Delegation at the U.S. Federal Appellate Courts: The Power to Remand As a Double-Edged Sword

Sarel, Roee; Demirtas, Melanie

doi:10.2139/ssrn.3094634

“…It bears repeating that this appeal is brought ... (2014), Sarel and Demirtas (2017) and used BOW and doc2vec (Le and Mikolov 2014) as the baselines. It is shown that our model has achieved a competitive performance.…”

Section: Methodsmentioning

confidence: 99%

Automated Fact-Value Distinction in Court Opinions

Cao

¹

,

Ash

²

,

Chen

³

2018

SSRN Journal

0

View full text Add to dashboard Cite

This paper studies the problem of automated classification of fact statements and value statements in written judicial decisions. We compare a range of methods and demonstrate that the linguistic features of sentences and paragraphs can be used to successfully classify them along this dimension. The Wordscores method by Laver et al. (Am Polit Sci Rev 97(2):311-331, 2003) performs best in held out data. In an application, we show that the value segments of opinions are more informative than fact segments of the ideological direction of U.S. circuit court opinions.

show abstract

“…Vectorization outputs are fed to a Logistic Regression classifier. (iv) ws: a BOW-like method implementing WordScore (Laver et al 2003;Sarel and Demirtas 2017). Each word in the training set is assigned a fact-or valueinclination score.…”

Section: Supervised Learningmentioning

confidence: 99%

“…Smith (2014) shows that judges are more likely to exercise policy preference in legal disputes focusing more on interpretations of facts, but less likely to do so in cases focusing more on interpretations of legal principles. In controlling the textual factors that might influence the likelihood of a case to be remanded from appellate courts to district courts, Sarel and Demirtas (2017) have considered whether a case raises more factual questions or more legal questions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated fact-value distinction in court opinions

Cao

¹

,

Ash

²

,

Chen

³

2020

View full text Add to dashboard Cite

This paper studies the problem of automated classification of fact statements and value statements in written judicial decisions. We compare a range of methods and demonstrate that the linguistic features of sentences and paragraphs can be used to successfully classify them along this dimension. The Wordscores method by Laver et al. (Am Polit Sci Rev 97(2):311-331, 2003) performs best in held out data. In an application, we show that the value segments of opinions are more informative than fact segments of the ideological direction of U.S. circuit court opinions.

show abstract