Introduction

---

When conducting inter-organizational information systems (IOIS) research, a decision needs to be made about which third-party organizations to analyze and how to collect the data. When analyzing buyer/seller relationships, many researchers ask one of the parties in the dyad to rate their perceptions of the relationship with a buyer/seller (e.g., Anderson & Narus, 1990; Batt & Rexha, 1999; Doney & Cannon, 1997; Gundlach, Achrol, & Mentzer, 1995; Heather, 2001; Kumar, Stern, & Achrol, 1992; Leuthesser & Kohli, 1995; Morgan & Hunt, 1994). Similarly, when studying information systems interactions between buyers and sellers, many researchers ask one party in the dyad to rate their perceptions of communication flows and how the information system operates (e.g., Bensaou, 1997, 1999; Bensaou & Venkatraman, 1995; Mohr, Fisher, & Nevin, 1996; Mohr & Sohi, 1995). Yet when looking at data collected from both parties in the dyad (Figure 1, “matched dyadic pairs”), it was found that buyers and sellers have significant differences in perceptions about the relationship and what information is shared (Spekman, Kamauff, & Myhr, 1998; Storer et al., 2002). From this, it may be concluded that it would be more valid to collect data from matched dyadic pairs or at least from samples of pools of buyers and pools of sellers (Figure 1, “pooled dyadic pairs”). However, it is proposed that an alternative may be to collect data from the one focal organization but to ask questions about two third-party organizations so that comparisons of responses can be made (Figure 1, “comparative pairs”).

Figure 1: Alternative units of data collection for IOIS chain research

A chain is normally defined as three or more linked organizations (Hines, 1998), such as the focal firm, a supplier, and a customer, e.g., a food processor, a primary produce supplier, and retailer customer. However, many quantitative chain studies only evaluate dyadic relationships such as a manufacturer and their suppliers (e.g., Ellinger, Daugherty, & Plair, 1999; Holland, 1995). Other than research done by Spekman et al. (1998), the main exception is when case studies of the chains or networks of a focal firm are examined (e.g., Gifford, Hall, & Ryan, 1998; Samuel & Hines, 1998; Trienekens, 1999; Van der Vorst, 2000). To gain a full chain perspective, some researchers start by studying cases and follow up with surveys of focal firms (e.g., Kanflo, 1998). This two-phase research process extends the time taken to conduct the research.

The concern in relying on quantitative chain research based on focal firm or dyadic studies is that only part of the chain is being studied. The differences in responses of sales and purchasing staff found by Spekman et al. (1998) and Storer et al. (2002) may have more to do with the greater bargaining power of customers compared to the lesser power of suppliers. Responses may be affected by the position of the organization in the chain.

As an alternative to multiple case studies, dyadic studies, or focal firm surveys, it is proposed that data can be collected from a focal firm about both customers and suppliers (Figure 1, “comparative pairs”). For example, data could be collected from a food processor about their customers (retailers, wholesalers, or other food processors) and their suppliers (primary producers, packaging suppliers, and other food processors).

---