The Good, the Bad and the Ugly: Affect and its Role for Renewable

Energy Acceptance

Barbara S. Zaunbrecher, Katrin Arning and Martina Zieﬂe

Chair for Communication Science, Human-Computer Interaction Center, RWTH Aachen University,

Campus Boulevard 57, 52074 Aachen, Germany

Keywords:

Emotions, Technology Acceptance, Energy Transition, Semantic Differential.

Abstract:

To foster a socially accepted energy transition, it is essential to gain insights into motives for acceptance or re-

jection of technologies related to renewable energies. The study aims to shed light on the emotional evaluation

of renewable energy technologies and in how far the affective responses are correlated with the acceptance

of those technologies. An empirical study is conducted in which a semantic differential is used to assess

emotional evaluation of wind power, solar power (PV) and biomass for electricity production. Furthermore,

general acceptance of the technologies was assessed. It was found that not only did the technologies differ in

terms of emotional responses they evoked, but that those responses also varied signiﬁcantly between groups of

high and low levels of acceptance concerning the respective energy technology. By analyzing spontaneous as-

sociations with the three energy sources, possible reasons for the affective evaluations were identiﬁed, which

can provide essential topics for the communication about those technologies. Overall, the three renewable

energy technologies revealed different emotional evaluations which might considerably impact the overall ac-

ceptance. It is argued that knowledge about such affective perceptions is useful to tailor energy technology

development in early phases and to steer public information and communication strategies.

1 INTRODUCTION

By decentralizing energy production infrastructure in

the course of the energy transition, it is moving closer

to residential areas. This often causes opposition in

the public, as people fear, e.g., health risks (Crich-

ton et al., 2014), spoiled landscapes (Johansson and

Laike, 2007; Upreti and van der Horst, 2004), and, as

a consequence, reduced property values (Elliott and

Wadley, 2002). Other reasons to oppose are related to

the project planning, for example, a lack of trust in de-

velopers, planners, and policy makers, alongside with

perceived unfairness of the decision making process

osch and Kaltschmitt, 1999; Gross, 2007; Zoellner

et al., 2008). It has been shown that opposition leads

to negative consequences for the planned infrastruc-

ture projects, but that early and systematic participa-

tion of the public can help to decrease these negative

effects (D

utschke et al., 2017). In order to support

this participation with targeted information and com-

munication concepts, it is essential to understand how

the public forms attitudes towards energy infrastruc-

ture to be able to address the issues raised. From a

technical point of view, an energy source is often eval-

uated along objective criteria, such as costs (Benitez

et al., 2008; Klaiß et al., 1995) or CO2 footprint (Sims

et al., 2003). From a social science point of view, the

public attitudes towards renewable energy technolo-

gies can additionally be inﬂuenced by subjective fac-

tors, such as affect (Huijts et al., 2012) and emotional

connotations of the energy source and its infrastruc-

ture, such as the perception of risks and uncertainty

(Frewer et al., 2002).

Affect is deﬁned as “[t]he speciﬁc quality of

“goodness” or “badness” (i) experienced as a feeling

state (with or without consciousness) and (ii) demar-

cating a positive or negative quality of a stimulus.”

(Slovic et al., 2007, p. 1333). Affect has a powerful

inﬂuence on risk perception and decision making by

providing mental “short cuts” which people rely on

for their assessment (Slovic et al., 2005). The impor-

tance of affect is underlined by ﬁndings which suggest

affective evaluations can even “overrule” cognitions:

“When cognitions and affect point in the same direc-

tion (e.g., are both positive or both negative), they

equally contribute to attitude, but when they contra-

dict, then feelings tend to dominate over cognitions

in the formation of attitudes. This shows the impor-

tance of including both cognitions and affects as an-

tecedents for attitudes.” (Huijts et al., 2012, p. 528).

Zaunbrecher, B., Arning, K. and Zieﬂe, M.

The Good, the Bad and the Ugly: Affect and its Role for Renewable Energy Acceptance.

DOI: 10.5220/0006795003250336

In Proceedings of the 7th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS 2018), pages 325-336

ISBN: 978-989-758-292-9

325

Despite its potential inﬂuence on attitudes and deci-

sion making, affect has, to the best of our knowledge,

not yet been profoundly studied in the context of re-

newable energy technologies. This applies in par-

ticular to a quantiﬁed measurement of affective re-

sponses to renewable energies and an attempt to an-

alyze the relation between affective responses and ac-

ceptance. The paper aims at uncovering both, affec-

tive responses to renewable energies on the one hand,

and the relation between those responses and accep-

tance. To this end, ﬁrst, studies on affect in the context

of energy will be reviewed and the remaining research

gap will be identiﬁed. The method of the study is then

pointed out, followed by detailed results on affect and

general acceptance of the energy technologies. In the

discussion, the results are critically analyzed with re-

gard to their contribution to acceptance research and

policy implications. Finally, the methodological ap-

proach is discussed.

2 ACCEPTANCE AND AFFECT IN

THE CONTEXT OF ENERGY

SUPPLY

Affect and emotional connotations of energy sources

play a particularly interesting role in the context of

the energy transition. First of all, their relevance in

the context of renewable energies is given by the com-

plexity of the topic of energy supply. Affect and un-

derlying emotional connotations have been shown to

play a key role in decision making, especially when

the context is complex and uncertain (Slovic et al.,

2007). It is therefore likely that affect will also play a

role for forming attitudes about renewable energies.

The complexity here arises, for example, due to a

generally positive image in the public (sustainabil-

ity), but, at the same time, possible negative effects

on the environment (“green on green conﬂicts”, (War-

ren et al., 2005)), leading to conﬂicting viewpoints.

This makes it difﬁcult for laypersons to come to an

informed decision (Frewer et al., 2002).

