1080/2330443X.2019.1574687","url":null,"abstract":"ABSTRACT Recently, Chikina, Frieze, and Pegden proposed a way to assess significance in a Markov chain without requiring that Markov chain to mix. They presented their theorem as a rigorous test for partisan gerrymandering. We clarify that their ε-outlier test is distinct from a traditional global outlier test and does not indicate, as they imply, that a particular electoral map is associated with an extreme level of “partisan unfairness.” In fact, a map could simultaneously be an ε-outlier and have a typical partisan fairness value. That is, their test identifies local outliers but has no power for assessing whether that local outlier is a global outlier. How their specific definition of local outlier is related to a legal gerrymandering claim is unclear given Supreme Court precedent.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"44 - 49"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443X.2019.1574687","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45832676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443x.2019.1688742
J. Ratner
Those of us who value analytic thinking about public policy and, in particular, about war, can learn a great deal from reading “Cost Benefit Analysis of Discretionary Wars” by Diane Hu and her coauthors.1 The article also raises many questions, and considering them spurs learning too. Their article contributes to the literature by formulating and implementing an approach to the cost-benefit analysis (CBA) of war that is tractable and amenable to empirical use. Notably, the authors add value by operationalizing several dimensions of war’s benefits, by introducing certain simplified methods of estimating the costs of war, and by applying their framework of measuring costs and benefits to five case-studies of discretionary war. As the authors note, they build on the work of Nordhaus (2002), Stiglitz and Bilmes (2008), and others regarding the costs to the United States of the Afghanistan and Iraq Wars, as well as on Hausken’s important theoretical framework for conducting a CBA of war (Hausken 2016). By abstracting from many complexities articulated by Hausken, the authors create an empirically oriented framework that can be populated with data from their case-studies of U.S. discretionary war.2 By examining a war’s benefits and assigning monetary values to them, the authors are able to juxtapose these monetized benefits to their estimates of these wars’ costs, thereby answering the question: Did the costs of these wars outweigh their benefits? The authors’ extensive attention to war’s benefits is distinctive, especially in estimating these benefits for five wars. (Other studies of a U.S. war’s monetized benefits focus on one war.3) Furthermore, they obtain a striking result: costs exceed benefits for all five wars. None, not even the First Gulf War or Korea, escapes the article’s grim verdict: negative net benefits should have ruled out these wars.
我们这些重视对公共政策，特别是战争进行分析思考的人，可以从戴安·胡(Diane Hu)及其合著者的《自由裁量战争的成本效益分析》(Cost - Benefit Analysis of Discretionary Wars)中学到很多东西这篇文章也提出了许多问题，思考这些问题也会刺激学习。他们的文章通过制定和实施战争成本效益分析(CBA)的方法对文献做出了贡献，这种方法易于处理，可用于实证应用。值得注意的是，作者通过对战争利益的几个维度进行操作，通过引入某些简化的估算战争成本的方法，以及通过将其衡量成本和收益的框架应用于自由裁量战争的五个案例研究，从而增加了价值。正如作者所指出的，他们建立在诺德豪斯(2002)，斯蒂格利茨和比尔梅斯(2008)的工作基础上，以及其他关于阿富汗和伊拉克战争对美国成本的研究，以及豪斯肯进行战争CBA的重要理论框架(豪斯肯2016)。通过从Hausken所阐述的许多复杂性中抽象出来，作者创建了一个以经验为导向的框架，可以用他们对美国自由裁量战争的案例研究中的数据进行填充通过研究一场战争的利益并赋予其货币价值，作者能够将这些货币化的利益与他们对这些战争成本的估计并置，从而回答了这个问题:这些战争的成本是否超过了它们的收益?作者对战争利益的广泛关注是与众不同的，特别是在估计五场战争的这些利益时。(其他关于美国战争货币化收益的研究集中在一场战争上。)此外，他们得出了一个惊人的结果:所有五场战争的成本都超过了收益。没有一场战争，甚至包括第一次海湾战争和朝鲜战争，逃不过这篇文章的残酷结论:负净收益本应排除这些战争。
{"title":"Discretionary Wars, Cost-Benefit Analysis, and the Rashomon Effect: Searching for an Analytical Engine for Avoiding War","authors":"J. Ratner","doi":"10.1080/2330443x.2019.1688742","DOIUrl":"https://doi.org/10.1080/2330443x.2019.1688742","url":null,"abstract":"Those of us who value analytic thinking about public policy and, in particular, about war, can learn a great deal from reading “Cost Benefit Analysis of Discretionary Wars” by Diane Hu and her coauthors.1 The article also raises many questions, and considering them spurs learning too. Their article contributes to the literature by formulating and implementing an approach to the cost-benefit analysis (CBA) of war that is tractable and amenable to empirical use. Notably, the authors add value by operationalizing several dimensions of war’s benefits, by introducing certain simplified methods of estimating the costs of war, and by applying their framework