sustainability
Article
How User-Innovators Pave the Way for a Sustainable
Energy Future: A Study among German
Energy Enthusiasts
Matti Grosse 1,2
1Faculty of Economics and Management, Technical University Berlin, 10587 Berlin, Germany;
2Alexander von Humboldt Institute for Internet and Society, 10117 Berlin, Germany
Received: 13 November 2018; Accepted: 14 December 2018; Published: 18 December 2018
Abstract:
The existence of user-innovators in the household sector is undeniable. Prior research
provided evidence on the vast scope of product developments by household sector users and
documented a large set of individual user-innovator characteristics to explain their behavior.
What has largely been neglected as a potential source of user innovation activities are product-
and technology-specific factors. This study aims to fill this gap by identifying and analyzing
user-innovators in Germany. On the basis of the results of a large-scale survey on German energy
enthusiasts, we find dissatisfaction with existing products and time-consuming implementation as the
main drivers of user innovation in our setting. The results show a negative correlation between data
security concerns and the likelihood of becoming a user-innovator, pointing towards the maker culture
among user-innovators. As an implication of our work, we provide a roadmap for all stakeholders
aiming to harness the potential of user-innovators for future open innovation eco-systems.
Keywords: user innovation; open innovation; smart energy; energy transition; Germany; survey
1. Introduction
The need for a clean, secure, and sustainable energy future is more urgent than ever. Not only
is Germany on its way to missing the climate targets for 2020, but also for the succeeding decades.
Long-lasting heat periods are becoming increasingly noticeable and can cause social effects and negative
consequences for nature, agriculture, and energy production [
1
]. Rifkin clarifies this importance and
argues that the progress of our entire society is related to the energy system, with users and consumers
playing an important role in shaping a sustainable energy future [2].
Smart energy products (e.g., smart home applications, smart meters, and energy monitors) are
designed to help identify power guzzlers in households, to conduct energy-intensive processes in
times of excess capacity, and to optimize energy consumption [
3
]. However, the sale figures for smart
energy products are low [
4
]. Research reveals that disinterest and disenchantment are the main drivers
of the lagging adoption [5]. Other studies identify data security concerns to explain the non-use [6].
In situations in which products available in the market do not fit user needs, consumers usually
start innovating for themselves. This phenomenon called user or household sector innovation, which
was comprehensively summarized in [
7
] for the first time, is an integral part of the open innovation
idea [
8
,
9
]. As recent examples show, the range of user-innovated products include gadgets and tools
facilitating the lives of people with disabilities [
10
] as well as solutions reducing the bacteria inside
a baby-bottle and bottles facilitating the preparation of the baby formula [
11
]. Representative studies
have been conducted in several countries since 2012. The results show that 1.5%–6.1% of individuals
engage in user innovation [
12
]. A recent re-estimation of these findings reveals that the numbers
Sustainability 2018,10, 4836; doi:10.3390/su10124836 www.mdpi.com/journal/sustainability
Sustainability 2018,10, 4836 2 of 16
of user-innovators may even be underestimated because of methodological issues concerning the
data collection [
13
]. In any case, the number of user-innovators and the collectively spent amount of
time and money are impressive and indicate the huge potential for creative ideas of new products
and features.
Aside from providing broad evidence, previous research mainly focused on the characteristics
and motivation of user-innovators at the individual level. The small part of variance explained by
the individuals’ attributes in national surveys [
14
] may be mainly due to the fact that other factors
drive user innovation effort, for example, daily practices [
15
], product ownership [
16
], or normative
considerations [
17
]. However, product- and technology-specific factors have been largely neglected
as potential triggers for user innovation. Therefore, we consider framework conditions in a specific
technological area to deliver an additional set of motives for users to start prototyping.
This study aims to identify and characterize user-innovators within the scope of smart energy
technologies. Thus, we conducted a large-scale survey on German energy enthusiasts. The primary
goal was to understand their reasons to innovate in this technologically complex product range as the
market penetration of smart energy products is lagging on the one hand but the energy market situation
(e.g., high-quality infrastructure, mainly low bills, and almost no black-outs) remains luxurious in
comparison with those of other countries on the other hand.
On the basis of the results, we contribute to the literature in three ways. First, we detect
an enormous potential of user-innovators among German energy enthusiasts. Second, we examine
the effects of product- and technology-specific factors on the likelihood of being a user-innovator.
That is, the results reveal that following intrinsic motivation, dissatisfaction with existing smart
energy products and time-consuming implementation are the primary motives of user-innovators
to spring into action. Interestingly, we find the variable data security concerns to be a negatively
significant predictor of user innovation, as it indicates that user-innovators are makers who do not
care much about laws and regulations. In this regard, our results are in accordance with previous
research that hints towards the vigor of user-innovators and the tendency to invoke change [
17
,
18
].
Third, we provide a roadmap for all stakeholders, including active and potential user-innovators
themselves and especially companies, policymakers, and society to emphasize the untapped potential
of these creative minds and present a way how to make user innovation an integral part of future open
innovation eco-systems. Moreover, we discuss why an understanding of user-innovators can help to
understand consumer needs in general. We conclude our work by presenting the limitations of our
study and providing suggestions for future research.
2. Literature Review
In our study, we understand user-innovators as individuals, that is, consumers: (1) practicing
innovation activities in their discretionary time without payment; and (2) aiming to benefit from
the usage of the innovation [
12
]. Following the Oslo Manual 2018, the output created by such
user-innovators is described as an user innovation [
9
]. As companies or research institutes are
usually not involved [
19
], user innovation activities commonly occur outside of formal research and
development (R&D) processes and are thus classified as extra-organizational innovation practices [
8
].
Therefore, in the context of our research, we do not consider firm user-innovators [
20
], lead users [
21
],
and user co-creation activities [
19
]. Moreover, we focus on single user-innovators and do not discuss
collaborative user innovation [22].
2.1. Measuring User Innovation
Measuring household sector innovation is important even if it has been largely excluded from
official innovation statistics [13]. To identify and measure consumer household innovations, national
surveys were conducted in six countries: the United Kingdom [
14
], the United States [
23
], South
Korea [
24
], Japan [
25
], Finland [
26
], and Canada [
27
]. An overview of the studies can be found in [
12
].
Sustainability 2018,10, 4836 3 of 16
All studies are representative and use a common questionnaire as basis for the survey [
12
].