Second, it has been shown that communicative

measures are desperately needed to integrate the pub-

lic in renewable energy planning, as this can help to

overcome opposition, and, as a result of this opposi-

tion, overcome delays of projects and dissatisfaction

in the public (D

utschke et al., 2017). It is argued that

there is a need to increase the public understanding

by involving the public in early phases of technology

development and local infrastructure planning and ad-

dressing the public’s speciﬁc information and com-

munication needs. In order to tailor those commu-

nicative measures, “it is crucial to gain insights into

the mental representations of renewable energy sys-

tems and their speciﬁc affective evaluation.” (S

utterlin

and Siegrist, 2017, p. 362). Exploring affect and men-

tal representations can, for example, help to uncover

seemingly paradox attitudes among laypersons (Leis-

erowitz, 2006).

For these reasons, the study of affect in the con-

text of energy and climate change has gained impor-

tance in recent years. A number of studies focused

on the inﬂuence of affect on attitudes towards nu-

clear energy, in order to understand reasons for op-

position and support. In a study by Finucane et al., it

was investigated how risks and beneﬁts of nuclear en-

ergy were perceived when affect as inﬂuential factor

is taken into account (Finucane et al., 2000). By pro-

viding biased information to the participants, it was

observed how an altered overall impression changed

assessment of risks and beneﬁts of the technology and

it was concluded that affect moderates both, perceived

risks and beneﬁts. While Leiserowitz’ and Finucane’s

studies provided a more conceptual approach to the

affect heuristic, other studies pursued more content-

driven approaches, in order to uncover the motives

behind attitudes towards different energy sources. In

Keller et al., the inﬂuence of affect, assessed through

spontaneous associations, on acceptance of nuclear

power plants was analyzed, also taking user factors

(gender) into account (Keller et al., 2012). There was

a relationship between the affective representation of

nuclear power plants and the acceptance of a new

generation of nuclear power plants: signiﬁcant differ-

ences were found between the associations of those

who opposed to nuclear power plants and those who

were more accepting towards them. Beyond the af-

fective evaluation, it was shown that also the content

of the associations is worth investigating to come to a

thorough understanding of acceptance.

Comparative approaches, in which more than one

energy source is investigated with regard to affect,

have also been undertaken. S

utterlin and Siegrist in-

vestigated attitudes towards hydroelectric power, so-

lar power, wind power, and, as a contrast, nuclear

power, but only elicited affective imagery for solar

power (S

utterlin and Siegrist, 2017). For the latter,

they found positive affective connotations. They also

found that affect inﬂuences abstract acceptance more

than concrete acceptance. In the study by Truelove

et al., nuclear energy was compared to wind energy,

natural gas and coal with regard to implicit associa-

tions and policy support (Truelove et al., 2014). It

was found that not only negative, but also positive,

implicit associations persisted. They had, however,

only moderate impact on policy support. Truelove et

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

326

al. also pointed out that the reference against which

an energy source is compared should be carefully cho-

sen (in this study, nuclear energy was for example

evaluated more positive than coal, but not preferred

over natural gas). Most similar to our approach, Renn

(1982) compared “the emotional component of an at-

titude” (p. 251) towards nuclear energy, coal and PV

on a semantic differential scale (Renn, 1982). In the

majority of dimensions, though not in all, PV was

evaluated most positively.

Finally, it has been shown that the inﬂuence of the

affect heuristic is not limited to decision making of

laypersons, but can also be found for experts. Chas-

sot et al. investigated how implicit cognition is re-

lated to the investment behavior of energy investors

(Chassot et al., 2015), and, also for this target group, a

link between positive associations and decision mak-

ing could be observed. In the context of the energy

transition, they highlight the necessity to look beyond

rational arguments: “Implicit associations may hin-

der investment in new energy technologies in ways

that are not obvious to the observer, and sometimes

inaccessible to the decision-maker himself.” (p. 291).

It is important to note that in none of the studies,

associations were evaluated for technical accuracy or

logic. Rather, “[the] intention was merely to measure

the structure of attitudes, to ﬁnd common types of rea-

soning in this matter and to investigate the processes

involved in making up one’s mind (...)” (Renn, 1982,

p. 243). Thus, although the mental representations

might not correspond to technical realities, they are,

nevertheless, important and decisive, because they re-

ﬂect laypersons’ realities and perceptions, which, in

turn, inﬂuence the decision for or against a speciﬁc

energy supply solution. In this sense, they are no less

“valid” than technical facts when it comes to decision

making for laypersons as they can even overrule tech-

nical “truths”. This calls for an open-minded commu-

nication with the public from a planning and policy

perspective, in which laypeople’s concerns are taken

seriously.

Even though the studies used different methods,

focused on different technologies and different tar-

get variables, they provide strong evidence for the as-

sumption that the affect heuristic plays a major role

for the acceptance of different types of energy and in

the energy transition in general. So far, a comparison

between different renewable energies with regard to

affect and attitude is still lacking, although especially

in this context, public perceptions are important to un-

derstand in order to better understand acceptance and

opposition and react in apporpriate ways.

3 EMPIRICAL EVIDENCE

In order to illustrate the interplay between affect and

attitudes towards renewable energies, an empirical

study was designed in which three renewable energy

sources (wind, solar power (PV), biomass) were ana-

lyzed with regard to affective attitudes (implicit) and

acceptance (explicit). In addition, a word association

task served as means to closer examine underlying

motifs, reﬂecting mental concepts and cognitions as

part of the affective evaluation.