of measuring costs and benefits to five case-studies of discretionary war. As the authors note, they build on the work of Nordhaus (2002), Stiglitz and Bilmes (2008), and others regarding the costs to the United States of the Afghanistan and Iraq Wars, as well as on Hausken’s important theoretical framework for conducting a CBA of war (Hausken 2016). By abstracting from many complexities articulated by Hausken, the authors create an empirically oriented framework that can be populated with data from their case-studies of U.S. discretionary war.2 By examining a war’s benefits and assigning monetary values to them, the authors are able to juxtapose these monetized benefits to their estimates of these wars’ costs, thereby answering the question: Did the costs of these wars outweigh their benefits? The authors’ extensive attention to war’s benefits is distinctive, especially in estimating these benefits for five wars. (Other studies of a U.S. war’s monetized benefits focus on one war.3) Furthermore, they obtain a striking result: costs exceed benefits for all five wars. None, not even the First Gulf War or Korea, escapes the article’s grim verdict: negative net benefits should have ruled out these wars.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"107 - 121"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443x.2019.1688742","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44637485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443X.2018.1555068
J. Nguyen, C. Tiu, J. Stewart, David H. Miller
Abstract Mixed-effects models were used to evaluate the global zoning concept using residue data from a comprehensive database of supervised field trials performed in various countries and regions on a variety of pesticide–crop combinations. No statistically significant systematic differences in pesticide residues were found between zones among the pesticide uses examined. In addition, we conducted a simulation to assess the impact of using regional versus global datasets for calculating maximum residue limits (MRLs). The conclusion of this assessment supports the concept of exchangeability of pesticide residue values across geographic regions and opens the possibility of improving harmonization of pesticide regulatory standards by establishing more globally aligned MRLs. Supplemental material for this article is available online.
{"title":"Global Zoning and Exchangeability of Field Trial Residues Between Zones: Are There Systematic Differences in Pesticide Residues Across Geographies?","authors":"J. Nguyen, C. Tiu, J. Stewart, David H. Miller","doi":"10.1080/2330443X.2018.1555068","DOIUrl":"https://doi.org/10.1080/2330443X.2018.1555068","url":null,"abstract":"Abstract Mixed-effects models were used to evaluate the global zoning concept using residue data from a comprehensive database of supervised field trials performed in various countries and regions on a variety of pesticide–crop combinations. No statistically significant systematic differences in pesticide residues were found between zones among the pesticide uses examined. In addition, we conducted a simulation to assess the impact of using regional versus global datasets for calculating maximum residue limits (MRLs). The conclusion of this assessment supports the concept of exchangeability of pesticide residue values across geographic regions and opens the possibility of improving harmonization of pesticide regulatory standards by establishing more globally aligned MRLs. Supplemental material for this article is available online.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"14 - 23"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443X.2018.1555068","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47736208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443x.2019.1688740
D. Hu, A. Cooper, Neel Desai, Sophie Guo, Steven Shi, David L. Banks
Abstract Policy-makers should perform a cost-benefit analysis before initiating a war. This article describes a methodology for such assessment, and applies it post hoc to five military actions undertaken by the United States between 1950 and 2000 (the Korean War, the Vietnam War, the invasion of Grenada, the invasion of Panama, and the First Gulf War). The analysis identifies three broad categories of value: human capital, economic outcomes, and national influence. Different stakeholders (politicians, generals, industry, etc.) may assign different weights to these three categories, so this analysis tabulates each separately, and then, as may sometimes be necessary, monetizes them for unified comparison.