The results show that the percentage of user-innovators differs between the examined countries,
ranging from 1.5% in South Korea [
24
] to 6.1% in the United Kingdom [
14
]. In total, 25.91 million
people can be identified as user-innovators who create and modify products for personal or family use,
excluding service and process innovation [
12
]. While the expenditures range from nearly zero to high
amounts on average, user-innovators spend between a few hundred and more than a thousand dollars
per year on their developments. In the United Kingdom, the amount of privately invested resources
even outnumbers the industry R&D spending for consumer products [14,28].
2.2. User-Innovators: Demographic Characteristics
As the interest in user innovation increased in recent years, the evidence landscape in the
determinants of user innovation has become more detailed but also more diverse. For example,
the results on user-innovator demographic characteristics are heterogeneous. Thus, previous studies
found that the education level is positively correlated with user innovation activities [
24
,
28
]. Other
studies did not find any significant effect of education on the likelihood of being a user-innovator [
25
].
Regarding gender, most studies agreed on the fact that male consumers are significantly more likely
to become user-innovators [
24
,
25
,
28
]. The study [
29
] found that user-innovators in the ideation and
prototyping phase are more likely to be male. However, age seems to be important only in case
different types of user-innovators are differentiated, namely, revealing, social, or silent innovators [
25
].
In general, the age of user-innovators does not seem to affect their innovation activities [
14
,
24
].
Regarding job possession, the study results vary. Whereas employed people are more likely to be
user-innovators in Korea [24], no effect was found in other countries [14,28].
Study results became more consistent in the analysis of two other determinants of user innovation.
User-innovators tend to be highly skilled in their respective fields of innovation, particularly in the
technical context [
14
,
24
,
29
]. Further, ownership seems to be a critical factor affecting user innovation,
both in the physical and digital world [
30
]. Recent results indicated that the separation of ownership
and control negatively affects user innovativeness [
16
]. Therefore, users commonly have ideas for
improving and modifying products but struggle with the realization under certain circumstances.
2.3. Users’ Motives to Innovate
The household sector user-innovators innovate for personal need [
31
]. Beyond this motivation,
users start innovating because the process itself is self-rewarding [
12
]. Generally, intrinsic motivational
drivers are identified as the key motives of users to innovate [
32
,
33
]. Aside from problem-solving
ambitions, feelings of accomplishment and enjoyment are also important [
34
]. These feelings of
inducing self-efficacy and self-expression, or the “I designed it myself” effect as framed in the literature,
describe a psychological benefit to users [17,35].
The findings from these studies confirm the results of the national surveys [
12
]. In Finland,
for example, the main motivations to innovate were “personal needs” and “fun and learning”, and
a minority of the respondents stated the “wish to help others” and “the prospect of selling the product”
as motivation [
26
]. Consequently, a previous work showed that this perceived personal benefit
of innovation positively predicts the likelihood to innovate and allocate private resources to these
tasks [36].
2.4. User Innovation in Complex Technologies
As the national surveys to measure user innovation are not limited in the scope of product
creations or modifications, they investigate the number of user-innovators in various innovation
categories [
12
]. In Japan and the United States, the category “dwelling-related” is the predominant
user innovation category (45.8% and 25.4%, respectively), and the other categories of “gardening”,
“vehicles”, “children”, and “sports and hobbies” range between 4.4% and 14.3% [
23
,
37
]. In the United
Kingdom and Canada, the category “crafts & shop tools” (23% and 22%, respectively) is the top
Sustainability 2018,10, 4836 4 of 16
innovation category, and “crafts & shop tools” and “dwelling” share the first position in Finland [
12
].
The category “household fixtures or furnishing” and “sports or hobby or entertainment related” share
the first position, with at both 17.9% in South Korea [24].
Other studies focused on a particular market circumstances in which classical producer innovation
could be lacking [
38
,
39
] or disenchantment and disinterest among consumers is the reason for slow
market diffusion [
5
]. In these situations, users commonly start to innovate for themselves because
local needs are not satisfied by the market [
40
,
41
]. Moreover, users identify technological gaps
and start creating their solutions that fit their particular environment. The area of open-source
hardware and software tools is predestined for these efforts, as both users are free to use, modify, and
redistribute solutions [
42
,
43
], and contributors to those projects are driven by learning, challenges,
and interaction with like-minded people [
44
]. Moreover, ideological reasons may be another driver
of user innovation [
17
]. Thus, studies pointed at an increased willingness to contribute and actively
participate in socially induced topics such as climate protection and sustainability [45,46].
3. Materials and Methods
This research adopts the approach of previous studies that determined the extent of household
sector innovation activities in various settings [
14
,
24
,
26
,
28
]. Following these studies, our study aims to
quantify the extent and understand user-innovators in the field of complex technologies in Germany.
To achieve this aim, we followed a three-stage research design: First, we created a questionnaire
appropriate to our research focus based on preparatory interviews. Second, we conducted a large-scale
survey on energy enthusiasts in Germany. Finally, we applied multivariate regression analysis to
identify the critical characteristics of German household sector innovations in complex technologies.
3.1. Data Collection
Our survey design was based on the questionnaire for identifying and surveying consumer
innovators that was developed by [
47
] and consistently used in six national surveys so far [
12
].
We decided to conduct six semi-structured interviews with potential user-innovators beforehand
to understand the specific characteristics of smart energy technologies. The interview experts were
randomly approached in the online forum of co2online and asked each one for a 60-min interview
session. Using qualitative content analysis [
48
], we used the findings to double-check and complement
the answer options of the questions focusing on technology-specific aspects. In doing so, we could
reduce the number of missing or irrelevant answer items.
To collect our data on user-innovators in complex energy technologies, we conducted an Internet-
based survey on the newsletter subscribers of the non-profit organization co2online. This think tank,
which is supported by the German Federal Ministry for Environment, Nature Conservation, and
Nuclear Safety, combines know-how, empirical analyses, and target-oriented online communication on
climate protection issues. It is one of the largest non-profit organizations in Germany supporting the
reduction of greenhouse gas emissions, climate protection, and energy transition. As our study focuses
on the identification of user innovation in the specific context of energy technologies, we assume
this data source to be of great value for our research objectives. We are fully aware of the fact that,
in contrast to those of the aforementioned studies, our sample is not representative at the national
level. Nevertheless, we assume to derive valid and reliable results in the context of our research focus.
The data collection process took place between April and June 2017. The final sample consisted of
1260 answers.
3.2. Survey Design
The survey was divided into three parts: (1) diffusion and acceptance of smart energy technologies;
(2) user-innovator identification; and (3) follow-up questions on demographic characteristics.
An introductory statement was initially provided to the respondents, including information on the
study purpose, aims, and confidentiality aspects.
Sustainability 2018,10, 4836 5 of 16
In the first part, we were interested in the different aspects of smart energy technologies.
