Social proximity is the inverse of social distance which is a subdimension of cognitive distance. It can be evident between individuals or between an individual and a social group who are not connected by ties, e.g. between a local, native citizen and members of a foreign community club (‘Kulturverein’). Several authors have shown that mental construal, i.e. representation, of socially distant people is more abstract than of socially
proximate individuals or groups (Liberman & Trope, 2008). I operationalise social proximity as individual social capital with the target group. Higher social capital, e.g.
through the joining of foreign community clubs is associated with increased social proximity or diminishing social distance between an individual and a social group (Glaeser et al., 2002). Social capital can be measured on different levels, e.g. for companies, groups or individuals. It can be regarded as the goodwill available to individuals or groups. The value of social capital “lies in the structure and content of the actor’s social relations. Its effects flow from the information, influence, and solidarity it makes available to the actor” (Adler & Kwon, 2002, p. 23). This definition highlights the information exchange facility of social capital, which is inversely related to social distance. When the social capital of a product developer to silver agers is higher, ceteris paribus, these individuals will have a higher likelihood of learning customer needs than less exposed persons. Thus, I can infer that social capital with members of the distant target group can increase customer need knowledge, which leads to lower social distance.
Social capital is characterised by structure (of one’s network, e.g. number of ties and position in network) and tie quality (e.g. content, closeness or information transmission through tie), the former denoting structural and the latter relational embeddedness (Moran, 2005). The focus of this study is on the individual product developer and his or her relation to a distant target group. Participants’ individual networks with members of the target group are expected to explain much of the hypothesised distance effect when developing products for distant target groups. For this purpose, egocentric networks were collected through the online participant pre-survey, consistent with established social capital and network literature (Burt, 2000; Carrasco, Miller, & Wellman, 2008;
Obstfeld, 2005). Egocentric networks present all the relevant exchange partners (alteri) from among all the contacts of the focal person (ego). Thus, social capital in this context disregards network position (structural holes and centrality) and differentiation between inter-organisation (company) and outside ties (Burt, 2000; Madjar, 2008; Madjar, Oldham, & Pratt, 2002).
In order to measure both ego-network size and tie quality, the widely acknowledged operationalisation of strong ties was used (Fredberg & Piller, 2011; Perry-Smith, 2006;
Rost, 2011). Strong ties are measured as the number of ties which are of at least a certain quality (e.g. in terms of duration: ties that have existed for longer than five years). Thus, this measure incorporates both structural as well as relational embeddedness. Strong ties entail the transfer of detailed, implicit knowledge (Hansen, 1999; Uzzi, 1997). Thus, the
transmission of customer needs, which are most likely not codified, from distant silver agers can be facilitated by strong ties. Operationalisation and the data collection method are used analogously to previous studies and collected for three tie quality subcategories that characterise the peculiarity of relational embeddedness (Nahapiet & Ghoshal, 1998) – tie closeness, duration of tie and tie frequency (Perry-Smith, 2006). Furthermore, the interaction type of all the ties was collected analogously to Sosa’s consultation-type interaction (2011). In order to control for relevance of domain information transmitted through each strong tie, a dichotomous item was introduced. Participants were asked to evaluate the following for each tie: I get valuable information on customer needs and trends from this contact (Yli-Renko, Autio, & Sapienza, 2001). In the subsequent analysis section, only ties that facilitate customer need relevant information transfer were counted.
Practical data collection was conducted using an established standard procedure (Perry- Smith, 2006; Reagans & Zuckerman, 2001; Rost, 2011; Sosa, 2011). Firstly, participants were asked to supply a list of all the silver agers they know, be it from their own workplace, other professional contacts or friends and family. In a second step, participants were shown their list of ties and asked to specify each individual relationship along certain criteria measured on Likert scales. These criteria include tie closeness (1 – acquaintance … 5 – very close friend), tie duration (1 – less than two years … 4 – more than 10 years), tie frequency (1 – daily … 6 – less often [than several times a year]) (Perry- Smith, 2006) and whether the specific ties help their own representation of the target group in terms of customer needs and trends (dichotomous, 1 – yes/0 – no).
Strong ties were defined as those with one of the top two characteristics for closeness (tie marked as good friend or very good friend), duration (tie exists for 5-10 years or more than 10 years) and frequency (interactions with tie occur several times a week or daily) (Perry- Smith, 2006). An example can be found in appendix A.