{"title":"Cost-Benefit Analysis of Discretionary Wars","authors":"D. Hu, A. Cooper, Neel Desai, Sophie Guo, Steven Shi, David L. Banks","doi":"10.1080/2330443x.2019.1688740","DOIUrl":"https://doi.org/10.1080/2330443x.2019.1688740","url":null,"abstract":"Abstract Policy-makers should perform a cost-benefit analysis before initiating a war. This article describes a methodology for such assessment, and applies it post hoc to five military actions undertaken by the United States between 1950 and 2000 (the Korean War, the Vietnam War, the invasion of Grenada, the invasion of Panama, and the First Gulf War). The analysis identifies three broad categories of value: human capital, economic outcomes, and national influence. Different stakeholders (politicians, generals, industry, etc.) may assign different weights to these three categories, so this analysis tabulates each separately, and then, as may sometimes be necessary, monetizes them for unified comparison.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"106 - 98"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443x.2019.1688740","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44198259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443X.2019.1574686
R. Amin, Alexander Bohnert, David L. Banks
ABSTRACT This study identifies pediatric cancer clusters in Florida for the years 2000–2015. Unlike previous publications on pediatric cancers in Florida, it draws upon an Environmental Protection Agency dataset on carcinogenic air pollution, the National Air Toxics Assessment, as well as more customary demographic variables (age, sex, race). The focus is upon the three most widely seen pediatric cancer types in the USA: brain tumors, leukemia, and lymphomas. The covariates are used in a Poisson regression to predict cancer incidence. The adjusted cluster analysis quantifies the role of each covariate. Using Florida Association of Pediatric Tumor Programs data for 2000–2015, we find statistically significant pediatric cancer clusters, but we cannot associate air pollution with the cancer incidence. Supplementary materials for this article are available online.
{"title":"Patterns of Pediatric Cancers in Florida: 2000–2015","authors":"R. Amin, Alexander Bohnert, David L. Banks","doi":"10.1080/2330443X.2019.1574686","DOIUrl":"https://doi.org/10.1080/2330443X.2019.1574686","url":null,"abstract":"ABSTRACT This study identifies pediatric cancer clusters in Florida for the years 2000–2015. Unlike previous publications on pediatric cancers in Florida, it draws upon an Environmental Protection Agency dataset on carcinogenic air pollution, the National Air Toxics Assessment, as well as more customary demographic variables (age, sex, race). The focus is upon the three most widely seen pediatric cancer types in the USA: brain tumors, leukemia, and lymphomas. The covariates are used in a Poisson regression to predict cancer incidence. The adjusted cluster analysis quantifies the role of each covariate. Using Florida Association of Pediatric Tumor Programs data for 2000–2015, we find statistically significant pediatric cancer clusters, but we cannot associate air pollution with the cancer incidence. Supplementary materials for this article are available online.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"24 - 35"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443X.2019.1574686","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42238617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443X.2019.1615396
M. Chikina, A. Frieze, W. Pegden
Abstract The article of Cho and Rubinstein-Salzedo seeks to cast doubt on our previous paper, which described a rigorous statistical test which can be applied to reversible Markov chains. In particular, Cho and Rubinstein-Salzedo seem to suggest that the test we describe might not be a reliable indicator of gerrymandering, when the test is applied to certain redistricting Markov chains. However, the examples constructed by Cho and Rubinstein-Salzedo in fact demonstrate a different point: that our test is not the same as another class of gerrymandering tests, which Cho and Rubinstein-Salzedo prefer. But we agree and emphasized this very distinction in our original paper. In this reply, we reply to the criticisms of Cho and Rubinstein-Salzedo, and discuss, more generally, the advantages of the various tests available in the context of detecting gerrymandering of political districtings.