3.1 Methodological Approach

Data were collected using a quantitative approach for

which an online questionnaire was designed. It was

distributed via mail and social networks to an inter-

ested German audience. Participation was not grati-

ﬁed and was voluntary. Participants were ensured that

their personal data and all of the results are treated

conﬁdentially and in accordance with the high pri-

vacy standards of the research institution. The ques-

tionnaire consisted of two major parts: the ﬁrst sec-

tion assessed demographic information and user char-

acteristics. Regarding the demographic information,

age, gender, education level, occupation, area of liv-

ing, and other energy infrastructure in the vicinity

were asked for. The user characteristics examined in

the study were general technical attitude (5 items, as

used and tested in (Himmel et al., 2014)) and envi-

ronmental attitude, (combined from attitude towards

renewables (adapted from (Siegrist, 1996)) and envi-

ronmental awareness (selected items from NEP scale

(Dunlap et al., 2000))

. Attitude towards technologies

in general and environmental attitude have been found

to inﬂuence acceptance of technology in general and

of renewable energy infrastructure in particular (Za-

unbrecher et al., 2014). Both attitudinal traits were

therefore hypothesized to also inﬂuence emotive re-

sponses to energy infrastructure and thus needed to be

controlled for. Cronbach’s Alpha for general attitude

towards technology was measured at 0.86, and for

attitude towards the environment at 0.79, the scales

were thus considered reliable.

The second section of the questionnaire assessed

attitude towards wind energy, PV and biomass. For

all questions, 6-point Likert scales were used as an-

swering format to measure agreement (1= do not

agree at all, 6= fully agree). Participants were asked

for their (subjective) knowledge about wind energy,

PV and biomass (“I feel well informed about” and

“I know how a (windpower plant/PV plant/biomass

Detailed questionnaire available from the authors by re-

quest.

The Good, the Bad and the Ugly: Affect and its Role for Renewable Energy Acceptance

327

plant) works”). These questions were used to exam-

ine inﬂuence of knowledge on emotive responses on

the semantic differential scale. In addition, partici-

pants were asked ﬁve questions on each renewable

energy technology to determine general acceptance of

the technologies (Items see Table 2).

The emotional assessment of wind energy,

biomass and PV was measured using a semantic dif-

ferential scale (Osgood, 1952) for each of the three

technologies . It is described as especially useful

to measure “meanings that are based on emotions”

(Bergler, 1975, p. 69). The semantic differential con-

sisted of 13 adjective-pairs which had previously been

tested by ﬁve independent participants for their suit-

ability in the context of energy-related infrastructure

(for dimensions see Figure 1). Furthermore, par-

ticipants were asked to note down spontaneous as-

sociations with the three energy technologies to en-

rich the quantitative with qualitative data and un-

cover acceptance-relevant aspects which had not been

treated in the questionnaire so far. For statistical eval-

uation of the results, SPSS (Version 21) was used.

The signiﬁcance level was set at 5%. Missing per-

centages adding up to 100% are due to missing an-

swers from some participants.

3.2 Sample

329 of originally 389 participants were included in

the sample for analysis (excluded participants had

not completed any of the questions on demographics

or user characteristics). 55.3% were female, 42.9%

male. The mean age was 33.7 years (SD= 15.7),

with respondents up to 30 years of age accounting

for around two thirds of the sample. The educa-

tional level was high, 57.1% of the respondents held

a university degree and further 27.1% school degrees

that qualify for university entrance (“Abitur”). 10.0%

had completed vocational training, and 3.3% had ob-

tained a basic school leaving certiﬁcate. 58.4% lived

in the city center, while around 27.7% lived in sub-

urbs, 12.5% lived in a village. To control for ex-

posure to (energy) infrastructure in their daily lives,

participants were asked which (energy) infrastruc-

ture was within view of their homes (multiple an-

swers possible). 13.4% reported to have transmis-

sion lines in view of their homes, 8.2% wind power

plants, 1.5% biomass plants, 21.3% PV installations.

Further 26.4% had a mobile phone base station in

view of their home and 54.4% indicated that none

of the above could be seen from their home. Re-

garding user diversity, participants showed an over-

all openness towards technology (general attitude to-

wards technologies: M=4.7, (Maximum: 6), SD=0.9)

and a high environmental awareness (M=4.8 (Maxi-

mum: 6), SD=0.7).

3.3 Results

First, the knowledge about and the general acceptance

of renewable energies will be reported. Afterwards,

results from the semantic differential as well as qual-

itative results are presented.

Participants reported overall average perceived

knowledge with mean values between 2.8 and 4.1

(Maximum: 6) (Table 1). They reported highest

scores for knowledge about and functionality of wind

power plants (M=3.6 and M=4.1, respectively) and

lowest perceived knowledge for biomass plants. Val-

ues for the feeling of being informed and perceived

knowledge of function correlated highly (PV: r=0.77,

p ≤ 0.01, Wind: r=0.68, p ≤ 0.01, Biomass: r=0.75,

p ≤ 0.01) and the two respective items are thus com-

bined (mean value) for subsequent analyses to one

factor (“knowledge”).

Table 1: Perceived knowledge about renewable energy in-

frastructure (N=326-328, Scale: Min: 1, Max: 6).

Feel informed Know about

functionality

PV 3.3 (SD=1.4) 3.6 (SD=1.5)

Wind 3.6 (SD=1.2) 4.1 (SD=1.3)

Biomass 2.8 (SD=1.2) 3.0 (SD=1.4)

Regarding general acceptance of the technologies

(Table 2), PV was overall most favored. Participants

were least unhappy if it would be built nearby, found it

least dangerous, accepted it most within view of their

house, expected least health risks and welcomed it

most for energy supply of their hometown. Biomass,

on the other hand, was least favored by the partici-

pants. People were most unhappy at the thought of a

biomass plant nearby or in view of their house, found

it most dangerous and thus also expected the most

health risks and support was weakest for their home

town to be supplied by this form of renewable energy.

Wind energy was rated worse than PV, but better than

biomass on all dimensions of acceptance.

While these ﬁrst questions asked for speciﬁc at-

titudes towards different renewable energy sources,

the semantic differential required participants to indi-

cate their intuitive rating of renewable energies with

regard to different dimensions. Results are depicted

in Figure 1. The semantic differential shows distinct

proﬁles for each renewable energy source. Biomass

was consistently rated worse than PV and wind en-

ergy. Ratings for biomass were especially negative (≤

3.5) for the dimensions “ugly-beautiful”, “disturbing-

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

328

Table 2: Acceptance of renewable energy infrastructure.

