Which retail services are co‐located?. Issue 4 (2014)

• Record Type: Journal Article
• Is Part Of: International journal of retail & distribution management. Volume 42:Issue 4(2014)
• Page Start: 281
• Page End: 297
• Publication Date: 2014
• Subjects: Retail -- Factor analysis -- Co‐location -- Retail location; Retail trade -- Management -- Periodicals; Physical distribution of goods -- Management -- Periodicals; 658.87
• DOI: 10.1108/IJRDM-11-2012-0105
• Languages: English
• ISSNs: 0959-0552

Abstract

Abstract : Purpose – Which retail services are co‐located in space? Is it possible to categorize retail stores of different kinds with respect to their location pattern? Acknowledging the spatial dependency between different and similar kinds of retailing activities, the aim of this paper is to find if and to what extent co‐location is present in a retail market and what kind of retailing activities are co‐located.Design/methodology/approach – The authors analyse the co‐location of different types of retail stores in Sweden by using geo‐coded data. The data they use allows them to pinpoint each establishment in Sweden down to a 250 by 250m square in space. First, they identify a measure of co‐location for each retail service by utilizing pairwise correlations between the different retail service establishments with respect to the squares in which they are present. Later, by using the finest level of industrial categorization for all physical retailing activities (and limiting their geographical unit to the Stockholm metropolitan market), they perform factor analysis to nest these retailing activities under relevant categories based on their co‐location pattern.Findings – In this analysis the authors obtain four major factors for the squares with retail stores, in which several kinds of retail activities are nested. These factors reveal a certain degree of location commonality for the markets in question.Originality/value – The authors' empirical design is based on a highly disaggregated spatial information and the methodology is novel in a sense that it has not been used to address a similar question. Rather than sampling, the authors use the total population, where they take all physical retailing activities into account to be able to draw a general picture for the co‐location phenomena in the entire retail market.