Temporal proximity
Temporal proximity is the inverse of temporal distance. It effects for ‘younger than silver ager’ product developers comprise mentally travelling from the here and now to their own silver age. This involves estimating one’s own customer needs in the future, when one is a silver ager. Thus, temporal distance depends on the individual’s own age. An easy- to-understand metric for temporal proximity is distance to silver age in years (e.g. 65 years – age of participant) or simply the participants age in years. Here, the latter
construct was used. Thus, the older the developer participants the higher the temporal proximity to the silver-ager target group.
Control variables – Domain-specific expertise and skills
Knowledge, i.e. human capital, in terms of the nature of the problem in focus, is an important antecedent to successful problem-solving (Volkema, 1983). Besides knowing what, knowing how (i.e. procedural knowledge) is also critical for problem-solving success. In this problem-solving task, how is associated with ideation task experience (Lovett & Anderson, 1996) and is controlled for by evaluating the status of the participant (university associate or professional). Knowing what refers to skills knowledge. Here, tenure in product and/or service development is collected as a measure for skill in developing ideas (Boudreau & Lakhani, 2015). Secondly, domain expertise can be critical for ideation success. Thus, tenure in aviation industry-related product or service development is collected. Thirdly, expertise with the specific target group of silver agers can have an effect on ideation outcomes. In line with several other authors, knowledge stocks were measured in years of experience (McFadyen & Cannella, 2004; Quiủones, Ford, & Teachout, 1995). The specific question was: “Do you have experience in product or service development (technical or commercial) (a) in the aviation industry/field, (b) in other industries/fields, (c) for the silver-ager target group?”. Each sub-question could be answered by choosing from “no experience yet”, “0-2 years’ experience”, “3-5 years’
experience” or “>5 years’ experience”.
Control variables – Use experience
Use experience is another domain-specific factor. Use experience presents the extent and frequency to which someone is involved in product/service usage. It is linked to use knowledge or need knowledge. Specifically, tacit knowledge is accumulated through direct use experience (Haldin-Herrgard, 2000), which accumulates the more products or services are used. Use experience is positively linked to lead-userness, i.e. being ahead of trend and having higher expected benefits from innovation in the field (Herstatt & Hippel, 1992). It is commonly operationalised through measurement of the product of time from initial usage (e.g. 16 years ago) and intensity of usage (e.g. times used per year) (Schreier
& Prügl, 2008). The product/service offerings surrounding the customer experience journey in air travel have been changing rapidly, e.g. through increasing market share of low-cost airlines and deregulation. Thus, use experience gained many years or decades ago might be rather irrelevant in today’s world. Consequently, this study emphasises
recent use experience (within the past five years). Therefore, use experience is condensed to one item termed: “How many times did you travel by air in the last five years?”
Control variables – General education
General cognitive capabilities are typically associated with a higher level of education.
These cognitive capabilities are beneficial in creative problem-solving (Bantel & Jackson, 1989). General education level was measured by asking for participants’ highest degree of academic attainment (e.g. some high school, A-levels, vocational degree, university degree).
Control variables – Cognitive empathy
Empathy is a character trait. It consists of an affective and a cognitive component.
Affective empathy is “an immediate emotional response of the empathiser to the affective state of the empathee” and is associated with emotional responses, identification and feeling (Kouprie & Visser, 2009, p. 442). Cognitive empathy is concerned with intellectually taking the position or perspective of the empathee (ibid.). The effect of the cognitive aspect is widely acknowledged in innovation management studies and is linked to need knowledge (Homburg et al., 2009; Schweisfurth, 2012). It is measured with a three-item construct (Barrett-Lennard, 1981). The construct yields a reliability value of 0.75 (Cronbach’s alpha); values above 0.7 can be regarded as reliable (Nunnally
& Bernstein, 1994).
Control variables – Creativity
Creativity is assumed to be a stable character trait and individuals’ capability to create novel solutions is largely influenced by their level of creativity (Amabile, 1996). Although this study’s focus is on useful ideas (i.e. accuracy of addressing customer needs), a psychometric construct for creativity was included as a control variable. The Adjective Check List was developed by linking self-attributed character adjectives to independent measures of creativity for the same person (Gough, 1979). This test was developed in two steps. First, 1,700 participants filled in a survey linking 300 adjectives to how they perceive themselves. In a second step, the same participants independently completed creative tasks that were assessed by experts. After that, connections between the adjectives and the outcomes of the creativity task were analysed. The adjectives that creative participants checked were linked to creativity and vice versa. Factor analysis
reduced the number of adjectives to 30. Gough and Heilbrun (1979) claim reliability in identifying creative and non-creative persons.
Further control variables
Besides these variables, the demographic factors of gender and status of participants were collected, i.e. whether they were university associates (mostly students), coded 1 or aviation industry associates (professionals), coded as 0.