For example, we asked for the motives for dealing with energy-related issues, the energy products
used by the respondents, the problems they encounter in using smart energy technologies, and the
knowledge sources of the respondents.
The questions in the second part followed the survey script taken from [
47
]. This script includes
a queue that helps respondents to recall innovation activities and thus provide information on their
household sector product innovations, namely, money and time expenditures, diffusion efforts, and
collaboration intentions [
12
]. This information is critical because it helps to exclude “false positives”
in the screening phase [
14
]. Following the suggestion of [
12
] in the case of a specific research scope,
we included questions related to particular problems faced by user-innovators in the field of complex
energy technologies and asked for the respondents’ assessments.
3.3. Data Screening
In the screening process, we applied the procedure initially developed by [
14
] and refined in [
26
].
This method identifies the respondents whose innovations are qualified as user innovations and are
thus (1) not job-related; (2) not available in the market; and (3) functionally novel and consequently
contains user-developed content [14].
Out of the 1260 respondents, 319 reported the realization of at least one idea in smart energy
technologies. Thus, 914 cases were excluded from the sample because the respondents could not
recall any innovation-related activity within the last three years. In the second step, we screened
27 cases in which innovation was part of the respondent’s job or business. As we are only interested
in creations or modifications that are at least new to the user, we excluded 145 respondents because
they only replicated existing products in the market. The remaining 147 were identified as potential
user innovations.
In the final screening phase, the remaining cases were analyzed in terms of the respondents’
answer to the following open-ended question: “What was new about your creation/innovation?”
This step is essential because it enables the researchers to assess the description of the respondents’
self-claimed innovations in terms of the creation’s degree of novelty. In this analysis, we found 31
“false positives”, which were discarded at this point. Thus, a sample of 116 validated cases of user
innovation was identified, accounting for 9.2% of the initial sample of 1260 respondents. Looking at
concrete examples from the survey data, it becomes clear how complex the solutions developed by
user-innovators can be, e.g., participants report on the development of control elements for a three-way
valve to support heating by solar panels, a self-developed dashboard to monitor and control the charge
of the electric car, solutions for heating their own swimming pool by solar energy, and the linkage of
proprietary standard components by their own interface solutions.
In all the evaluation steps, the two raters were strict in applying the judging criteria, and thus all
the cases were initially assessed independently in terms of their fulfillment levels. In case a respondent
had a different assessment, we consulted a third rater and made the final decision after a discussion.
Further, we particularly excluded instances in which the respondent’s information was insufficient in
the final stage. Figure 1summarizes the results of the screening procedure.
3.4. Variables and Models
3.4.1. Dependent Variable
On the basis of the screening, we generated the variable userinnovator, which is a binary variable
that takes the value of one if the respondent’s user innovation successfully passed the screening
procedure. This variable served as the dependent variable in all subsequent regression models.
Sustainability 2018,10, 4836 6 of 16
Sustainability 2018, 10, x FOR PEER REVIEW 6 of 17
Figure 1. Screening procedure to identify user innovations.
3.4 Variables and Models
3.4.1. Dependent Variable
On the basis of the screening, we generated the variable userinnovator, which is a binary variable
that takes the value of one if the respondent’s user innovation successfully passed the screening
procedure. This variable served as the dependent variable in all subsequent regression models.
3.4.2. Independent Variables
As our study focused on the motives and characteristics of user-innovators in complex
technology settings, we collected relevant data in the survey and coded a set of context-related
independent variables. Table 1 summarizes these variables.
Figure 1. Screening procedure to identify user innovations.
3.4.2. Independent Variables
As our study focused on the motives and characteristics of user-innovators in complex technology
settings, we collected relevant data in the survey and coded a set of context-related independent
variables. Table 1summarizes these variables.
Table 1. Description of variables.
Variable Description Values/Range
userinnovator
The respondent successfully worked on a user innovation within the last three years.
0 = no; 1 = yes
intrinsic The respondent’s work is characterized by intrinsic motivation. 0 = no; 1 = yes
techknow Respondent’s level of technical knowledge. 0–10
opensource Usage of open-source software or hardware. 0 = no; 1 = yes
costs The purchase and implementation costs are high. 0 = no; 1 = yes
ignorance I do not know which product is suitable for me. 0 = no; 1 = yes
dataconcern I have concerns about what happens to my data. 0 = no; 1 = yes
functrange The functionalities are not enough. 0 = no; 1 = yes
operatdiff The operation is difficult. 0 = no; 1 = yes
dissatisfaction The product does not deliver what it promises. 0 = no; 1 = yes
impltime The implementation is time-consuming. 0 = no; 1 = yes
gender Gender of the respondent. 0 = female; 1 = male
age Age of the respondent. 17–83
ownership Binary variable indicating if the respondents own the house or flat, they live in. 0 = tenant; 1 = home owner
largecity Binary variable indicating if the for respondents live in a city with at least
20,000 inhabitants 0 = no; 1 = yes
unidegree Binary variable indicating if the respondents own a university degree. 0 = no; 1 = yes
income
Binary variable that is one if the household income is 4000 Euro per month or higher.
0 = no; 1 = yes
First, we asked the respondents to rate the importance of various reasons to engage with the
topic of smart energy technologies on a Likert scale from totally unimportant to very important (e.g.,
Sustainability 2018,10, 4836 7 of 16
cost- and energy-saving considerations, a professional interest, him/herself a producer of energy).
A cluster analysis using Ward linkage clustering [
49
] revealed three underlying clusters of motivation:
cost-saving, job, and fun. For subsequent analyses, we generated the variable intrinsic, which is
a binary variable that takes the value of one if the respondent falls into the third motive cluster that
includes individuals who are driven by enjoyment, fun, and willingness to learn.
Second, we suspect technology-related drivers of innovation activities to be part of the
motivational landscape of potential user-innovators aside from personal motives [
5
]. Thus, we asked
the respondents for an assessment of the following reasons to be potentially dissatisfied with
smart energy technologies available in the market: costs (costs), ignorance in the usage (ignorance),
data protection concerns (dataconcern), limited functional range (functrange), operational difficulties
(operatdiff), dissatisfaction in quality (dissatisfaction), and a considerable implementation time (impltime).
For the analyses, we coded the motives as dichotomous variables taking the value of one if the
respondents rated the respective reason as relevant or highly relevant in their context.
Third, we were interested in the type of technology that the respondents used. Thus, we generated
the variable opensource, which is a binary variable that takes the value of one if the respondent stated
using either open-source hardware or software products.