PV Wind energy Biomass

N M SD N M SD N M SD

I would be unhappy if X was built nearby. 313 2.4 1.4 314 2.9 1.3 302 3.2 1.3

I ﬁnd X dangerous. 311 1.9 0.9 312 2.2 1.0 298 2.5 1.0

I would accept X in view of my house. 314 4.7 1.3 315 4.2 1.4 305 3.4 1.2

I fear that X could cause health risks. 312 1.8 1.0 313 2.1 1.2 300 2.5 1.0

I would support it if my town

decided to use X for energy supply

314 5.0 1.1 315 4.9 1.2 300 4.1 1.2

Figure 1: Semantic Differential for renewable energy in-

frastructure (N=292-315).

pleasant”, “alien-familiar”, “repulsive-attractive” and

“dirty-clean”. Biomass received the most positive rat-

ing for usefulness. For wind energy and PV, the pic-

ture was less distinct than for biomass. In some di-

mensions, PV was rated more positive than wind en-

ergy, in others, the opposite occurred. While wind

energy was rated more mature than PV, PV was rated

more beautiful. There were only minor differences in

the rating of their usefulness with high overall values

(≥ 5.0). PV was rated more pleasant, modern and

fascinating than wind energy. Regarding complex-

ity and familiarity, wind energy was rated to be sim-

pler and more familiar. PV reached higher values for

attractiveness, peacefulness, and was perceived more

positive and safer. Regarding cleanliness, both energy

sources were rated almost equally positive (≥ 5.0).

In a next step, to investigate the relation between

ratings on the semantic differential and acceptance,

the participants were split according to the levels of

general acceptance towards the respective technolo-

gies to compare semantic differential evaluations be-

tween the two groups. For each participant, a “general

acceptance” score was calculated for each technology

using the mean score of the acceptance items (Table 2,

negatively worded items were recoded), taking M=3.5

as cut-off value between the “high acceptance” (HA)

and the “low acceptance” (LA) group. This was done

with reference to the 6-point Likert scale, for which

3.5 presented the midpoint of the scale and the tip-

ping point between agreement and rejection. Partic-

ipants who had not answered all questions related to

acceptance of the respective energy sources did not

receive a “general acceptance” score and were there-

fore excluded from the group analysis.

For all energy sources, the HA groups scored

higher than the LA groups on all dimensions of the

semantic differential (Figure 2 to 4). This indicates

that affective ratings are related to the explicitly stated

acceptance, supporting the assumption that the af-

fect heuristic holds true for the context of the renew-

able energy sources. Depending on the speciﬁc en-

ergy source, the differences between the groups, as

well as groups sizes and characteristics varied (Ta-

ble 3, Appendix). For biomass, differences between

the LA group (n=73) and the HA group (n=236)

were largest for the dimensions “positive-negative”,

“safety” and “threatening-peaceful” (Figure 2). The

two groups differed signiﬁcantly in terms of gender

distribution, attitude towards technologies and knowl-

edge about biomass: In the HA group, there were

more male participants and group members had a

more positive attitude towards technologies as well

as higher reported knowledge about biomass. For

PV, the differences between the groups (LA group:

n=16, HA group: n= 299) were the larger than for

biomass, with the largest differences occurring on

the same dimensions (“positive-negative”, “safety”

and “‘threatening-peaceful”) (Figure 3). The groups

according to PV-acceptance differed signiﬁcantly in

gender distribution, attitude towards technology and

knowledge about PV: The HA group had a higher

share of male participants, a more positive attitude

towards technology and reported better knowledge

about PV. For wind energy, the two groups (LA group:

n=41, HA group: n= 275) produced extreme differ-

ences with respect to the rating on the semantic differ-

ential similar to the PV groups (Figure 4). The largest

differences occurred on the dimensions “positive-

negative”, “threatening-peaceful”, and “disturbing-

pleasant”. The two groups differed signiﬁcantly in

age, educational level and environmental attitude:

The Good, the Bad and the Ugly: Affect and its Role for Renewable Energy Acceptance

329

Figure 2: Semantic differential for biomass for high accep-

tance and low acceptance groups.

Figure 3: Semantic differential for PV for high acceptance

and low acceptance groups.

Figure 4: Semantic differential for wind power for high ac-

ceptance and low acceptance groups.

The HA group was younger, better educated and had

a more ecologically oriented attitude.

Summarizing results for the three different renew-

able energy sources, feelings of danger or threat were

most distinguishing between the HA and LA groups

for all energy sources. For wind energy, the visual ap-

pearance of wind turbines (evoking feelings of being

“pleasant” or “disturbing”) was an additional distin-

guishing factor. These ﬁndings show the underlying

affective evaluations which can help to explain ex-

plicit acceptance ratings.

In addition to rating renewable sources on se-

mantic differential scales, participants were asked for

their spontaneous associations with PV, wind energy

and biomass. Associations and remarks were cate-

gorized, summarizing those associations which were

similar to each other in one category (e.g.: “un-

known”, “not enough information”, “no knowledge

about it” and similar associations were summarized

in “lack of knowledge”). Association categories with

four or more mentions are depicted in Figures 5 to 7

For PV, 251 participants produced 458 associa-

tions (Figure 5). The two most dominant associa-

tions were “sun”, referring to the source of the en-

ergy, followed by the typical location in which PV

panels are found (“roofs”). Further frequent asso-

ciations included references to the type of energy

(heat/electricity), as well as general favorable eval-

uations of PV as a type of renewable energy (e.g.,

“clean”, “useful”, “modern”), with the exception of

“expensive” and “ugly”. Participants also frequently

referred to the dependence on the weather, as well as

unfavorable climatic conditions in Germany for PV as

a general problem. In contrast to biomass and wind

power, PV was seen as a renewable energy “for ev-

eryone”, referring to the possibility of installing it at

one’s own home. In general, PV was seen as a rather

unobtrusive technology, with the downsides of price

and weather-dependency.

