Measures and operationalization

This section describes the central measures and the operationalization of each measure to clarify the difference between the studied concepts. The variables utilized in this dissertation are based on established measures deployed by prior studies. Entrepreneurial orientation, absorptive capacity, and competitive intensity are borrowed directly from prior studies as such, whereas new product and service portfolio advantage and success are developed based on prior NPD research for purposes of this study. It was necessary to develop the NPSP advantage measure because prior research had investigated the advantageous characteristics of new products but neglected to investigate the characteristics of services or product–service combinations that drive success. All measures captured through the online survey utilize retrospective approach suggested by prior studies (Kumar, Petersen, & Leone, 2013; Miller, Cardinal, & Glick, 1997) to reflect the same time period with objective financial data from years 2010, 2011, and 2012. The items deployed by measures are presented in the appended articles in the second part of this dissertation. Sales growth, profitability, slack resources, firm age, and firm size are actual values for each company obtained from the ORBIS database.

Sales growth was calculated as the average annual change in turnover between 2009 and 2012. Turnover information was adapted from ORBIS database and thus represents the real values. Although objective financial measures are argued to be misleading in multi-industry samples (Covin, Slevin, & Schultz, 1994) because different industries tend to differ in terms of growth and profitability, objective performance measures can offer an accurate way to capture firm performance as they are free from the respondents’ perceptions and opinions. Article 1 investigates the EO–sales growth relationship.

Profitability (EBIT %) refers to the average EBIT percentage rate from the years 2010, 2011, and 2012. Profitability was also adapted from data drawn from the ORBIS database. The EBIT percentage was selected as the profitability measure because it is not affected by national taxation, so enabling comparison between studies using samples from other countries. Articles 2 and 4 utilize the profitability measure.

Entrepreneurial orientation (EO) was adapted from a recent study by Patel et al. (2015), which builds on the most commonly deployed operationalization of a 9-item scale devised by Covin and Slevin (1989) (Rauch et al., 2009). The EO measure is deployed in Articles 1, 2 and 4. By following the suggestions in prior EO studies (Lumpkin & Dess, 2001; Richard, Barnett, Dwyer, & Chadwick, 2004), the

authors deployed a multidimensional construct structure reflecting three dimensions: proactiveness, innovativeness, and risk taking. It may be worth mentioning that a unidimensional measure reflecting the same three dimensions has also been used by EO researchers (Covin et al., 1994; Stam & Elfring, 2008;

Wiklund & Shepherd, 2005) and that both structures are widely accepted.

Absorptive capacity (ACAP) was captured through the 22-item scale developed by Jansen et al. (2005) that is built on the 7-point scale devised by Zahra and George (2002) reflecting four dimensions: knowledge acquisition, assimilation, transformation, and exploitation. Knowledge acquisition captures a firm’s ability to acquire potentially valuable external knowledge. Assimilation refers to the practices and processes through which the acquired knowledge is internalized. The transformation dimension captures organizational practices turning assimilated knowledge into a utilizable format. The fourth dimension, exploitation, refers to a firm’s ability to apply the transformed knowledge for commercial ends. Accordingly, whereas EO is a disposition toward entrepreneurial behavior, ACAP can be considered a capability facilitating knowledge processing and utilization that assists in the execution of entrepreneurial initiatives. The ACAP measure is used in Articles 1 and 2.

Slack resources (SR) were measured through the current ratio average of the years 2010, 2011, and 2012 as drawn from the ORBIS database. Accordingly, the measure reflects the financial slack resources that could be deployed to support the development of an organization. Whereas EO is considered an organizational strategic posture and mindset affecting responsiveness to new market opportunity identification and capture, slack resources provide the required pool of potential utilizable resources to engage in innovative and risky entrepreneurial endeavors.

The slack resources variable is used as a moderator in Article 1.

New product and service portfolio (NPSP) advantage was built on the new product advantage scales deployed in prior studies (Atuahene-Gima, 1995;

Chen, Reilly, & Lynn, 2012; Cooper & Kleinschmidt, 1987; Im & Workman, 2004;

McNally et al., 2010; Rijsdijk et al., 2011). Article 3 reports the development of NPSP advantage in detail and Article 4 investigates the role of NPSP advantage in the EO–profitability relationship. As NPSP advantage may be considered a desired innovation outcome, the development of the NPSP advantage measure and the study (Article 4) investigating EO’s impact on NPSP advantage was particularly important to ensure the main objective of this dissertation was met. Based on the established measures, a 15-item scale was built to reflect three dimensions: the novelty, meaningfulness, and superiority of new products and services introduced in the past three years (2010, 2011, and 2012). The three-dimensional construct

structure was found to offer the best fit with the data, indicating that these three advantageous characteristics are distinct from each other but together constitute the NPSP advantage.

New product and service portfolio (NPSP) success was built on the measure of five new product success items introduced by Cooper and Kleinschmidt (1987). The original items were rephrased to reflect the success of both new products and services and then revalidated. NPSP success indicates the perceived success of new products and services introduced within a three-year period (2010, 2011, and 2012) in terms of sales, market share, return on investment, profitability, and senior management satisfaction.

Competitive intensity (COMIN), a 5-item scale, was adapted from a prior study by Jaworski and Kohli (1993). Competitive intensity is used in Articles 1 and 4 as a control variable. Competitive intensity refers to the characteristics of the business environment reflecting to what extent the competition is perceived as fierce and aggressive. Despite the single-industry sample, the authors wanted to control for possible differences in the competitive environment of food manufacturers because the companies serve customers in different geographical areas, which can directly affect the competitive environment. In addition, the food manufacturing industry consists of nine sub-industry classes, so for example, manufacturers of dairy products and pet food producers represent distinct sub-industries and serve completely different customer groups; meaning the producers are operating in a dissimilar competitive environment. Finally, as the EO–

performance relationship is apparently influenced by the nature of the business environment (Rauch et al., 2009), even in single industry studies it is wise to control for the possible differences in environmental characteristics within the industry.

Firm age (AGE) is used in Articles 1 and 4 as a control variable and was derived from the ORBIS database by calculating the difference between the year of establishment and the year 2012. In investigations of firm performance, age is commonly used to control the results (Kollmann & Stöckmann, 2014) as younger companies can enjoy above-average sales-growth performance in some samples.

Prior studies have found EO and firm age to have a negative correlation (Engelen et al., 2014). This could mean that companies are more likely to lose their proactive, innovative, and risk-taking attitude toward new market opportunities as they age. The results of this study offer no exception to that notion, as reported in Article 1; in the data, EO has a minor statistically significant negative correlation with firm age (p<.05).

Firm size (SIZE) was derived from the ORBIS database by calculating the average number of employees in the years 2010, 2011, and 2012. Size was used as a control variable in Articles 1 and 4. Similarly to age, firm size is also often controlled for when investigating the performance effects of EO (Kollmann &

Stöckmann, 2014).