As the probability of being a user-innovator is likely to be determined by further characteristics,
we included a number of control variables in the subsequent regression models. Following previous
studies [
14
,
26
], we included the variables gender (binary variable), age (continuous variable), education
(unidegree, which is a binary variable that takes the value of one if the respondent stated having
a university degree), and income (a binary variable that takes the value of one if the respondent
belongs to the highest income class in the survey, i.e., with a monthly household income of above
4000 euros). Apart from these demographic characteristics, we considered three context-specific factors,
namely, ownership, level of technical knowledge, and city size. As a previous work identified that the
separation of ownership and control negatively affects user innovativeness [
16
], we included another
binary variable (ownership) that takes the value of one if the respondent stated owning a flat or a house
and zero if the respondent is a tenant. Further, we asked the respondents for a self-assessment of
their degree of technical knowledge. As we assume technical knowledge (techknow) to be an essential
prerequisite of user-innovators in the field of complex technologies, we included this variable in the
subsequent analyses. Finally, we controlled for city size, as discussions in the energy transition context
usually differentiate between large-city-solutions and remote solutions for rural areas [
50
,
51
]. Thus,
we coded a dichotomous variable (largecity) that takes the value of one if the respondent lives in a city
with more than 20,000 inhabitants.
3.4.3. Regression Models
As the dependent variable in all our model specifications is a binary viable, we modeled the
probabilities of the dependent variable userinnovator with respect to our independent variables xby
πi≡Prob(userinnovatori= 1|x) = F(x0
iβ) (1)
where xis the (kx1) regressor vector, and
β
is the (kx1) vector of coefficients to be estimated. Further,
we specified Fas the cumulative distribution function of the logistic distribution [
49
]. Subsequently,
the probit regression model coefficients were estimated using the maximum likelihood approach. Our
standard errors were robust to heteroskedasticity.
We estimated three model specifications. The first specification focuses on the effect of intrinsic
motivation on the likelihood of being a user-innovator. In this step and all other model specifications,
we included the level of technical knowledge and all controls. In the second specification, we added
all technology-related motives to innovate. Then, we checked for the effect of using open-source
technologies on the likelihood of being a user-innovator.
Sustainability 2018,10, 4836 8 of 16
4. Results
4.1. Amount and Expenditures of User-Innovators
We started our analyses by identifying the share of user-innovators in Germany that created or
modified a product or a solution in the context of smart energy technologies within the last three
years. In our sample, we found 9.2% of our respondents, aged 18 or older, to be innovating users.
As described earlier, these individuals were successfully audited in the previously discussed identification
procedure [
12
]. Comparing our results with those of previous studies, i.e., 6.1% in the UK [
14
], 5.2%
in the US [
28
], 3.7% in Japan [
28
], 5.4% in Finland [
26
], 5.6% in Canada [
27
], and 1.5% in Korea [
24
],
the share of user-innovators in Germany was relatively high. This finding is mainly due to two reasons.
First, as the registration for the mailing list of co2online is voluntary and self-initiated, registered
people may be more likely to be interested in energy and energy-related aspects than the average
German population (self-selection bias). Thus, we assume that the probability of being a user-innovator
is higher in our sample than the likelihood in Germany on average. Similarly, we expect an oversampling
of people with technical expertise, as the topic of smart energy products is a technical issue. As people
with a technical background are more likely to be user-innovators [
14
,
29
], we conclude that the share
and number of user-innovators found in our sample may be overestimated.
Second, in contrast to previous results [
12
], we did not aim to draw from a representative
sample but to investigate instead the potential of user innovation in the specific field of smart energy
technologies in Germany. Thus, we sent out our questionnaires to energy enthusiasts who subscribed
to the mailing list of co2online. We induced a sampling bias because we are not able to obtain valid
demographic data of the mailing list subscribers owing to the data protection regulations. Further,
we did not apply a weighting scheme to our data set, as done in previous studies [
14
], because we
expected the weighting scheme techniques not to compensate for the distortion of our sampling. Thus,
we generated results in the frame of our sample dataset.
Regarding the time investment and expenditures of user-innovators, we found relatively high
numbers compared with previous studies. The identified user-innovators in our sample spent 39.8 days
per year on average for innovation activities in their private contexts. We calculated on average
€
1456
per year as the annual material expenditures by user-innovators. Again, both numbers may be
overestimated. Nevertheless, we suspect that even powerful correction methods would still result in
above-average spending in terms of the time and money of user-innovators in energy technologies as
we look at a specific, cost-intensive technological field.
4.2. Motives of German User-Innovators in Complex Energy Technologies
As this study aimed to explore the drivers of user-innovators in complex technologies,
we performed multivariate probit regression analyses using three model specifications (Table 2).
To investigate the extent to which user-innovators are driven by intrinsic motivation, we included the
variable intrinsic and all the control variables in the first model specification.
The results show that the respondents who have a high level of intrinsic motivation are
significantly more likely to be user-innovators than those acting out of external motivational drivers,
that is, job or cost-saving considerations. Among the controls, age is significantly negatively correlated
with the dependent variable, indicating that the older the people are, the less likely they come up with
innovative solutions in the field of smart energy technology. Further, income is a positive predictor
for being a user-innovator. This finding hints at the fact that user innovation activities in the field
of complex technologies are costly and require a certain degree of financial resources, although
user-innovators generally rely on inexpensive, home-made solutions [14,24,52].
Sustainability 2018,10, 4836 9 of 16
Table 2. Regressions results.
Variables Specification 1 Specification 2 Specification 3
intrinsic 0.527 *** 0.570 *** 0.601 ***
(0.149) (0.183) (0.189)
costs −0.0921 −0.122
(0.188) (0.192)
ignorance 0.0964 0.193
(0.183) (0.190)
dataconcern −0.432 ** −0.439 **
(0.174) (0.179)
functrange 0.109 0.107
(0.180) (0.185)
operatdiff −0.327 −0.347
(0.228) (0.234)
dissatisfaction 0.380 ** 0.318
(0.189) (0.195)
impltime 0.419 ** 0.349 *
(0.199) (0.204)
opensource 0.739 ***
(0.203)
gender 0.313 0.298 0.156
(0.290) (0.408) (0.409)
techknow 0.0893 ** 0.0773 0.0528
(0.0373) (0.0472) (0.0483)
age −0.0138 ** −0.0124 * −0.00950
(0.00565) (0.00710) (0.00729)
ownership 0.0933 0.275 0.229
(0.205) (0.274) (0.282)
largecity 0.203 0.243 0.223
(0.146) (0.182) (0.187)
unidegree −0.0880 −0.164 −0.174
(0.153) (0.189) (0.193)
income 0.250 * 0.268 0.195
(0.144) (0.180) (0.186)
Constant −1.871 *** −1.894 *** −1.796 ***
(0.469) (0.669) (0.678)
Observations 746 400 400
Pseudo R2 0.0765 0.116 0.156
log likelihood −206 −143.2 −136.7
Wald Chi2 34.11 37.47 50.52
Wald df 8 15 16
Wald p-value 3.88 ×10−50.00108 1.90 ×10−5
Standard errors in parentheses. *** p< 0.01, ** p< 0.05, * p< 0.1.