{"title":"Understanding Our Markov Chain Significance Test: A Reply to Cho and Rubinstein-Salzedo","authors":"M. Chikina, A. Frieze, W. Pegden","doi":"10.1080/2330443X.2019.1615396","DOIUrl":"https://doi.org/10.1080/2330443X.2019.1615396","url":null,"abstract":"Abstract The article of Cho and Rubinstein-Salzedo seeks to cast doubt on our previous paper, which described a rigorous statistical test which can be applied to reversible Markov chains. In particular, Cho and Rubinstein-Salzedo seem to suggest that the test we describe might not be a reliable indicator of gerrymandering, when the test is applied to certain redistricting Markov chains. However, the examples constructed by Cho and Rubinstein-Salzedo in fact demonstrate a different point: that our test is not the same as another class of gerrymandering tests, which Cho and Rubinstein-Salzedo prefer. But we agree and emphasized this very distinction in our original paper. In this reply, we reply to the criticisms of Cho and Rubinstein-Salzedo, and discuss, more generally, the advantages of the various tests available in the context of detecting gerrymandering of political districtings.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"50 - 53"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443X.2019.1615396","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41320895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443x.2019.1660285
Yi-jie Tang, Nicole M. Dalzell
ABSTRACT Social media and other online sites are being increasingly scrutinized as platforms for cyberbullying and hate speech. Many machine learning algorithms, such as support vector machines, have been adopted to create classification tools to identify and potentially filter patterns of negative speech. While effective for prediction, these methodologies yield models that are difficult to interpret. In addition, many studies focus on classifying comments as either negative or neutral, rather than further separating negative comments into subcategories. To address both of these concerns, we introduce a two-stage model for classifying text. With this model, we illustrate the use of internal lexicons, collections of words generated from a pre-classified training dataset of comments that are specific to several subcategories of negative comments. In the first stage, a machine learning algorithm classifies each comment as negative or neutral, or more generally target or nontarget. The second stage of model building leverages the internal lexicons (called L2CLs) to create features specific to each subcategory. These features, along with others, are then used in a random forest model to classify the comments into the subcategories of interest. We demonstrate our approach using two sets of data. Supplementary materials for this article are available online.
{"title":"Classifying Hate Speech Using a Two-Layer Model","authors":"Yi-jie Tang, Nicole M. Dalzell","doi":"10.1080/2330443x.2019.1660285","DOIUrl":"https://doi.org/10.1080/2330443x.2019.1660285","url":null,"abstract":"ABSTRACT Social media and other online sites are being increasingly scrutinized as platforms for cyberbullying and hate speech. Many machine learning algorithms, such as support vector machines, have been adopted to create classification tools to identify and potentially filter patterns of negative speech. While effective for prediction, these methodologies yield models that are difficult to interpret. In addition, many studies focus on classifying comments as either negative or neutral, rather than further separating negative comments into subcategories. To address both of these concerns, we introduce a two-stage model for classifying text. With this model, we illustrate the use of internal lexicons, collections of words generated from a pre-classified training dataset of comments that are specific to several subcategories of negative comments. In the first stage, a machine learning algorithm classifies each comment as negative or neutral, or more generally target or nontarget. The second stage of model building leverages the internal lexicons (called L2CLs) to create features specific to each subcategory. These features, along with others, are then used in a random forest model to classify the comments into the subcategories of interest. We demonstrate our approach using two sets of data. Supplementary materials for this article are available online.","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"80 - 86"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443x.2019.1660285","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41530975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-01-01DOI: 10.1080/2330443x.2019.1688741
David L. Banks
