Williams College

About: Williams College is a education organization based out in Williamstown, Massachusetts, United States. It is known for research contribution in the topics: Population & Politics. The organization has 2257 authors who have published 5015 publications receiving 213160 citations. The organization is also known as: Williams.

PolyTOIL: A Type-Safe Polymorphic Object-Oriented Language

Abstract: PolyTOIL is a new statically-typed polymorphic object-oriented programming language which is provably type-safe. By separating the definitions of subtyping and inheritance, providing a name for the type of self, and carefully defining the type-checking rules, we have obtained a language which is very expressive while supporting modular type-checking of classes. The matching relation on types, which is related to F-bounded quantification, is used both in stating type-checking rules and expressing the bounds on type parameters for polymorphism. The design of PolyTOIL is based on a careful formal definition of type-checking rules and semantics. A proof of type safety is obtained with the aid of a subject reduction theorem.

The Ease-of-Processing Heuristic and the Stability Bias Dissociating Memory, Memory Beliefs, and Memory Judgments

Abstract: Judgments about memory are essential in promoting knowledge; they help identify trustworthy memories and predict what information will be retained in the future. In the three experiments reported here, we investigated the mechanisms underlying predictions about memory. In Experiments 1 and 2, single words were presented once or multiple times, in large or small type. There was a double dissociation between actual memory and predicted memory: Type size affected predicted but not actual memory, and future study opportunities affected actual memory but scarcely affected predicted memory. The results of Experiment 3 suggest that beliefs and judgments are largely independent, and neither consistently resembles actual memory. Participants’ underestimation of future learning—a stability bias—stemmed from an overreliance on their current memory state in making predictions about future memory states. The overreliance on type size highlights the fundamental importance of the ease-of-processing heuristic: Information that is easy to process is judged to have been learned well.

Contractual Mix in Southern Agriculture since the Civil War: Facts, Hypotheses, and Tests

Abstract: In the South after 1865, workers and property owners employed a variety of contracts—wage payment, crop sharing, and land rental—to bring together cooperating resources in agricultural production. The contractual mix varied over time and space, depending on the relative resource endowments of the contracting parties, the prevailing risk conditions, and the transactions costs of alternative contractual arrangements. To understand the contractual mix, certain empirical distinctions must be made, and the major hypotheses advanced to explain the mix must be seen as complementary rather than mutually exclusive. These hypotheses, however, differ in their demonstrated ability to account for the empirical variance. In addition to factual clarification and theoretical explication, the paper presents a new sample of plantation data and a new econometric procedure for performing more detailed and better controlled tests of hypotheses.

Roe V. Wade and American Fertility

Abstract: We consider the effect of abortion legalization on births in the United States. A simple theoretical model demonstrates that the impact of abortion legalization on the birth rate is ambiguous, because both pregnancy and abortion decisions could be affected. We use variation in the timing of legalization across states in the early 1970's to estimate the effect of abortion on birth rates. Our findings indicate that states legalizing abortion experienced a 5% decline in births relative to other states. The decline among teens, women over 35, and nonwhite women was even greater: 13%, 8%, and 12% respectively. Out-of-wedlock births declined by twice as much as births in wedlock. If legalization in some states affected birth rates in neighboring states (through travel to obtain an abortion), comparing births between states will underestimate the actual reduction. Using more distant comparison states increases the estimated impact of abortion legalization on birth rates to about 8%. Applying this estimate to the current level of births, a complete recriminalization of abortion would result in 320,000 additional births per year.