Kurt C. Schaefer

The uses and misuses of data and models : the mathematization of the human sciences

PART ONE: FOUNDATIONS Oracles, Norms and Science Modeling Dreams and Disappointments PART TWO: THE INFORMATION CYCLE A Priori Influences on the Information Cycle Measurement of Human Information Limitations of Measurement in the Social Sciences Information for Inferences What Are Social Science Data? Causality Models and Policy Making

Monte Carlo Appraisals of Gravity Model Specifications

Many improvements have been proposed for the basic gravity model specification, most of which are confirmed by standard statistical tests due to the large number of observations often used to estimate such models. We use Monte Carlo experiments to examine situations in which features of models may be found statistically significant (or insignificant) when it is known ex ante that they are absent (or present) in the underlying data process. Erroneous assumptions about the presence or absence of lagged dependent variables, fixed effects, free-trade associations and customs unions are shown to introduce economically important bias in estimates of the coefficients of interest, and in some cases to be confirmed spuriously. Policy effects, such as for free trade associations and currency unions, can also be confirmed spuriously when they do not exist in the data-generating process.

Rural Community Characteristics, Economic Hardship, and Peer and Parental Influences in Early Adolescent Alcohol Use

The study explores how differences in rural community contexts relate to early adolescent alcohol use. Data were gathered from 1,424 adolescents in the sixth through eighth grades in 22 rural Northern Plains communities, as well as 790 adults, parents, teachers, and community leaders. Multilevel modeling analyses revealed that community supportiveness, as perceived by adolescents, but not adults, was associated with less lifetime and past month alcohol use, and for past month use, this relationship was stronger than perceived peer drinking or parental closeness. Perceived peer drinking and parental closeness were not associated with past month use. Adolescents experiencing family economic strain did not report greater lifetime or past month use, but living in a disadvantaged community was associated with greater past month use. Relatively affluent adolescents reported greater past month use when living in a poor community than did poorer adolescents, highlighting relationship complexity between economic disadvantage and alcohol use.

Can Price Dispersion Reveal Distance-Related Trade Costs? Evidence from the United States

Abstract We offer insights on how distance-related trade costs may best be inferred from price-dispersion measures. Using a simple spatial model of price dispersion, we argue that measures of price dispersion that are not spatially informed can mislead researchers into concluding that distance-related costs are small even when such costs are the major determinant of price dispersion. With intra-United States data on eleven goods, we find that distance-related costs are large and are indeed underestimated when inferred from standard, non-spatial, price dispersion measures. Our empirical findings have implications for studies of market integration policies (such as trade liberalization and currency unions) and the significance of economic geography.

Reverse Regression and Orthogonal Regression in Employment Discrimination Analysis

In a recent review article, White and Piette provide an overview of the use of reverse regressions in discrimination-related litigation. They explain the technique, provide a model application, summarize its advantages and disadvantages, and identify litigation in which it has been used. We point out weaknesses in common uses of reverse regression, some of which might cause serious misinterpretations of the data. We suggest that typical interpretations of reverse-regression results are incorrect. We also question the practice of conducting both direct and reverse regressions when studying employment discrimination, since the two approaches make mutually-inconsistent assumptions about the nature of the stochastic error; these assumptions generally can not be corroborated from the data and bias the regressions’ results. We suggest orthogonal regression as a potential alternative to the problems associated with direct and reverse regressions in some cases. We then provide a model application of orthogonal regression to discrimination-related investigations.

Price Dispersion in Spatial Perspective: Theory and Evidence

We develop a simple Hotelling spatial-trade model of price dispersion to examine how distance-related costs affect price dispersion, and we offer some insights on how such costs may best be inferred from price-dispersion measures. Our theoretical model suggests that measures of price dispersion that are not spatially-informed can mislead researchers into concluding that distance-related costs are small even when such costs are the major determinant of price dispersion. In estimates based on price dispersion across U.S. cities of eleven goods, we find that distance-related costs are large and are indeed under-estimated when inferred from standard, non-spatial, price dispersion measures.

AFDC and births to unwed women

Abstract Most studies by economists have been inconclusive when seeking a consistent relationship between income-support programs [like aid to families with dependent children (AFDC)] and births to unwed women (or, as the literature traditionally terms it, illegitimacy). But a recent study [Southern Econ. J. 62 (1995) 44] reports a large, positive and statistically significant relationship when data are weighted to reflect differences in propensities toward illegitimacy. We find that the 1995 study appears to rely upon erroneous data and irregular econometric technique. When these are remedied, the major results are reversed. We then suggest that a switching regimes methodology, with parameters influenced by other variables, is more appropriate to the issue. Our empirical results confirm the literatures consensus that AFDC and illegitimacy do not appear to be strongly related.

An analytic model for emergency food provision

Abstract Donor agencies employ a range of food provision and cost-sharing policies, including food linked to cash distribution grants, food sales (“monetization”), and distribution of food through a local agency. This paper develops a systematic methodology for choosing the proper blend of policies. The model requires estimates of several simple economic indicators in the recipient country. Two numerical applications of the model are developed, and extensions of the model for peculiarities of specific fields are suggested.