Research summary: This article demonstrates the potential of computer-aided text analysis (CATA) as a technique to operationalize hard-to-measure constructs in international business research, provided that a rigorous set of validity tests is applied. CATA allows researchers to perform content analyses on large textual databases by constructing indicators using deductively and inductively derived keywords. We show the critical validity steps that have to be followed to arrive at valid CATA indicators. We illustrate the CATA technique through an application to the concept of global mind-set, which has received substantial attention in the international business and strategy literature. We conclude that CATA analysis is a valuable method for international business research, but that its potential can be unleashed only with proper procedures and due attention to construct validity.Managerial summary: With the increasing availability of large textual databases such as collections of press releases, newswire archives, and SEC filings, computeraided text analysis (CATA) creates new opportunities to analyze otherwise unobserved firm and managerial traits. CATA can be applied to a broad range of firm activities and industries across long time periods, while eschewing the low response rates typical for surveys. In this article, we demonstrate the critical validity steps that have to be followed to arrive at valid CATA indicators of firm and managerial traits, with an application to international business. As an empirical illustration, we build a keyword-based indicator of firms' global mind-sets using a large dataset of news articles.

This paper demonstrates the potential of computer-aided text analysis (CATA) as a technique to operationalize hard-to-measure constructs in international business research, provided that a rigorous set of validity tests is applied. CATA allows performing content analyses on large textual databases by constructing indicators using deductively and inductively derived keywords. We show the critical validity steps that have to be followed to arrive at valid CATA indicators. We illustrate the CATA technique through an application to the concept of global mindset, which has received substantial attention in the international business and strategy literature. We conclude that CATA analysis is a valuable method for international business research, but that its potential can only be unleashed with proper procedures and due attention to construct validity.

MNCs often engage in international research collaborations with foreign universities through one of their central R&D laboratories (at headquarters or elsewhere) even though they operate a local R&D unit close to that university, and hence forego the benefits of geographic proximity and local collaboration. Drawing on the knowledge-based theory of the firm, we hypothesize that the choice between distant and local collaboration systematically relates to the knowledge capabilities of the firms’ R&D units, the characteristics of the focal knowledge, and local knowledge leakage risks. Analysis of close to 13,000 research collaborations with foreign universities by the world’s major biopharmaceutical firms (1995–2015) confirms that collaboration at distance occurs if this allows the firm to benefit from scale and knowledge diversity advantages, if the central unit has strong basic research capabilities, and if collaboration is in a core research domain of the MNC while rival firms are locally present. Maturity of the focal research domain is associated with local collaboration. Our findings qualify the common arguments in favor of collaboration in proximity and suggest that (distant) central R&D units are important orchestrators of research collaboration with universities around the globe.

No abstract

No abstract