Economic importance and statistical significance: Guidelines for communicating empirical research

Abstract

A critical objective for many empirical studies is a thorough evaluation of both substantive importance and statistical significance. Feminist economists have critiqued neoclassical economics studies for an excessive focus on statistical machinery at the expense of substantive issues. Drawing from the ongoing debate about the rhetoric of economic inquiry and significance tests, this paper examines approaches for presenting empirical results effectively to ensure that the analysis is accurate, meaningful, and relevant for the conceptual and empirical context. To that end, it demonstrates several measurement issues that affect the interpretation of economic significance and are commonly overlooked in empirical studies. This paper provides guidelines for clearly communicating two distinct aspects of “significance” in empirical research, using prose, tables, and charts based on OLS, logit, and probit regression results. These guidelines are illustrated with samples of ineffective writing annotated to show weaknesses, followed by concrete examples and explanations of improved presentation.     

An Empirical Investigation into the Perception of Vehicle Source Effects

Advertising managers, educators and regulators alike have observed the growth of federal programs mandating the form or content of consumer information disclosures. Although this activity often has been warranted, at times these programs may have had unintended “deceptive” effects. Examples might include the USDA's food grading systems, some of the FTC's corrective advertising activities and certain programs of affirmative disclosure of product information (e.g, gasoline octane ratings and the cigarette health warning). To avoid these problems one must give attention to principles learned from communication and attitude research. Further, such programs will benefit greatly from before-after impact evaluation studies to insure that results correspond with objectives.     

How to learn new tasks: Shop floor performance effects of knowledge transfer and performance feedback

We investigate how learning and the task performance of individuals are affected by different forms of knowledge transfer. Whereas previous research has proven the positive performance impacts of knowledge transfer, self-observation and feedback mechanisms individually, we explore the cumulative effect of these factors on learning and performance. With the help of two laboratory experimental studies reproducing manufacturing tasks that are typical for industrial production, we show that explicit knowledge transfer is superior to other forms of knowledge transfer. Externally provided performance feedback in the form of cost information and non-financial performance indicators has no effect on the order of different forms of knowledge transfer. Moreover, external feedback does not even have an additional significant performance effect on learning new tasks irrespective of the type of knowledge transfer.     

Challenges for land system science

While considerable progress has been made in understanding land use change, land system science continues to face a number of grand challenges. This paper discusses these challenges with a focus on empirical land system studies, land system modelling and the analysis of future visions of land system change. Contemporary landscapes are contingent outcomes of past and present patterns, processes and decisions. Thus, empirical analysis of past and present land-use change has an important role in providing insights into the socio-economic and ecological processes that shape land use transitions. This is especially important with respect to gradual versus rapid land system dynamics and in understanding changes in land use intensity. Combining the strengths of empirical analysis with multi-scale modelling will lead to new insights into the processes driving land system change. New modelling methods that combine complex systems thinking at a local level with macro-level economic analysis of the land system would reconcile the multi-scale dynamics currently encapsulated in bottom-up and top-down modelling approaches. Developments in land use futures analysis could focus on integrating explorative scenarios that reflect possible outcomes with normative visions that identify desired outcomes. Such an approach would benefit from the broad and in-depth involvement of stakeholders in order to link scientific findings to political and societal decision-making culminating in a set of key choices and consequences. Land system models have an important role in supporting future land use policy, but model outputs require scientific interpretation rather than being presented as predictions. The future of land system science is strongly dependent on the research community's capacity to bring together the elements of research discussed in the paper, via empirical data collection and analysis of observed processes, computer simulation across scale levels and futures analysis of alternative, normative visions through stakeholder engagement.     

Changes over time in the return to education in urban China: Conventional and ORU estimates

Studies of the return to education in urban China have reported that this has increased over time, and that females typically have a higher return than males. In this paper we adopt a framework provided by the over education/required education/under education literature, and the decomposition developed by Chiswick and Miller (2008), to investigate the reasons for these findings. The finding by Chen and Hamori (2009), from analysis of data for 2004 and 2006, of the return to schooling for males exceeding that for females, is also examined using this decomposition.     

Travelling to the Ancestral Homelands: The Aspirations and Experiences of a UK Caribbean Community

This paper is based on an ethnographic study of the Caribbean community of Moss Side, Manchester (UK). Its aim is to reveal, interpret and analyse the personal meanings which members of the community attach to visiting the ancestral homeland in the Caribbean. This form of travel is defined in terms of 'ethnic reunion', which involves travelling for the purpose of visiting friends and relatives and/or searching for one's cultural roots. The study, which is based on an interpretive analysis of a range of ethnographic material, initially examines the reasons why first- and second-generation Caribbeans wish to participate in the 'homeland experience', and then illustrates ways in which they reconstruct an identity of themselves through their travel perceptions and experiences. The latter part of the paper discusses how people's travel encounters serve to illustrate how ethnic differences and boundaries between groups are constructed and/or reconstructed. It is argued that established perspectives of tourism motivation and behaviour do not fully account for the role of ethnicity as a significant variable in influencing specific forms of travel. The conclusion asserts that ethnic reunion should be conceptually viewed as a distinct form of 'travel', socioculturally dissimilar to conventional forms of 'tourism'.     

Reasonableness and Section 4 of the Employment Act 1980

This article explains the origins of s.4 of the Employment Act 1980 and assesses the place of this provision in the corpus of labour legislation enacted by the Conservative government since 1980. Its principal aim is to concentrate upon the concept of reasonableness which is central to the operation of the provision and to evaluate critically the role of the EAT and the industrial tribunals in interpreting it. Finally, the article seeks to assess the impact of the tribunaľs interpretation of s.4 on union admission and expulsion procedures.     

Regularity conditions and scope estimates: The case of large-sized U.S. banks

We consider economies and diseconomies of scope for large U.S. banks by employing ordinary and hybrid translog cost functions. We examine the regularity conditions in output space where scope estimates are calculated and reject all models for which these conditions fail. The translog model always possesses violations. For the hybrid translog, violations occur in every case except one. In this one case, we find economies of scope.     

Production Externalities and Urban Configuration

We formally explore Jacobs' idea that uncompensated knowledge spillovers are crucial for agglomeration by modeling location-dependent interfirm production externalities in a general-equilibrium linear-city framework. Good and factor allocation, firms' and households' locational choice, and wages and land rents are all endogenously determined. The unique equilibrium urban configuration may be concentrated (with monocentric firm locations), dispersed (with completely mixed firm/household locations), or a combination, depending on the population of firms, transportation cost, and firm-interaction parameters. Due to distance-dependent production externalities, firms will be clustered together in any equilibrium, thus ruling out the possibility of any multicentric urban configuration. Journal of Economic Literature Classification Numbers: D51, R12.     

Swinburne's Challenge: A Note on Probability Estimates for Nontypical Potential Evidence Instances

General confirmation theory, and especially its Bayesian variant, has never been able to adequately address the issue of how to handle qualitative evidence instances. Such statements encompass a wide class including historical claims, those of the human sciences not incorporating quantitative models, legal statements and the like. The issue was recognized by the philosopher Swinburne (1973) who puzzled how such statements as, Caesar crossed the Rubicon, could be assigned a meaningful probability estimate. The present paper suggests that such statements can be transformed into, at least, plausible probability estimates. This requires a two step process: judgements made by credible raters, and the transformation of judgements into one or more reliability co-efficients. These reliability estimates can then be utilized in the standard Bayesian model to yield plausible degrees of belief between hypothesis and evidence.