In the second step, we determined the reason why users innovate for themselves, included
technology-related factors, and considered the context of the innovation activities. Based on the results,
three technology-specific factors affect the likelihood of starting user innovation: dissatisfaction,impltime,
and dataconcerns. First, dissatisfaction with the existing products and solutions in the field of smart
energy technologies is a significantly positive predictor for starting user innovation activities. This
finding is in accordance with that of a previous work that proves that personal problems or needs are
the leading source of user innovation [
41
]. Second, the extensive time required for the implementation
Sustainability 2018,10, 4836 10 of 16
of the prevailing energy product is another factor that significantly raises the probability of starting the
development of personally designed solutions to energy problems. This finding hints at the conclusion
that user-innovators are creators and prefer to do it themselves instead of buying off-the-shelf products
mainly when the implementation is time consuming. Third, respondents with data security concerns
are significantly less likely to start innovating. In other words, user-innovators do pay less attention to
regulatory processes and begin to create or modify solutions even before policies or the industry has
agreed on a regulatory framework and implemented standards and privacy protection requirements.
Again, user-innovators are creators and seem to be less concerned about a cultivated field of action.
Among the other variables, intrinsically motivated people are more likely to become
user-innovators again, and age is significantly negatively related to the dependent variable. Income
becomes insignificant in this model specification, as we expect the monetary aspect to be covered by
the technology-related factors that we included in this analysis step. Similarly, the variable related
to the degree of technical knowledge became insignificant. We suspect that the technology-related
variables absorb the effects of the technical expertise of the respondents on the probability of being
a user-innovator.
Finally, we estimated a model specification including all the presented variables, as we were
interested in determining the effect of open-source usage on the likelihood of being a user-innovator
in the context of energy technologies. As expected, the respondents who claimed to use open-source
hardware or software components are significantly more likely to become user-innovators than those
who used predominantly proprietary off-the-shelf solutions. The effects of the other variables remain
fairly constant in this specification. Interestingly, the variable dissatisfaction becomes insignificant
possibly because of the nature of open-source users. As these users are usually characterized by having
a high level of technical knowledge and experience in engaging in this movement and thus in the
community [
53
], we assume that open-source users do not rely on proprietary products from the
beginning and are therefore less dissatisfied because they do not use many off-the-shelf solutions.
4.3. Robustness Checks
First, standard measures, such as the Wald x
2
test statistics and pseudo R
2
, provide evidence
that our model specifications fit the data quite well. We tested our estimated models against various
alternatives containing few variables by employing likelihood ratio tests [
49
]. As the null hypothesis
was rejected in all cases, the test results confirm that additional variables (e.g., employment status,
living space, and persons per household) have no added explanatory power and do not increase
the model fit significantly. Thus, our model specifications fit the data significantly better than the
alternative models.
Second, the initial robustness checks are strongly supported by the fact that our results, including
the estimated coefficients and the standard errors, are relatively constant over all the three models,
thus indicating a sufficient degree of robustness of our findings. There are a few exceptions for which
we have provided intuitive explanations.
Third, we checked for correlations among the variables considered in the various estimations
(Results are available from the authors upon request.). Overall, the relevant correlation matrix seems
satisfactory. We found some increased correlation values in the range of 0.3–0.4 in the variables that
portray the problems users have with existing energy products (e.g., dissatisfaction,impltime, and
operatdiff). However, as users seem to face more than one single problem with an existing solution and
the questionnaire is allowed to assess various problems with high relevance, we accept the identified
correlation values and assume robust results.
Fourth, the fact that only 9.2% of the respondents were identified as user-innovators could be
considered as a search for rare events (small-sample bias). Therefore, the problem of separation may
occur [
54
]. This is a case in which the maximum likelihood estimates may tend toward infinity and
become inestimable. A method introduced by [
55
] addresses this problem and allows convergence to
finite estimates. We re-estimated our model specifications using Firth’s method (Results are available
Sustainability 2018,10, 4836 11 of 16
from the authors upon request.). The significance levels remained constant in all specifications, with
the estimated coefficients changing only marginally. The problem of separation seems to be absent in
our analyses.
5. Discussion
5.1. The Need to Support User-Innovators
Our results show the potential of user innovation in smart energy technologies. We prove the
existence of user-innovators in a complex technological field. Therefore, 10% of German energy
enthusiasts can be identified as user-innovators in the area of smart energy products. Despite the
potential overestimation of the share in our study, the results indicate the expected rise in household
sector innovation because of better education and access to innovation-required resources [28,56].
A vast but largely untapped potential is apparent. Users have ideas and usually start implementing
them. Our results are in accordance with those of previous work confirming the importance of intrinsic
motives as drivers of user-innovators [
43
]. However, previous research has not sufficiently covered
all aspects of the motives of user-innovators to innovate. Technology-related factors are another set of
important triggers for household sector innovation activities, especially in complex technology scenarios.
Regarding the question related to the realization motives again, the data indicate the lack of diffusion
efforts among user-innovators. The majority innovated for themselves, whereas only a few respondents
opted for dissemination. Here, our data confirm the results of previous research showing the tendency
of user-innovators to act for intrinsic reasons and not to invest funds in the diffusion [
57
,
58
]. From
a welfare–theoretical perspective, the diffusion is essential because the innovation of one individual can
be useful for many others. Consequently, the efficiency of user innovation efforts increases as diffusion
takes place.
The goals must be to increase the macrosocial awareness of user innovation and to stimulate
discussions on how to better support these creative people. We suggest support at five levels:
•
Technical support (e.g., user-friendly toolkits for designing and prototyping, creating open
developer platforms);
•Informational support (e.g., information campaigns, establishing communication channels);
•
Financial support (e.g., financial support and reward programs for user-innovators through policy
and business);
•
Structural support (e.g., information events for users with their own ideas, user innovation
workshops, hackathons);
•Regulatory support (e.g., new working-time models such as the 20% rule, Free Friday).
The basis of all these efforts is the rising awareness and the associated increase in the self-efficacy
of user-innovators. Therefore, we intend to demonstrate the opportunities and action space to
user-innovators and give them a vocabulary for their actions. Doing so helps them to understand
the phenomenon of user innovation in their self-perception as a user-innovator and in companies,
policymaking, and society.
