Study Findings

All the findings of our study are summarized in table 9.1. In the top half of the table, the significances of four "control variables” are presented. In the bottom half, significant relationships between Big Five personality factors and likelihood of success at each innovation process stage are shown.

Findings regarding control variables

In order to see the effects of personality traits clearly, one has to "control for” the effects of other variables known to have a strong relationship to innovation process success; hence the term control variables. (By including control variables explicitly in our study model, we addressed what is called omitted-variable bias. This would result from the absence of an independent variable correlated with both the dependent variable and one or more included independent variables.)

The effects of the first two control variables in table 9.1 have been studied and found important in the national surveys of consumer innovation described in chapter 2 (von Hippel, Ogawa, and de Jong 2011; de Jong 2013; de Jong, von Hippel, Gault, Kuusisto, and Raasch 2015; and Kim 2015). In line with the findings of those studies, on the first row of the table we see that male gender is significantly associated with both successful idea generation and prototyping. Gender may also be statistically associated with successful diffusion. However, because most of the individuals who had succeeded at the earlier phases and so were entering the third and final diffusion stage were male, there was not enough variation in the sample entering stage 3 to assess the significance of that control variable in the diffusion stage.

In the second row of table 9.1, we next see that technical background is significantly associated with successful idea generation. As was the case with gender, technical background is so strongly associated with successfully passing stage 1 that most of the individuals who move on to further stages have technical backgrounds. For this reason, the importance of technical background to success at stages 2 and 3 cannot be analyzed. However, we do know from other research that technical background is also very important to stage 2 prototype development (Luthje, Herstatt, and von Hippel 2005).

Table 9.1

Effect of personality on the likelihood to successfully pass through stages of the household sector innovation and diffusion process.

Ideation (stage 1)

Prototyping (stage 2)

Peer-to-peer diffusion (stage 3a)

Commercial diffusion (stage 3b)

Control variables

Gender (male)

.39 (.11)***

.62 (.14)***

-.21 (.27)

.44 (.26)

Technical background

.34 (.11)**

-.05 (.13)

.29 (.21)

-.27 (.19)

Inspiring social environment

.49 (.12)***

.14 (.15)

.50 (.28)

-.02 (.21)

Frequency of unmet needs

.62 (.12)***

.61 (.15)***

-.10 (.25)

.30 (.21)

Big Five personality traits

Openness to experience

.35 (.11)**

.08 (.14)

.21 (.24)

-.09 (.20)


.12 (.11)

-.51 (.16)**

-.28 (.27)

.12 (.22)


-.13 (.11)

.31 (.15)*

-.64 (.28)*

.57 (.28)*


.03 (.11)

-.06 (.14)

-.40 (.25)

-.28 (.24)


-.07 (.11)

-.13 (.15)

-.35 (.32)

.42 (.22)


.59 (.10)***

.13 (.13)

-1.89 (.30)***

-1.53 (.29)***

Model Fit

Wald test statistic (degrees of freedom)

96.36 (9)***

Source: Stock, von Hippel, and Gillert 2016, table 3. n = 547. Analysis method: sequential logit regression. Coefficients reported in log-odds units; robust standard errors in parentheses; degrees of freedom = 9. *p < .05, **p < .01, ***p < .001.

The control variable "inspiring social environment” was included because the social environment within which innovation takes place has been found to be important to innovation likelihood and success. An inspiring environment involves strong social ties (Perry-Smith 2006) and also supportive attitudes toward innovation (Amabile, Conti, Coon, Lazenby, and Herron 1996; Scott and Bruce 1994). For example, a supportive family would say to an ill family member attempting to innovate, "How wonderful that you are being creative in that way, how can we help?,” as opposed to, "Why are you doing something so foolish? You should just follow your doctor's orders!” As can be seen from the third row of the table, this variable is significantly correlated with successful idea generation.

The fourth control variable, "frequency of unmet needs,” refers to the degree to which a respondent felt that he or she had needs not satisfied by products on the market, and so would have a reason to innovate. The association of this variable with innovation likelihood has been documented in numerous studies of innovation by lead users (e.g., Morrison, Roberts, and Midgely 2004; Franke and von Hippel 2003). As can be seen in row four of table 9.1, this control variable was significantly associated with both successful completion of the idea generation phase and completion of the prototype phase too.

< Prev   CONTENTS   Source   Next >