In the case of wind energy, 263 participants pro-

duced 535 associations (Figure 6). From the results,

the bigger impact of wind power plants on their envi-

ronment, in contrast to PV, is clearly visible: although

considered to be “clean”, the “size”,“noise emissions”

and “shadows” of wind power plants are frequently

cited. Besides, there is evidence that the “green on

green” conﬂict (Warren et al., 2005) concerning pos-

sible harm to animals (birds, bats, etc.) as well as

“landscape impact”, is deeply anchored in peoples’

minds. In contrast to PV, however, there were also

very positive statements, in which wind power plants

were described as “majestic” or even “elegant”. This

ties in with the ﬁndings from the semantic differential

results, revealing that HA and LA groups produced

extreme results using almost the entire width of the

scale, which reﬂects extremely opposing viewpoints.

For biomass, 232 participants produced 356 asso-

ciations (Figure 7). The most striking result is that –

among the 22 most frequently mentioned categories–

only two (!) are clearly positive (“ecological” and

“useful”). “Odor/stench” was the most dominant as-

sociation, which is coined negatively. This is in strong

Full list of associations available upon request from the

authors.

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

330

contrast to wind power and PV, for which “clean” was

among the most frequent associations. Further neg-

ative associations contained references to ecological

(“monocultures”) as well as ethical concerns (“waste

of food”). Besides, the difference to PV and wind

power was that no speciﬁc infrastructure was among

the most frequent associations for biomass, in contrast

to, e.g., “wind turbines” or “solar panels”.

Summarizing, the results from the direct accep-

tance measurement (using items), the semantic dif-

ferential and the word association task uncovered a

differentiated acceptance-pattern for biomass, wind

power and PV. According to the results, PV was most

accepted, while wind power was seen more controver-

sial, and biomass was least accepted. While the items

to measure acceptance already brought this pattern

to light, the semantic differential helped to system-

atically compare the three renewable energies along

affective dimensions. Based on the theory of the af-

fect heuristic, the semantic differential and the spon-

taneous word associations helped to uncover the hid-

den drivers of these attitudes: while PV was valued

for its unobtrusiveness, wind power was seen more

critical because of its high impact on landscapes and

fauna. Biomass was frequently associated with neg-

ative concepts (dirt, manure, stench), thus receiving

an overall more negative evaluation. Besides, the re-

lation between affective ratings and acceptance was

shown.

4 DISCUSSION

In this research, the public acceptance of three dif-

ferent renewable energy sources, wind power, solar

power and biomass, was empirically studied. The

overall aim was to gain insights into motives for ac-

ceptance or rejection of technologies related to renew-

able energies. Previous studies (Joffe, 2003; Renn,

1982; Slovic et al., 2005; Slovic et al., 2007) have

shown that –apart from “objective” or “factual” ar-

guments, such as costs, known sustainability effects,

or CO2 footprints– acceptance of the public also re-

lies on implicit affective evaluations. In order to shed

light on these hidden emotional evaluations of renew-

able energy technologies and to understand in how far

the affective responses are related to the acceptance

of those technologies, mental models and associations

of participants were investigated and analyzed in re-

lation to acceptance groups. The study revealed that,

overall, the technologies differed distinctly in terms

of emotional responses they evoked. In addition, the

affective evaluations varied signiﬁcantly between LA

and HA groups, supporting the theory that affect in-

ﬂuences attitudes. The analysis of the spontaneous

associations revealed possible reasons for attitudes to-

wards the energy sources, which might contribute to

future information strategies for the respective tech-

nologies directed at laypersons. Following, key ﬁnd-

ings are summarized as well as their policy implica-

tions, followed by a discussion of the effectiveness

of the methodology. The section closes by outlining

some limitations and future research duties.

4.1 Key Findings: Perceptions of

Energy Technologies

The results support the hypothesis that there exists a

link between the affect heuristic and perceptions of

renewable energies: the energy sources with positive

ratings on the semantic differential and in word as-

sociation tasks were perceived more positive regard-

ing risks and acceptability than the one with nega-

tive ratings and negative associations. Biomass re-

ceived, overall, the most negative evaluation, PV the

most positive. This is reﬂected in the acceptance-

judgments: Biomass was seen as bearing the largest

risk (most dangerous, most fear of health risks), PV

as being least dangerous and bearing the least health

risks, as well as being supported most for communal

energy supply. Wind power, for which results were

in the middle of the answering-scales on most affec-

tive dimensions, showed the lowest distances between

the risk and beneﬁt items. This corresponds to results

by Alhakami and Slovic, who found on the one hand

that “items with intense positive or intense negative

evaluations had the largest distance between risk and

beneﬁt judgments.

” and that “items in the middle of

the evaluation scale were associated with smaller dis-

tances.” (Alhakami and Slovic, 1994, p. 1095).

From the results of the semantic differential and

the acceptance groups, it can be deduced that PV rep-

resents the overall most “agreeable” renewable en-

ergy. It is thus not only more positive evaluated than

fossil sources (Renn, 1982), but also than the renew-

ables it was compared against. From the associa-

tions, the reasons for this can be assumed: it does nei-

ther have an impact on landscape, nature, nor humans

through emissions of any kind. The group separation

according to acceptance showed large differences on

the semantic differential scales for PV, however, the

LA-group comprised of only 16 participants, as the

overall acceptance of PV was high. The very selec-

tive group could therefore account for the extreme

views. Wind power, according to the associations,

provoked equally strong feelings in the two groups,

“Item” is to be understood here as “artefact under

study”, rather than “question”.

The Good, the Bad and the Ugly: Affect and its Role for Renewable Energy Acceptance

331

Figure 5: Associations with the term “photovoltaics”, total: 460 associations.

Figure 6: Associations with the term “wind energy”, total: 535 associations.

Figure 7: Associated with the term “biomass”, total: 357 associations.