5.2. Understanding User-Innovators Entails Understanding Customers
Understanding user-innovators in the field of complex energy technologies means understanding
the needs of people in the context of smart energy products, as user innovations arise primarily where
existing market solutions do not satisfy the customers. Moreover, user-innovators can help to better
understand the role of people in the future energy system. Our results illustrate the diversity of users’
needs and concerns as well as their different requirements for smart energy products and solutions.
There is no “the” user or “the” customer of smart energy products. Instead, there are heterogeneous
user groups with different questions and problems, diverging levels of knowledge, and varying
degrees of willingness to invest time and money in their own “small-scale energy transition”. Thus,
Sustainability 2018,10, 4836 12 of 16
we need a more substantial heterogeneity of smart energy products and solutions that address the
specific requirements and questions of the corresponding user groups. Energy-related user innovations
can act as an idea pool and thus a source of innovation [
14
] in the creation of smart energy products
and services. As our results show, users do have many ideas while using products at home. These
ideas must be pursued and realized either by the users themselves or by companies that can assist and
scale up solutions. In this context, an appropriate incentive, risk, and organization management is
important as various user-innovator types react differently to diverse stimuli and approaches [25,59].
Open platforms and developer environments, that is, those that we know from the smartphone
and tablet sectors, can help to speed up the creation of innovation eco-systems [
60
,
61
]. In these
environments, users find a place to channel their ideas and developments, the producer can co-create
new product developments with customers and user-innovators, and policymakers can observe
user-producer-eco-systems to monitor and evaluate framework programs dedicated to foster open
innovation. Nevertheless, these open eco-systems only work if standards are harmonized and data
interfaces are created. In this regard, we see the need for policymakers, businesses, and standard-setting
organizations to work together to build the necessary framework and to provide the technical
infrastructure. To speed up the development, we suggest a roadmap defining the goals for each
stakeholder group (Table 3).
Table 3. Roadmap for integrating user-innovators in innovation eco-systems.
Short-Term Mid-Term Long-Term
Active
user-innovators
recognize the relevance of
one’s activity
strengthen self-efficacy;
activate others;
circulate ideas and achievements
increase self-confidence;
contribute to professionalization efforts;
networking;
become a role model
Potential
user-innovators
get inspiration;
perceive examples of user
innovation
identify starting points;
start first innovation activities
strengthen self-efficacy;
collaborate with others
Business
be aware of user
innovation activities;
give user-innovators credit
understand the potential in one’s
own context;
evaluate possible scaling;
provide access to volunteer help
(e.g., experts)
offer testing sites for user-innovators;
create developer toolkits and websites;
build open innovation eco-systems
Policy be aware of user
innovation activities
empower user-innovator actions;
identify possibilities to support and
assist (e.g., funding or data access)
integrate user innovation support in
funding and framework programs;
build user competences through
education initiatives
Society
observe user innovation as
a phenomenon
recognize the activities of
user-innovators benefit from user innovation diffusion
6. Conclusions
Users often innovate for themselves. As a rule, user-innovators do not act from an economic
point of view but try to solve their problems; therefore, they are mainly intrinsically motivated. When
individual users or groups of users become innovative themselves, the trigger is usually that existing
solutions do not or no longer satisfy their needs. In cases in which consumers or users of products
and technologies become innovators, they know their problems better than anyone else, but they are
usually isolated. Consequently, the goal is to make the ideas and solutions of user-innovators available
to all members of society, to promote broad diffusion, and thus to support the active participation of
interested parties in a sustainable energy future.
By identifying user-innovators among German energy enthusiasts, we intend to further increase
the attention of industry and policymakers for user innovation. On the basis of our work, we conclude
that better framework conditions in the form of standards, uniform interfaces, and clear regulations in
data security can help user-innovators to exploit the potential of their work. Further, we promote these
processes and initiate a debate on how user innovation can play a more central role in both business
decisions and political discussions. R&D funding programs should also consider user innovation
Sustainability 2018,10, 4836 13 of 16
activities more explicitly. If making ideas and solutions accessible to individuals and small groups of
innovators in society is possible, then innovation efforts can benefit society as a whole.
Limitations and Future Research
Our study design and the results are not without limitations, which leave room for future research.
Method-wise, the social desirability bias may play a role in the responses of our survey in terms of
issues related to sustainability, climate-friendly behavior, and energy transition. However, as we used
a survey approach, which is non-intrusive and presents no interviewer–interviewee contact, the bias
could have a minor effect on our results [62,63].
Further, we distributed our call to participate through the subscriber base of the co2online
newsletter. Therefore, we assume that most of the respondents in our sample are interested in both
energy and climate issues. Consequently, our study is not representative of the German population
but nevertheless reveals interesting results for energy enthusiasts. Future study designs may opt for
nationally representative data to confirm our results and avoid the self-selection bias.
Similarly, the findings of our study may be specific to the German situation. As energy-related
discussions and developments are country specific, we need more research on the effects of digitization
on consumers, users, and society in energy-related issues in various scenarios. Identifying the
mechanisms on how user innovation helps to reduce market uncertainty in new markets would be
interesting. Further, we believe that innovating users should play a more central role in standardization
and policymaking processes.
In any context, the active involvement of user-innovators in open innovation eco-systems induces
some general implications. We need more research on the actual commercial potential of user ideas for
the development of new products and services. The goal should be to create assessment models that
provide user-innovators, companies, and policymakers with guidance on how to decide on effective
strategies to manage and exploit collaborations. As user-innovators usually do not protect their
innovations with intellectual property rights, a main topic of future researchers should be to address
this issue that challenges the primary incentive system of investments in new product development.
Finally, as prior research shows, user-innovators increasingly exploit the opportunities of
information and communication technologies and share insights and innovation outcomes with
like-minded enthusiasts in Internet forums or other formats [
52
,
64
]. Conversely, our study analyzes
and discusses the innovation activities of user-innovators at the individual level. Therefore, questions
on collaborative innovation or peer-to-peer support were not addressed. The variables modeling these
trends could have added explanatory power to our regression results.
Funding: This research received no external funding.
Acknowledgments:
We would like to thank the editor of the journal as well as three anonymous reviewers for
their valuable and helpful comments. Further, this paper has benefitted from discussions with my wonderful
colleagues Hendrik Send, Thomas Richter, Jakob Pohlisch, and Moritz Zoellner. Finally, we acknowledge support
by the German Research Foundation and the Open Access Publication Funds of TU Berlin.
