Peter E. Kennedy

The Use (and Abuse) of Meta-Analysis in Environmental and Natural Resource Economics: An Assessment

Motivated by the 2006 report of a Work Group appointed by the Environmental Protection Agency (EPA), this paper examines the present state of meta-analysis in environmental economics and offers recommendations for its future use. To this end we summarize and assess 140 meta-analyses from 125 published and unpublished studies, covering 17 topical categories in environmental and resource economics. First, we provide several generic meta-analysis models as reference points and discuss major estimation issues. Five econometric issues are identified as part of a complete analysis: (1) sample selection criteria; (2) basic data summary; (3) primary data heterogeneity; (4) heteroskedasticity; and (5) non-independence of multiple observations from primary studies. Second, a tabular summary is presented for the 140 meta-analyses with respect to estimation methods. Third, a narrative summary is presented for 19 meta-analyses, including the three value-of-statistical-life studies examined by the EPA Work Group and one analysis from each of 16 other categories. Fourth, we offer a set of “best practice” guidelines for future meta-analyses in this and other areas of economics. Last, the paper comments on the use of meta-analytic methods for benefit transfers of environmental values.

Testing for Unit Roots: What Should Students Be Taught?

Abstract Unit-root testing strategies are unnecessarily complicated because they do not exploit prior knowledge of the growth status of the time series, they worry about unrealistic outcomes, and they double- or triple-test for unit roots. The authors provide a testing strategy that cuts through these complications and so facilitates teaching this dimension of the unit-root phenomenon. F tests are used as a vehicle for understanding, but t tests are recommended in the end, consistent with common practice.

Randomization Tests in Econometrics

Numerous shufflings of data produce a distribution of test-statistic values that can be used to assess the degree to which the test-statistic value produced by the actual data is unusual. Because this controversial randomization-testing methodology, made practical by the computer revolution, has begun to appear in applied econometric studies, econometricians should become familiar with its mechanics, rationale, and interpretation, all of which are quite different from the status quo. This article exposits randomization tests in an econometric context, discusses their advantages, and alerts practitioners to pitfalls.

Randomization tests for multiple regression

Four generic means of conducting randomization tests in the context of multiple regression are analysed. Based on their performance in traditional repeated samples, three of these are shown to be inappropriate or applicable only in special circumstances; their shortcomings are illustrated via Monte Carlo studies

A Graphical Exposition of the Ordered Probit

The use of probit and logit models has become quite common whenever the dependent variable in a regression is qualitative. These models have been used to explain either/or choices and decisions involving multiple alternatives. A two-dimensional graphical interpretation of these different models has been provided by Johnson [3]. The purpose of this paper is to provide a three-dimensional graphical exposition of the ordered probit model, which was first estimated by McKelvey and Zavoina [4] and is now built into computer packages, such as LIMDEP [1].

Are Multiple-Choice Exams Easier for Economics Students? A Comparison of Multiple-Choice and “Equivalent” Constructed-Response Exam Questions

Many economics students believe that multiple-choice (MC) exams are easier than constructed-response (CR) exams. If true understanding of economics is reflected by performance on CR exams, then easier MC exams would give rise to misleading measures of student understanding of economics, creating a false sense of complacency among both instructors and students. This article investigates this issue by comparing student scores on MC and &ldquo=uivalent” CR questions. Our results show that, for certain types of MC questions, students do indeed score better, even after correcting for guessing, and on other types of questions, they produce similar scores. These results are unaffected by comparing males versus females or “good” versus “poor” students.

F Versus T Tests for Unit Roots

F tests which test jointly for a unit root and a zero intercept, and so compete against Dickey-Fuller t tests, are shown not to enhance power because they are invariant to the intercept value in the absence of a unit root. Monte Carlo results in the literature that indicate otherwise are shown to have resulted from the use of special starting values. Testing procedures that employ these F tests to enhance power should be revised.

Combining qualitative forecasts using logit

This paper introduces a computationally-convenient means of combining qualitative forecasts, through use of logit regression, applicable in dichotomous, polychotomous and ordered poluchotomous contexts.

Does Teaching Enhance Research in Economics

An extensive literature in education shows little relationship between teaching and research. Authors advance the notion that research enhances teaching, but typically ignore the prospect of teaching enhancing research. We report the results of a survey of active researchers in economics who were asked if they could cite a specific instance in which their research was substantively influenced by their teaching. About 50% reported such an experience, with a further 35% indicating, without prompting, that their teaching had a very positive effect on their research in general. Their comments identify a variety of channels through which teaching influences research.

An Agenda for Research on Economic Education in Colleges and Universities.

Future research on economic education at the postsecondary level should be directed toward the measurement of multiple outputs, the economics major, and replication.