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

332

on the positive as well as on the negative end of the

scale. From the group comparison and the associa-

tions it can be inferred that it is the most debated of

the energy sources, this will be especially true for peo-

ple with concerns for the welfare of animals and those

sensitive towards landscape change. Biomass repre-

sents the renewable energy source on the most neg-

ative end of the scale from the perspective of social

acceptability. Besides scoring comparably negative

on the affect-scales (semantic differential), the word

associations were mostly unfavorable. Besides odor

emissions, negative associations contained references

to ecological (monocultures) as well as ethical con-

cerns (waste of food). These barriers are in line with

other studies on the acceptance of biomass (R

osch

and Kaltschmitt, 1999; Kortsch et al., 2015). Ac-

cording to the associations, biomass did not provoke

many positive associations, which could explain the

overall hesitant attitude towards it, in contrast to other

studies, in which positive aspects such as job secu-

rity were frequently mentioned (Kortsch et al., 2015).

Concerning the impact of user diversity, no clear pat-

tern emerged in which one demographic variable or

user factor was similarly inﬂuential in all technology

contexts. It thus seems that next to a differentiated

acceptance pattern on the level of content, also the

inﬂuence of user characteristics is different for each

technology. These results should, however, be treated

with care and need further evidence, as some of the

groups observed were only small.

4.2 Methodological Discussion

As in this study, qualitative and quantitative mea-

sures were used to uncover users’ attitudes and af-

fective drivers of energy technology acceptance, the

methodology needs a thorough discussion regarding

its strengths and reliability. The application of the

semantic differential had the advantage that a di-

rect, quantitative comparison between the three en-

ergy sources was possible in terms of affect, as they

were rated on the same scales. In combination with

free associations, the method was useful for a compar-

ative view on affect in the domain of energy supply.

Other studies have compared affect related to wind

and solar energy with coal, nuclear and energy from

gas (e.g., (Lee, 2015)), while we focused speciﬁcally

on a comparison between renewables only. Method-

ologically, this approach was chosen because even re-

newables, although generally having a positive image,

have considerable drawbacks, which might have been

diminished when comparing them to fossil fuels. In

contrast to quantitative approaches, in which attitudes

are measured and used to make predictions follow-

ing a theoretical model, the use of qualitative mea-

sures and the ﬁndings might leave, though insightful,

a fuzzy and somewhat elusive impression behind in

terms of arbitrariness and randomness of the ﬁndings.

One might argue that these qualitative and indirect

measures to uncover affective evaluations might miss

the “numerical” rigor in comparison to quantitative

analyses which use inferential statistics to validate

ﬁndings. Even if we cannot fully rule out this justi-

ﬁable argumentation, only by such “unregulated” and

“free” methodology it is possible to uncover the hid-

den drivers that underlie affective evaluations which

impact acceptance and decision making.

Thus, the combination of qualitative and quantita-

tive methods is mandatory to detect affective accep-

tance drivers, at least in early phases of acceptance

modeling. In the next step, a cross validation with a

more detailed quantitative model is needed. Regard-

ing policy, appealing to the “conscious level of deci-

sion making”, e.g. by presenting technical facts on

new renewable energy projects, may not be the most

effective way of promoting renewable energy infras-

tructure. This also highlights the need for an interdis-

ciplinary approach to such projects which takes the

human factor into account rather than base decisions

on technical parameters only.

4.3 Implications for Policy and

Communication

Considering the implications of the results for public

communication and policy, different topics are rele-

vant for the communication about the different energy

sources. For wind, the impact on landscape and nature

was frequently associated, indicating that perceived

ﬁt within the landscape and visual intrusion are topics

which need to be addressed. This is in line with other

acceptance studies on wind energy, some of which

have even found that perceived negative landscape

impact can “spill over” to evaluations of efﬁciency

(Waldo, 2012). Biomass was not well-known among

the participants, indicating a need for more informa-

tion on this energy source. This is supported by the

lack of a speciﬁc, symbolic infrastructure as part of

the mental model of biomass of laypersons (in con-

trast to, e.g., wind turbines for wind power or solar

panels for PV). It can be assumed that a widespread

diffusion of biomass plants will be difﬁcult to achieve

on the basis of the current level of knowledge of the

public, as participants spontaneously had mostly neg-

ative associations concerning various aspects (eco-

logical, ethical and health concerns). PV, on the

other hand, seemed most agreeable, some participants

even demanded that it should be “obligatory for every

The Good, the Bad and the Ugly: Affect and its Role for Renewable Energy Acceptance

333

house owner”. The concerns regarding dependence

on weather conditions could be addressed by better

informing the public about storage possibilities. From

a social acceptance point of view, results indicate that

PV is likely to provoke the least opposition. Apart

from the implications of the ﬁndings on public infor-

mation strategies, a ﬁnal note is directed at a metaper-

spective of the utility of the mental models prevailing

in the public. Naturally, mental models and affective

evaluations of energy sources of the uninformed pub-

lic might be “illogical” or even “incorrect” from ex-

perts’ points of view. Not considering these mental

models seems ill-advised from a social science point

of view. If experts take those affective evaluations as

a starting point to integrate the public, giving them the

chance to participate in an open and transparent dis-

course, this might help to overcome affective barriers

and to shape public knowledge systematically. In this

case, uncovering implicit mental models and address-

ing them deliberately in an open discourse between

experts, planners, and citizens, represents a kind of

discourse instrument and an educative tool to negoti-

ate accepted renewable and sustainable energy transi-

tion processes.

