Henry Renski

Clusters and Economic Development Outcomes An Analysis of the Link Between Clustering and Industry Growth

Much of the existing empirical research on industry clusters focuses on the detection of clusters for economic development purposes. There are comparatively few studies that relate identified clusters to business and industry growth or that trace changes in designated clusters over time. This article seeks to better understand the link between industry clustering and regional economic outcomes. In a comprehensive study sponsored by the Appalachian Regional Commission and released in 2002, the authors identified technology-based clusters within and on the border of the Appalachian region. The Appalachian technology clusters constituted subregional concentrations of related industrial, research and development, and university-based strengths as of the middle to late 1990s. In this article, the authors investigate how the industries in the identified clusters fared over the subsequent several years in terms of employment and new business formation. They find evidence that clustering is associated with new business formation for selected technology industries but not with employment growth.

Differences in Labor versus Value Chain Industry Clusters: An Empirical Investigation

Regional analysts often identify industry clusters according to a single dimension of industrial interdependence, typically by trading patterns as revealed in national or regionalized input-output data. This is despite the fact that the theory underpinning regional industry cluster applications draws heavily on Marshalls theory of external economies, including the important role of labor pooling economies and knowledge spillovers in addition to spatially co-located suppliers. This article investigates whether industry clusters identified based on trading relationships (value chain clusters) are meaningfully different in industrial composition and geography than those derived from an analysis of occupational employment requirements (labor-based clusters). The results suggest that value chain linkages are a weak proxy for shared labor requirements, and vice versa.

The Influence of Industry Mix on Regional New Firm Formation in the United States

Renski H. The influence of industry mix on regional new firm formation in the United States, Regional Studies. Per capita rates of entry are commonly used to measure the regional entrepreneurial climate. Yet entry rates vary widely by industry and tend to mirror existing regional specializations. Without controlling for industry mix, factors associated with regional differences in entry may describe the industry base rather than explain variations in entrepreneurial climate. This study finds that although most of the factors commonly associated with the regional entrepreneurial climate remain significant after purging the data of industry-mix effects, others – namely educational attainment, homeownership, university research and development, and unemployment – are contingent upon industry structure.

A (Short) History of Social Media Sentiment Analysis

In this chapter, we introduce the concept of sentiment analysis as a way to study the prevalence of positive or negative sentiments in expressed attitudes and opinions. We review the ways researchers broadly have used sentiment analyses of Twitter data and other digital data and the established benefits and limits of this method for urban social science research. We then discuss how Twitter data and other similar forms of social media data have been applied in a wide variety of urban planning issues and projects across the globe.

Evaluating Alternative Migration Estimation Techniques for Population Estimates and Projections

Accurate population estimates are critical to effective planning and policy. This study evaluates county population estimates using three different approaches to estimate domestic migration—typically the most volatile component of population change. Using the 2010 Census as a benchmark, it compares the standard net-migration approach to two gross migration approaches—a biregional approach and a multiregional approach that models migration between county pairs. The biregional model produces the lowest average error and is a good choice when producing estimates for a large number of diverse areas. The multiregional model works well for many counties, but is prone to extreme errors.

A National Comparison: Twitter versus the American Housing Survey

We examine differences in resident perceptions of neighborhood quality of life, as well as expressed positive and negative sentiment while accounting for changes in population among cities between 1970 and 2010. We find no evidence that population loss leads to a lower evaluation of life satisfaction. Additionally, we find while tweets are a source for consistently determining the positive and negative affect of individuals on a geographic basis and that people generally have a positive feeling about their neighborhood, there are no significant relationships between the Twitter data sets and traditional ones. Thus, planners or policy makers should not presume that a singular measure will provide a complete picture of well-being.

Estimating the Returns to Professional Certifications and Licenses in the U.S. Manufacturing Sector

This study uses recently released Current Population Survey microdata to estimate the earnings premium associated with professional certification and licenses. The author finds that full-time manufacturing workers with a certification or license earn close to

Taking Microblogging Data for a Test Drive

200 more in median weekly earnings compared to those without. However, this does not account for differences in pay that are associated with worker endowments, such as education and gender. A Blinder–Oaxaca decomposition is used to distinguish the portion of the earnings gap that is attributable to the credential from the portion associated with endowments. Endowments explain 62% of the total earnings gap, meaning that the actual returns to a certification or license are closer to

A Close Look at Urban Immigrant Communities

70 per week. The author also finds that workers with no high school or college education receive a relatively larger increase in weekly earnings, compared to those with more advanced degrees.

A new era of federal involvement in regional economic development? The case of the WIRED initiative

In this chapter we provide an account of our attempt to analyze Twitter data. We describe our methods for creating a database of over 100,000 tweets produced by users in the city of New Bedford, Massachusetts. We attempt to analyze the way in which the Twitter messages engaged with the topics of urban policy and find there is a cursory overlap. We also compare the commercially available IBM SPSS Modeler to our custom-designed sentiment analyzer. Both methods showed a relatively low percentage of sentiment overall but a greater prevalence of positive tweets. Overall, we note that the magnitude of microblogging data and the ability to capture it readily and improvements in analysis techniques may allow for quantity to compensate for low percentages of sentiment.