Dr. Jonathan Ratner’s discussion is amazing and a valuable commentary (and sometimes a corrective) upon the work in our article. We are grateful for his thoughtful examination and testing of the assumptions and methodology we have used. His contribution goes far beyond a typical discussion and is an article in its own right, or at the very least a provocative essay. He makes many important points and builds out our reasoning and expands its scope in numerous ways. This response attempts to briefly address some of the key points and suggestions that he makes. Dr. Ratner is quite correct that we made the enormously simplifying assumption of a unitary decision-maker, the “president,” who need only consult his or her utility function, and whose analysis is rational and unselfish but completely proAmerican. Like everyone, we appreciate that the political realities are far more complex than that, but we believe that our deliberate simplification has the advantage of focusing attention on the simple question of whether the five wars (or military actions) under consideration led to good or bad economic outcomes for the United States as a whole. Clearly, one could address a more realistic decision-theoretic framework in which multiple stakeholders (Congress, generals, intelligence analysts, Halliburton, and many others) negotiate or coalesce or diverge in reaching a military decision, and that would surely lead to fascinating work in sociology and political science. But such modeling was not our intent. And we appreciate Dr. Ratner’s recognition that our primary goal was the cost-benefit analysis. Our emphasis on “the U.S.-centric utility function” bothered Dr. Ratner, and we readily acknowledge that it makes us morally uncomfortable too. We would prefer to live in a world in which the United States is not indifferent to the suffering of others and where altruism is part of the calculus of leadership. And we also think that considerations of decency are usually given some weight in the corridors of power. However, we also believe that a callous calculation of the bottom line is a necessary component of military and other policy decisions. Absent that starting point, there seems to be no principled basis for prioritizing cases and causes. Dr. Ratner would prefer to see “a sensitivity analysis, with an alternative, semi-altruistic utility function.” We think that would be interesting and useful, and effective altruism is always important. But (as Dr. Ratner points out later), our article is already heavily freighted with assumptions that have varying degrees of plausibility. Trying to monetize the lives of non-American
{"title":"Response to “Discretionary Wars, Cost-Benefit Analysis, and the Rashomon Effect”","authors":"David L. Banks","doi":"10.1080/2330443x.2019.1688741","DOIUrl":"https://doi.org/10.1080/2330443x.2019.1688741","url":null,"abstract":"Dr. Jonathan Ratner’s discussion is amazing and a valuable commentary (and sometimes a corrective) upon the work in our article. We are grateful for his thoughtful examination and testing of the assumptions and methodology we have used. His contribution goes far beyond a typical discussion and is an article in its own right, or at the very least a provocative essay. He makes many important points and builds out our reasoning and expands its scope in numerous ways. This response attempts to briefly address some of the key points and suggestions that he makes. Dr. Ratner is quite correct that we made the enormously simplifying assumption of a unitary decision-maker, the “president,” who need only consult his or her utility function, and whose analysis is rational and unselfish but completely proAmerican. Like everyone, we appreciate that the political realities are far more complex than that, but we believe that our deliberate simplification has the advantage of focusing attention on the simple question of whether the five wars (or military actions) under consideration led to good or bad economic outcomes for the United States as a whole. Clearly, one could address a more realistic decision-theoretic framework in which multiple stakeholders (Congress, generals, intelligence analysts, Halliburton, and many others) negotiate or coalesce or diverge in reaching a military decision, and that would surely lead to fascinating work in sociology and political science. But such modeling was not our intent. And we appreciate Dr. Ratner’s recognition that our primary goal was the cost-benefit analysis. Our emphasis on “the U.S.-centric utility function” bothered Dr. Ratner, and we readily acknowledge that it makes us morally uncomfortable too. We would prefer to live in a world in which the United States is not indifferent to the suffering of others and where altruism is part of the calculus of leadership. And we also think that considerations of decency are usually given some weight in the corridors of power. However, we also believe that a callous calculation of the bottom line is a necessary component of military and other policy decisions. Absent that starting point, there seems to be no principled basis for prioritizing cases and causes. Dr. Ratner would prefer to see “a sensitivity analysis, with an alternative, semi-altruistic utility function.” We think that would be interesting and useful, and effective altruism is always important. But (as Dr. Ratner points out later), our article is already heavily freighted with assumptions that have varying degrees of plausibility. Trying to monetize the lives of non-American","PeriodicalId":43397,"journal":{"name":"Statistics and Public Policy","volume":"6 1","pages":"122 - 123"},"PeriodicalIF":1.6,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/2330443x.2019.1688741","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42677444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：