4.4 Limitations

Regarding the inﬂuence of the affect heuristic on ac-

ceptability of renewable energy sources, it should be

further analyzed how stable the mental models of the

participants are, for example by giving biased infor-

mation. In the case of biomass, for which a lack of

speciﬁc information was identiﬁed, the inﬂuence of

biased information could be different than for wind

power or PV. Additionally, the extent of the inﬂu-

ence of affect on acceptance was not quantiﬁed in

this study. Comparative research on the three en-

ergy sources showed tendencies for affect having a

different inﬂuence on acceptance depending on the

energy source (Zoellner et al., 2008). Furthermore,

user factors and their inﬂuence on affect were not an-

alyzed in detail in this study, however, S

utterlin and

Siegrist found that affective images for solar power

are inﬂuenced by gender, so it is likely that this is

also the case when comparing across different en-

ergy sources (S

utterlin and Siegrist, 2017). They also

suggest a cultural inﬂuence on affect. Future studies

should therefore include more countries, concentrat-

ing on user diversity and the question which individ-

ual factors might impact affective responses towards

energy technologies (e.g., extent of risk tolerance, in-

terest in innovation, extent of ecological awareness).

In addition, out sample was quite young and –in com-

parison to the German population – highly educated.

As education levels, social norms and cultural tradi-

tions also impact the environmental awareness and the

consciousness about societal welfare, a more detailed

look into user factors is needed (Kollmuss and Agye-

man, 2002). Finally, electricity production is only one

part of the grid which needs to be adapted to the en-

ergy transition. Storage facilities for electricity also

play an important role for securing sustainable elec-

tricity supply, also here, mental models can help to

uncover possible barriers to acceptance (Zaunbrecher

et al., 2016).

5 CONCLUSIONS

The study showed that emotional responses to renew-

able energies can, next to cognitive responses, ex-

plain attitudes and uncover hidden motivations. In-

vestigating those affective responses can thus help to

improve understanding for attitudes of laypersons to-

wards energy infrastructure. As research has shown

that an integrative approach which values public opin-

ions and participation can inﬂuence the success of

such projects, it is advisable to integrate public men-

tal models and perceptions into early phases of energy

infrastructure planning and communication strategies.

ACKNOWLEDGEMENTS

Special thanks to Jonas Hemsen for research support.

This work was funded by the Excellence initiative of

the German federal and state governments and the

Strategy fund of RWTH Aachen University (Projects

UFO and KESS).

REFERENCES

Alhakami, A. S. and Slovic, P. (1994). A psychological

study of the inverse relationship between perceived

risk and perceived beneﬁt. Risk Analysis, 14(6):1085–

1096.

Benitez, L. E., Benitez, P. C., and Van Kooten, G. C. (2008).

The economics of wind power with energy storage.

Energy Economics, 30(4):1973–1989.

Bergler, R. (1975). Das Eindrucksdifferential: Theorie und

Technik. Huber.

Chassot, S., Kl

ockner, C. A., and W

ustenhagen, R. (2015).

Can implicit cognition predict the behavior of profes-

sional energy investors? An explorative application of

the implicit association test (IAT). Journal of Applied

Research in Memory and Cognition, 4(3):285–293.

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

334

Crichton, F., Dodd, G., Schmid, G., Gamble, G., and

Petrie, K. J. (2014). Can expectations produce symp-

toms from infrasound associated with wind turbines?

Health Psychology, 33(4):360.

Dunlap, R. E., Van Liere, K. D., Mertig, A. G., and Jones,

R. E. (2000). New trends in measuring environmental

attitudes: Measuring endorsement of the new ecolog-

ical paradigm: a revised NEP scale. Journal of Social

Issues, 56(3):425–442.

utschke, E., Schneider, U., and Wesche, J. (2017).

Knowledge, use and effectiveness of social accep-

tance measures for wind projects. Zeitschrift f

ur En-

ergiewirtschaft, 41(4):299–310.

Elliott, P. and Wadley, D. (2002). The impact of transmis-

sion lines on property values: Coming to terms with

stigma. Property Management, 20(2):137–152.

Finucane, M. L., Alhakami, A., Slovic, P., and Johnson,

S. M. (2000). The affect heuristic in judgments of

risks and beneﬁts. Journal of Behavioral Decision

Making, 13(1):1–17.

Frewer, L. J., Miles, S., Brennan, M., Kuznesof, S., Ness,

M., and Ritson, C. (2002). Public preferences for in-

formed choice under conditions of risk uncertainty.

Public Understanding of Science, 11(4):363–372.

Gross, C. (2007). Community perspectives of wind energy

in australia: The application of a justice and commu-

nity fairness framework to increase social acceptance.

Energy Policy, 35(5):2727–2736.

Himmel, S., Zaunbrecher, B. S., Wilkowska, W., and

Zieﬂe, M. (2014). The youth of today designing

the smart city of tomorrow. In International Con-

ference on Human-Computer Interaction, pages 389–

400. Springer.

Huijts, N. M., Molin, E. J., and Steg, L. (2012). Psycholog-

ical factors inﬂuencing sustainable energy technology

acceptance: A review-based comprehensive frame-

work. Renewable and Sustainable Energy Reviews,

16(1):525–531.

Joffe, H. (2003). Risk: From perception to social rep-

resentation. British Journal of Social Psychology,

42(1):55–73.

Johansson, M. and Laike, T. (2007). Intention to respond

to local wind turbines: The role of attitudes and visual

perception. Wind Energy, 10(5):435–451.

Keller, C., Visschers, V., and Siegrist, M. (2012). Affective

imagery and acceptance of replacing nuclear power

plants. Risk Analysis, 32(3):464–477.

Klaiß, H., K

ohne, R., Nitsch, J., and Sprengel, U. (1995).

Solar thermal power plants for solar countriestechnol-

ogy, economics and market potential. Applied Energy,

52(2):165–183.

Kollmuss, A. and Agyeman, J. (2002). Mind the gap: why

do people act environmentally and what are the bar-

riers to pro-environmental behavior? Environmental

Education Research, 8(3):239–260.

Kortsch, T., Hildebrand, J., and Schweizer-Ries, P. (2015).

Acceptance of biomass plants–results of a longitudi-

nal study in the bioenergy-region Altmark. Renewable

Energy, 83:690–697.

Lee, R. P. (2015). Stability of energy imageries and af-

fect following shocks to the global energy system:

The case of Fukushima. Journal of Risk Research,

18(7):965–988.