Conflicts of Interest: The author declares no conflict of interest.
References
1.
Future Impacts of Climate Change Across Europe. Available online: https://papers.ssrn.com/sol3/papers.
cfm?abstract_id=1604531 (accessed on 14 December 2018).
2.
Rifkin, J. The Zero Marginal Cost Society: The Internet of Things, the Collaborative Commons, and the Eclipse of
Capitalism; St. Martin’s Press: New York, NY, USA, 2014.
3.
Wind and Solar Power are Disrupting Electricity Systems. Available online: https://www.economist.com/
leaders/2017/02/25/wind-and-solar-power-are-disrupting-electricity-systems (accessed on 11 November 2018).
4.
German Federal Statistical Office. Smart Home Report 2016. Statista DMO. 2017. Available online: https://de.
statista.com/outlook/279/137/smart-home/deutschland#market-revenue (accessed on 30 September 2018).
Sustainability 2018,10, 4836 14 of 16
5.
Kahma, N.; Matschoss, K. The rejection of innovations? Rethinking technology diffusion and the non-use of
smart energy services in Finland. Energy Res. Soc. Sci. 2017,34, 27–36. [CrossRef]
6.
Kaspersky. Kaspersky-Studie zur IFA: Deutsche Skeptisch bei Virtual Reality, Drohnen und Digitalem Bezahlen.
2016. Available online: https://www.kaspersky.de/about/press-releases/2016_kaspersky-studie-zur-ifa-
deutsche-skeptisch-bei-virtual-reality-drohnen-und-digitalem-bezahlen (accessed on 30 September 2018).
7. von Hippel, E. Democratizing Innovation; MIT Press: Cambridge, MA, USA, 2005.
8. Bogers, M.; Zobel, A.-K.; Afuah, A.; Almirall, E.; Brunswicker, S.; Dahlander, L.; Frederiksen, L.; Gawer, A.;
Gruber, M.; Haefliger, S.; et al. The open innovation research landscape: Established perspectives and
emerging themes across different levels of analysis. Ind. Innov. 2017,24, 8–40. [CrossRef]
9.
OECD. Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th ed.; OECD
Publishing: Paris, France; Eurostat: Luxembourg, 2018.
10.
Woyke, E. The Best Baby Gadgets of 2017 Were All Built by First-Time Parents. MIT Technol. Rev.
2018
.
Available online: https://www.technologyreview.com/s/609645/the-best-baby-gadgets-of-2017-were-all-
built-by-first-time-parents/ (accessed on 10 December 2018).
11.
Arieff, A. Designing for Access. 2017. Available online: https://nyti.ms/2kuzSgP (accessed on
10 December 2018).
12. von Hippel, E. Free Innovation; MIT Press: Cambridge, MA, USA, 2017.
13.
Franke, N.; Schirg, F.; Reinsberger, K. The frequency of end-user innovation: A re-estimation of extant
findings. Res. Policy 2016,45, 1684–1689. [CrossRef]
14.
von Hippel, E.; de Jong, J.P.J.; Flowers, S. Comparing Business and Household Sector Innovation in Consumer
Products: Findings from a Representative Study in the United Kingdom. Manag. Sci.
2012
,58, 1669–1681.
[CrossRef]
15.
Fursov, K.; Thurner, T.; Nefedova, A. What user-innovators do that others don’t: A study of daily practices.
Technol. Forecast. Soc. Chang. 2017,118, 153–160. [CrossRef]
16.
Tietze, F.; Pieper, T.; Herstatt, C. To own or not to own: How ownership impacts user innovation—An
empirical study. Technovation 2015,38, 50–63. [CrossRef]
17.
Dóci, G.; Vasileiadou, E. “Let’s do it ourselves” Individual motivations for investing in renewables at
community level. Renew. Sustain. Energy Rev. 2015,49, 41–50. [CrossRef]
18.
Jeacle, I. The popular pursuit of DIY: Exploring the role of calculative technologies in an actor network.
Manag. Account. Res. 2017,35, 99–109. [CrossRef]
19.
Flowers, S.; von Hippel, E.; de Jong, J.; Sinozic, T. Measuring User Innovation in the UK: The Importance of
Product Creation by Users; NESTA Research Report: London, UK, 2010.
20.
Agarwal, R.; Shah, S.K. Knowledge sources of entrepreneurship: Firm formation by academic, user and
employee innovators. Res. Policy 2014,43, 1109–1133. [CrossRef]
21. von Hippel, E. Lead Users: A Source of Novel Product Concepts. Manag. Sci. 1986,32, 791–805. [CrossRef]
22.
Hienerth, C.; von Hippel, E.; Berg Jensen, M. User community vs. producer innovation development
efficiency: A first empirical study. Res. Policy 2014,43, 190–201. [CrossRef]
23.
Ogawa, S.; Pongtanalert, K. Exploring Characteristics and Motives of Consumer Innovators: Community
Innovators vs. Independent Innovators. Res.-Technol. Manag. 2015,56, 41–48. [CrossRef]
24. Kim, Y. Consumer user innovation in Korea: An international comparison and policy implications. Asian J.
Technol. Innov. 2015,23, 69–86. [CrossRef]
25.
Pongtanalert, K.; Ogawa, S. Classifying user-innovators—An approach to utilize user-innovator asset. J. Eng.
Technol. Manag. 2015,37, 32–39. [CrossRef]
26.
de Jong, J.P.J.; von Hippel, E.; Gault, F.; Kuusisto, J.; Raasch, C. Market failure in the diffusion of
consumer-developed innovations: Patterns in Finland. Res. Policy 2015,44, 1856–1865. [CrossRef]
27.
de Jong, J.P.J. User Innovation by Canadian Consumers, Analysis of a Sample of 2,021 Respondents; Commisioned
by Industry Canada: Zoetermeer, The Netherlands, 2013; Unpublished.
28.
von Hippel, E.A.; Ogawa, S.; de Jong, J.P.J. The age of the consumer-innovator. MIT Sloan Manag. Rev.
2011
,
53. Available online: http://hdl.handle.net/1721.1/77937 (accessed on 11 November 2018).
29.
Stock, R.M.; von Hippel, E.; Gillert, N.L. Impacts of personality traits on consumer innovation success.
Res. Policy 2016,45, 757–769. [CrossRef]
30.
Zhou, M.; Leenders, M.A.A.M.; Cong, L.M. Ownership in the virtual world and the implications for
long-term user innovation success. Technovation 2018,78, 56–65. [CrossRef]
Sustainability 2018,10, 4836 15 of 16
31.