Leiserowitz, A. (2006). Climate change risk perception and

policy preferences: The role of affect, imagery, and

values. Climatic Change, 77(1):45–72.

Osgood, C. E. (1952). The nature and measurement of

meaning. Psychological Bulletin, 49(3):197.

Renn, O. (1982). Nuclear energy and the public: Risk per-

ception, attitudes and behaviour. In Uranium and Nu-

clear Energy: 1981. Proceedings of the Sixth Interna-

tional symposium held by the Uranium Institute, Lon-

don, 2-4 September 1981.

osch, C. and Kaltschmitt, M. (1999). Energy from

biomass – Do non-technical barriers prevent an in-

creased use? Biomass and Bioenergy, 16(5):347–356.

Siegrist, M. (1996). Questionnaire of ecocentric and

anthropocentric attitudes towards the environment.

Zeitschrift f

ur Sozialpsychologie, 27(4):290–294.

Sims, R. E., Rogner, H.-H., and Gregory, K. (2003). Carbon

emission and mitigation cost comparisons between

fossil fuel, nuclear and renewable energy resources for

electricity generation. Energy Policy, 31(13):1315–

1326.

Slovic, P., Finucane, M. L., Peters, E., and MacGregor,

D. G. (2007). The affect heuristic. European Jour-

nal of Operational Research, 177(3):1333–1352.

Slovic, P., Peters, E., Finucane, M. L., Hawaii, K. P., and

Macgregor, D. G. (2005). Affect, risk, and decision

making. Health Psychology, pages 35–40.

utterlin, B. and Siegrist, M. (2017). Public acceptance of

renewable energy technologies from an abstract ver-

sus concrete perspective and the positive imagery of

solar power. Energy Policy, 106:356–366.

Truelove, H. B., Greenberg, M. R., and Powers, C. W.

(2014). Are implicit associations with nuclear energy

related to policy support? Evidence from the brief

implicit association test. Environment and Behavior,

46(7):898–923.

Upreti, B. R. and van der Horst, D. (2004). National re-

newable energy policy and local opposition in the UK:

The failed development of a biomass electricity plant.

Biomass and Bioenergy, 26(1):61–69.

Waldo,

A. (2012). Offshore wind power in Sweden - A

qualitative analysis of attitudes with particular focus

on opponents. Energy Policy, 41:692–702.

Warren, C. R., Lumsden, C., O’Dowd, S., and Birnie, R. V.

(2005). Green on Green: Public perceptions of wind

power in Scotland and Ireland. Journal of Environ-

mental Planning and Management, 48(6):853–875.

Zaunbrecher, B. S., Bexten, T., Wirsum, M., and Zieﬂe,

M. (2016). What is stored, why, and how? Mental

models, knowledge, and public acceptance of hydro-

gen storage. Energy Procedia, 99:108–119.

Zoellner, J., Schweizer-Ries, P., and Wemheuer, C. (2008).

Public acceptance of renewable energies: Results

from case studies in Germany. Energy Policy,

36(11):4136–4141.

The Good, the Bad and the Ugly: Affect and its Role for Renewable Energy Acceptance

335

APPENDIX

Table 3: Demographics and user characteristics of acceptance groups (WP: wind power plant, BM=biomass plant), *sign.

differences (p≤ 0.05)

Biomass-acceptance groups

HA group (n=73) LA group (n=236)

Gender* female: 52.4% male: 47.6% female: 68.5% male: 30.1%

Age M=33.3 (SD=15.5) M= 35.2 (SD=17.1)

University degree 59.70% 47.90%

Place of residence City center: 57.2%,

Suburbs: 28.8%

Village: 12.3%

City center: 57.5%,

Suburbs: 30.1%

Village: 12.3%

Infrastructure within view WP: 8.9%,

BM: 2.1%,

PV: 19.9%

WP: 5.5%,

BM: 0%,

PV: 27.4%

Environmental attitude M=4.9 (SD=0.7) M=4.8 (SD=0.6)

Attitude towards technology* M=4.8 (SD=0.8) M=4.5 (SD=0.9)

Knowledge biomass* M=3.0 (SD=1.2) M=2.5 (SD=1.3)

Windpower-acceptance groups

HA group (n=275) LA group (n=41)

Gender female: 56.0% male: 42.9% female: 61.0% male: 36.6%

Age* M=32.7 (SD=15.1) M=40.2 (SD=16.8)

University degree* 60.70% 39.00%

Place of residence City center: 58.9%,

Suburbs: 28.4%

Village: 11.3%

City center: 53.7%,

Suburbs: 29.3%

Village: 17.1%

Infrastructure within view WP: 8.7%,

BM: 1.5%,

PV*: 18.9%

WP: 4.9%,

BM: 2.4%,

PV*: 34.1%

Environmental attitude* M=4.9 (SD=0.7) M=4.5 (SD=0.8)

Attitude towards technology M=4.7 (SD=0.9) M=4.7 (SD=0.9)

Knowledge wind power M=3.9 (SD=1.1) M=3.7 (SD=1.2)

PV-acceptance groups

HA group (n=299) LA group (n=16)

Gender* female: 54.8% male: 43.8% female: 81.3% male: 18.8%

Age M=33.9 (SD=15.8) M=31.5 (SD=15.7)

University degree 60.20% 25.00%

Place of residence City center: 57.5%,

Suburbs: 28.8%

Village: 12.4%

City center: 62.5%,

Suburbs: 31.3%

Village: 6.3%

Infrastructure within view WP: 8.4%,

BM: 1.3%,

PV: 21.49%

WP: 6.3%,

BM: 6.3%,

PV: 12.5%

Environmental attitude M=4.8 (SD=0.7) M=4.9 (SD=0.7)

Attitude towards technology* M=4.8 (SD=0.7) M=3.9 (SD=1.0)

Knowledge PV* M=3.5 (SD=1.3) M=2.2 (SD=1.5)

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

336