Shah, S. Sources and Patterns of Innovation in a Consumer Products Field: Innovations in Sporting Equipment;
WP-4105; Sloan School of Management, Massachusetts Institute of Technology: Cambridge, MA, USA, 2000.
32.
Hienerth, C. The commercialization of user innovations: The development of the rodeo kayak industry.
R&D Manag. 2006,36, 273–294. [CrossRef]
33. Schulz, C.; Wagner, S. Outlaw Community Innovations. Int. J. Innov. Manag. 2008,12, 399–418. [CrossRef]
34.
Ross, T.; Mitchell, V.A.; May, A.J. Bottom-up grassroots innovation in transport: Motivations, barriers and
enablers. Transp. Plan. Technol. 2012,35, 469–489. [CrossRef]
35.
Franke, N.; Schreier, M.; Kaiser, U. The “I Designed It Myself” Effect in Mass Customization. Manag. Sci.
2010,56, 125–140. [CrossRef]
36.
Bin, G. A reasoned action perspective of user innovation: Model and empirical test. Ind. Mark. Manag.
2013
,
42, 608–619. [CrossRef]
37.
Ogawa, S.; Pongtanalert, K. Visualizing Invisible Innovation Continent: Evidence from Global Consumer
Innovation Surveys. SSRN J. 2011. [CrossRef]
38. Gupta, A.K. Innovations for the poor by the poor. IJTLID 2012,5, 28. [CrossRef]
39.
Yadav, V.; Goyal, P. User innovation and entrepreneurship: Case studies from rural India. J. Innov. Entrep.
2015,4, 1399. [CrossRef]
40.
Hyysalo, S.; Juntunen, J.K.; Freeman, S. User innovation in sustainable home energy technologies.
Energy Policy 2013,55, 490–500. [CrossRef]
41.
Lüthje, C.; Herstatt, C.; von Hippel, E. User-innovators and “local” information: The case of mountain biking.
Res. Policy 2005,34, 951–965. [CrossRef]
42.
von Krogh, G.; von Hippel, E. The Promise of Research on Open Source Software. Manag. Sci.
2006
,52,
975–983. [CrossRef]
43.
Lakhani, K.R.; von Hippel, E. How Open Source Software Works: “Free” User-to-User Assistance.
In Produktentwicklung Mit Virtuellen Communities: Kundenwünsche Erfahren und Innovationen Realisieren;
Herstatt, C., Sander, J.G., Eds.; Gabler Verlag: Wiesbaden, Germany, 2004; pp. 303–339.
44.
Lakhani, K.; Wolf, R.G. Why Hackers Do What They Do: Understanding Motivation and Effort in Free/Open
Source Software Projects. SSRN J. 2003. [CrossRef]
45.
Nielsen, K.R.; Reisch, L.A.; Thøgersen, J. Sustainable user innovation from a policy perspective: A systematic
literature review. J. Clean. Prod. 2016,133, 65–77. [CrossRef]
46.
de Vries, G.W.; Boon, W.P.C.; Peine, A. User-led innovation in civic energy communities. Environ. Innov.
Soc. Transit. 2016,19, 51–65. [CrossRef]
47.
de Jong, J.P.J. Surveying innovation in samples of individual end consumers. Eur. J. Innov. Manag.
2016
,19,
406–423. [CrossRef]
48.
Mayring, P. Qualitative content analysis: Theoretical foundation, basic procedures and software solution.
Klagenfurt 2014,1, 1–143.
49.
Wooldridge, J.M. Econometric Analysis of Cross Section and Panel Data; MIT Press: Cambridge, MA, USA, 2010.
50.
Kennedy, C.; Stewart, I.D.; Facchini, A.; Mele, R. The role of utilities in developing low carbon, electric
megacities. Energy Policy 2017,106, 122–128. [CrossRef]
51.
Green, J.; Newman, P. Citizen utilities: The emerging power paradigm. Energy Policy
2017
,105, 283–293.
[CrossRef]
52.
Hyysalo, S.; Usenyuk, S. The user dominated technology era: Dynamics of dispersed peer-innovation.
Res. Policy 2015,44, 560–576. [CrossRef]
53.
von Krogh, G.; Spaeth, S.; Lakhani, K.R. Community, joining, and specialization in open source software
innovation: A case study. Res. Policy 2003,32, 1217–1241. [CrossRef]
54.
Heinze, G.; Schemper, M. A solution to the problem of separation in logistic regression. Stat. Med.
2002
,21,
2409–2419. [CrossRef] [PubMed]
55. Firth, D. Bias reduction of maximum likelihood estimates. Biometrika 1993,80, 27–38. [CrossRef]
56.
Baldwin, C.; von Hippel, E. Modeling a Paradigm Shift: From Producer Innovation to User and Open
Collaborative Innovation. Organ. Sci. 2011,22, 1399–1417. [CrossRef]
57.
Gambardella, A.; Raasch, C.; von Hippel, E. The User Innovation Paradigm: Impacts on Markets and Welfare.
Manag. Sci. 2017,63, 1450–1468. [CrossRef]
58.
Shah, S.K.; Tripsas, M. The accidental entrepreneur: The emergent and collective process of user
entrepreneurship. Strateg. Entrep. J. 2007,1, 123–140. [CrossRef]
Sustainability 2018,10, 4836 16 of 16
59.
Rayna, T.; Striukova, L. Open innovation 2.0: Is co-creation the ultimate challenge? Int. J. Technol. Manag.
2015,69, 38. [CrossRef]
60.
Hienerth, C.; Lettl, C.; Keinz, P. Synergies among Producer Firms, Lead Users, and User Communities:
The Case of the LEGO Producer-User Ecosystem. J. Prod. Innov. Manag. 2014,31, 848–866. [CrossRef]
61.
Kapoor, R.; Agarwal, S. Sustaining Superior Performance in Business Ecosystems: Evidence from Application
Software Developers in the iOS and Android Smartphone Ecosystems. Organ. Sci.
2017
,28, 531–551.
[CrossRef]
62.
Nederhof, A.J. Methods of coping with social desirability bias: A review. Eur. J. Soc. Psychol.
1985
,15,
263–280. [CrossRef]
63.
Richman, W.L.; Kiesler, S.; Weisband, S.; Drasgow, F. A meta-analytic study of social desirability distortion in
computer-administered questionnaires, traditional questionnaires, and interviews. J. Appl. Psychol.
1999
,84,
754–775. [CrossRef]
64.
Hyysalo, S.; Juntunen, J.K.; Martiskainen, M. Energy Internet forums as acceleration phase transition
intermediaries. Res. Policy 2018,47, 872–885. [CrossRef]
©
2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
